Computer generated transcript

Warning!
The following transcript was generated automatically from the content and has not been checked or corrected manually.

Hi, everyone. We'll, we'll get started in a few minutes. We'll let a few more people trickle in if that's alright. Hi, everyone. Er, I hope you're good. I hope you can all hear and see me. All right. And thank you very much for coming to this academic surgery even today. Um, my name is Bala Rat Singh. So I'm one of the final of your medical students from K university and I'm also a regional lead for K star surge and recently been on the, uh, I'm on the Star Surge steering committee and this is obviously called the West Midlands Academic Surgery evening. And this serves as a kind of insight into surgical academia. And so we have a host of, um, great speakers today. Um, and I really hope that you enjoy what you're gonna, what you're gonna hear today and it gives you a bit more of an insight into academia and gives you a bit of an idea of how to get involved in this, um, for the future as well as it is extremely important. So, in terms of the aims of this, um, evening, generally, we want to promote surgery and academic surgery in particular as potential career paths to, to yourselves as medical students and any, uh, junior doctors that I've attended today as well. Um, we want to facilitate local networking opportunities and collaboration between those interested in surgery and academic surgery. For instance, most of the speakers, er, today personally have actually, er, colab with the on research and they all, and they've all been involved in research projects as well. So it's a really good opportunity to see kind of what the local area is producing in a way. And we also want to promote awareness and opportunities for engagement in regional national groups for surgery and academic surgery. Cos for those of you that don't know, er, Star Sur is a regional, er, is a national collaborative network where everyone from medical students, um junior doctors and consultants are involved in Multicenter studies um that are very impactful and being published in very good journals as well. So we're really um excited to extend this opportunity to everyone to see um for you to see um if you're interested, first of all in uh in academic surgery as especially at medical school, no one really talks about the academic side of things as much to be honest. So these are the speakers we have today. So we have myself, we have, er, Doctor Emily Hall who's a um academic foundation trainee at West Midlands North. We have Doctor Nan Nan Kumar, who's a core fellow in trauma orthopedic surgery at Adam Brookes Hospital and also a teaching fellow at Cambridge. Um, we also have Mr Sheehan Havana that will be, um, joining us later on who's an S TA general and have had bi and pancreatic surgery in the West Midlands Deanery. And we also have Professor Paul called as our keynote speaker, who's a consultant tr orthopedic surgeon, professor of orthopedic surgery at the Robert Jones and Agnes her orthopedic hospital in the Oswestry. So without further ado this is how we're gonna format the day. Um And that's kind of the order of the speaker. So we'll have pro speaking first. Um And then we'll get involved in get involved in surgery as a student in research. Um And then talking about the specialized foundation program which many of you may not have heard of. And then Miss a Banda's talk on balancing surgery and academia. So first up, we've got um, Prof Paul co who's a professor of orthopedists at K University and consultant TN O surgeon at um the I JH Hospital in Os Street. Um His special interests include diagnostics, statistics, genomics, and machine learning. And in this talk, um Procol will highlight examples of machine learning in clinical practice. Um And we'll go ahead now if that's ok. So, Procol, if you're able to um share your screen, I'll share mine. Hi. Yeah, thanks so much for a nice introduction. I be me. It's very kind. Um Yeah, so my name um is uh Paul? Cool and on an Ortho or something went wrong with the sharing here. Um Hang on a second. This was working swimmingly a second ago. Uh But now it's all gone bad shape. I'll tell you again, um share a screen, this one share. So you should be able to see my screen. Now, you see your desktop at the moment. You only got the wrong. Well, I don't know why this has gone off. Yeah, there you go. So, so that's what I wanted to show you. OK. Um So artificial intelligence um show you 20 minutes or so, talk about this. Uh It's a bit of name calling and artificial intelligence is a bit of a, a sort of modern name and people like using the A I sort of um abbreviation as well, but I would like to show you the difference between machine learning, deep learning statistics and actually all these things are pretty much the same um artificial intelligence. It just got all this cloud of, of mystery around it. And uh II just wanted to show you uh uh about that and also to make you a bit wary of it. So just about this, this the terminology, you know, you know about statistics, it is model forming, for example, a regression model you can form and then you can try to predict your uh your data and, and you know the relation between variables. This machine learning is a little bit more of a, of a black box. And there are different uh there, there are different um uh machine learning algorithms you could use. For example, the Vanden Forest is very common, commonly used. And I'll give you an example of that in a, a blood sample data set. You in a second uh deep learning. Um I'll give you also an example of that uh and computer vision which uh is, is of course for, for medical purposes, very exciting neo development. But I'll also show you some, some concerns about that. And then to me, artificial intelligence and people go, oh we, we're doing a II mean, I think to me medicine, there is a way of um to me artificial intelligence when the whole computer is completely autonomous. And so it takes its own input and its own output. So it's self-driving car, for example, and we're nowhere near that your, the patient logs into a computer and, and, and, and they'll do uh the computer will do an operation on the patient. I think they are going to be a long way off. Um So um even if the diagnosis are and I'll come back to that in a uh in a second also, I think that will be really difficult. So if you look at here some problems and you go, if you, if you see here on the left, you see some, some data on the, on the scatter plot and you see with regression modeling with statistics, you could put a straight line through it, but you can see, oh actually there is a sort of different pattern to it. Um And you could maybe get reasonable predictions here and reasonable predictions here. Uh But you're sort of out here and there, you can get a better fit by using some sort of uh uh poly uh fit as you can see here. Or you can use that often do with this um um uh A I models or is machine learning models whereby you can have to worry is that you can do this data and that you can sort of over FT and you may say, OK, they will fit well on your training data, but that does not necessarily mean that it will generate well outside what you're trying to compute up. And that is one of the big problems with uh A I or machine learning that it doesn't generalize very well in, in, in uh in, in many cases. So the types of learning you have, you can have supervised learning and unsupervised learning. And uh supervised learning is what we, we often use. But actually I find unsupervised learning much more interesting. So for example, supervised learning is er if you have different cars, you can say, well, that's a Ferrari, that's a, a Renault, that's a, a Ford uh that's a jaguar or whatever or, or, or, or maybe for the ladies uh different uh shoes. Um, I'm not so familiar with the, with the brand name but, uh, I'm sure you can, can see that so you can picture in or something and it'll tell you which, which brand it is. Um, the problems is you don't really learn anything, you recognize the feature. Uh, but our understanding of diseases and problems change and, and then how do you change your model? Um, so I think our understanding of the whole thing becomes limited and it is just a categorization. And actually what you do is you, you fix in time, uh your current opinion of something, but that doesn't change. And if you don't keep that as a a as given for, for the rest of the time, then, then I think you don't really learn anything. So um that is a, a problem, not a problem with this thing is that it's like a black box and that you actually don't know uh what the system does. And I mean, obviously, if you do a linear algorithm, a linear regression model, then you know what the, what the model does uh you need as, as long as you understand the statistics. But with a random forest and, and, and even more so with deep learning that works uh than excuse me, that's much more difficult. There are some feature importance you can extract and then we'll show you that. But, but this clustering, the, this unsupervised learning, this clustering is probably more important because that actually shows you if you have say um different patients with uh say a bone cancer or so, and you have different X rays, then you can see on, on imaging features, whether you can pull them apart and where they cluster. And I'll show you an example of cluster of clustering um in a second. And so the computer will have no future knowledge and you just see if you can discover patterns and they can be quite useful. Um So this is an example here of what you can or the problem you can get with this. Now, if you look when you zoomed in picture, you can see that, oh, well, if you would classify it on that as a Ferrari, isn't it? But actually, it's just the golf, it's a golf GTI and somebody put some flashy wheels and some nice calipers on it with say Ferrari. Now you can see that the computer will very easily uh misclassified as, as a, as a Ferrari. Unless it gets maybe a bit more information uh about the about the shape of the car, then it may sort of say, maybe not. You can also see perhaps that if you would get um I really, you know AAA picture of the paint here, if it was a Ferrari, then the type of color the computer could recognize. Oh, that's a Ferrari because only Ferraris have that Ferrari red. And, and you could see that but on the other hand, it is perfectly possible to get the same color and, and, and spray a gold GTI with that color. Um So that doesn't prove it's, it is, it is a Ferrari, you know, and so there, there are always going to be limitations to this, to this approach. Um I think so, starting with the statistical modeling, um I'll show you a regression model which I'm I'm sure you're all aware of is the predictive variables on the X axis and the outcome variables on the Y axis. Uh There's other fancy um techniques you can use as principle component analysis whereby you can try to uh understand the variance in the data can be very useful uh for exploring techniques but not normally for model forming but linear discriminant analyst, for example, is another uh technique you can use. And I'll show you an example of that in a second going a bit clinical now because II know you like the clinical bit and, and so do I and so you've got a lady with breast cancer who's got uh a left sided hip uh pain. And you can see here in the hip here uh There are, there is some abnormality and you see that that's a metastasis. Um So now what are your options for this lady? She comes with pain in the hip, you can do nothing and say uh tell her you've got cancer, go away. Uh Not a very nice approach but um uh you could say we do surgically nothing. Um or you can do some something surgically. Uh But um do you need to do something? Is the bone going to break? We don't really know that uh you can give drugs, chemotherapy, radiotherapy or a combination uh of all this. Um And so what would be the right approach and, and as a surgeon and it will be different if you're a radiotherapist, I think, although they will probably have the same approach. Is this going to break? That is the for me, the critical question in the first place. And if you read the literature, then you have this um uh this scoring system here called the meal score. You may have heard about this. Uh It's a bit um um yeah, it's a bit uh niche to be honest with you. But nevertheless, uh it's a scoring system between four and 12 based on the site, the pain, the size and the lesion itself. And if you score more than nine, it's gonna break and if it's less than nine, then you're gonna be uh you're gonna be all right. And so the site, if it's the upper limb, the lower limb or the hip and then the pain where it's mild, moderate functional, the size a third, two, third or more than two, third. And then plastic mixed or lytic sounds quite straightforward. But if you go back um to this, then saying from? Well, is it hip now? Well, yeah, I think most people would agree. It's hip. But could it be femur? Where does the hip stop? Where does the hip start? Where do you classify this as hip? Same as the size you say? Is this a third, a third of what, a third of the cortex? A third of the femur, if you actually read the original um description, it, it actually doesn't say that the pain, somebody has mild or moderate pain. What is that? Um what's mild pain? What's moderate pain? It's very difficult and very open to interpretation this classification and thing. I think most people would agree that this is probably mixed this lesion and in the sense that there are some bone formation here and there's some lysis here. So I think most people would agree with that. This is mixed. But if you look at the scoring, um yeah. So if it is a lot of sorry, it's not a subjective and it, and, and it limits your options and this is the, the the this is what you know, I don't know if you like, I like this thing to reading about quantum mechanics. This is when they talk about the collapse of the wave function. So if you call something, one thing, then you have to call the other thing the other. So I'll give you an example of that. So if you got here in the middle, you've got a say a pixel and pixels in, in color are from 0 to 255. Um So 255 is white and zero is black. Um So most people will, will say this is white. OK. That's right. So what's this middle bit then? Is this white or is this black? Oh The interesting thing is I think if you people push people, you would say, well, actually it's gray and I would agree, I would also call this gray, but you can't call it gray. You can only say it's white or it's black because that's the categories we have now 100 and th this actually, I've, I've, I've done this on the computer. This actually is a gray scale of 100 and 30 the middle of 256 gray scale is actually 100 and 28. Yeah. So this is more than 228 2 pixels higher. This actually you should classify as white, but I don't think many people would do that, you know, and that's one of the problems with classifying that once you start classifying things, um we can get into trouble with, with borderline cases. And, you know, if you look at the variation of this meal score that we've done in lots of hips uh between um between observers, you can see here that observers sometimes three points either way. Um That's quite a lot of difference. There's a huge variation in um in agreement between er observers and you think? Oh, well, but I, at least I will be consistent. Well, sadly, including myself, I'm not and even the, the, the, the same observer will um misclassify or vary uh between its own observations, one or 1.5, 2 points either way showing that this is actually not a very good system. And so what we used, for example, we worked on this with uh Emma Howard, who was one of the uh one of your predecessors in uh in Q. She did her integral masters with me. And actually what we've done is we measured the proportion of weight bearing and this is a rock curve which can show you what's the best predictor is the nine, here is the mineral score. And you can see here this 85 is actually a better predictor uh for fracture than um than the meal score. So we're now using this, this, this mayor score. Uh oh sorry, this weight bearing percentage as, as, as a uh um estimator of um uh of fracture. And now this lady could only put 72% of her weight on the leg. So in other words, it's gonna break. So they say, OK, should we do an operation then, which is then the next question he said, well, actually, that depends still a bit as she, she obviously is going to break. But if her survival would be three months or less, then you should worry about doing an operation. It takes about two months to get over an operation like that. So what's gonna be the benefit for her? And it may be better to have some supportive uh measures. But that's something to consider because obviously these operations are, are quite um uh quite dangerous in terms of you is definitely a mortality involved in them. Um If these survives up to about a year, then fixation is generally seen a reasonable approach. But if you're going to live for more than a year, then we also know that the fracture is never gonna heal if it fractures, er and er then the metalwork will eventually fail just like a paper clip if you keep bending it. Uh And so you will then need to replace it. So then the next thing we did is we use this, this uh quadruple a on the blood test. Uh So you can do blood tests and you do this is linear discriminant analysis. I was on about a little bit earlier and you can, this is a coefficient and so what you can do is just say the albumin and you multiply this by dis coefficient and you get a result and the adjusted calcium, you multiply it by dis coefficient, you get a result and the alkaline phosphatase, you find this coffi you get another result and the age you do this, you know, you see age and albumin if you can imagine if your albumin is good. If you've got high album and you're gonna, you're obviously fit and healthy uh age. Um ladies with older ladies with breast cancer tend to do much better than younger patients. So that's why age is an important aspect here. You probably know that calcium and phosphatase. A alkaline phosphatase are both measures of bone breakdown. And you can see here, calcium, if your calcium is high, the coefficient is very high, that means very bad news. But if you add these all up and uh these uh you get to AAA five, so this is bigger than one. So she's gonna live for more than one. she's gonna live for more. Well over a year is our prediction which actually turned out to be, right? And you can see here uh the, the, the, the big doorknocker we put in for her. Um and this did very well for her and she walked out of the hospital and she was delighted with uh with this result. So next machine learning uh machine learning is a little bit uh different. As I said, the, the model, the algorithm behind it are a little bit more obscure. Uh and, and based in mystery mystery, but nevertheless, there are patterns and they are there to uh show patterns in the data because the data is not random. That's the whole idea about it, you know, so it's not a mystery. Um It is just that the um the algorithms to come to it are more difficult for us as humans to understand, but very easy for computers to understand. And so this is, for example, a study we did in 200 or 2000, 371 patients on blood test. And there were some had hematological cancer, some had sarcoma, some had metastatic bone disease and some had benign uh bone disease. And if you use and it was all about this unsupervised crusting technique here and it look does look a bit like a shark. Um I think anyway, but, but you can see the um the clustering, the blue is the benign and the red is the malignant, you can see that they, they definitely cluster together. So if you are in this, that or that area, you're most likely to benign. And if you are here there or there, you're most likely to have malignant disease. So there is definitely a pattern in the date of the blood test which it sort of makes sense. And you know, just because one blood test may not be diagnostic for cancer, maybe a combination of it uh can be and this may be patterns which we as humans can't see. And this is where a random forest can be very useful. So for example, here this goes like a, a forest is like a, this is called a leaf and then you've got a decision tree uh, which goes down. And so you've got here, if the, er, is less than six, then if that's true, then you're almost certainly benign. Yeah. And if the E sr is more than six, you're certainly being malignant. You see that. And so this is how you can, with all the other blood tests can grow the three and come to AAA sort of categorization in the end. But more complicated, you can then use the shock values um to um to see what your model does. For example, here, the alkaline phosphatase, no surprise. If that is high, the future value is high, then you are highly likely to be malignant disease. On the other hand, if your red blood cell count here um is is high, then you've got, you've got lots of good blood, you can imagine, you know, so you're not anemic. Uh that means that you uh actually have got a very, you're very unlikely to have cancer. You can see that. So it is quite useful um in that respect. And uh that's recently been uh uh been submitted at work as well. Um Deep learning is the next thing getting a bit more into mystery a bit like the Titanic, which is also uh clouded in uh in, in mystery. But this is a famous data set. And if you look at the data set of uh Titanic in adults, you can see here, this, this left table here is the, the ones who died and the right table is the ones who um survived. You can see the first class if you were a first class lady. Um you are not gonna die on the Titanic. You see that uh whilst if you are um third class, yeah, then there were quite a few ladies who who died as you can see here. So not only class but also gender but all where the cabin is is located. You can imagine that the first class cabins were located somewhere else. Um Also the way the tickets were allocated was not random. Some people uh embarked in Cherbourg, some people bar embarked in Southampton. So if you have all this information, you can actually er with 100% certainty, er say er or if somebody's going to die, er or not, which is quite impressive. Um And, and so, you know, with that data, you can actually er say for you're gonna die or you're not gonna die on the Titanic. The problem is you don't really understand how this algorithm comes to it. Um And that's the, the problem with uh this deep learning. It's a bit more er cloaked in mystery computer vision. Um This is um something I wanted to show you because II think it's quite nice and then I'll, I'll stop boring you. Um So just um computer vision, you've got image classification, which is here. So this is bottle cup and a cube or you can have localization. There's the bottle, there's the cube, there is not a cube, there is not a cube or you can have what you call segmentation where you have the actual pixels of the cube outlined rather than a box around it or it instant segmentation where even the different cubes have different colors. So you can count the number of cubes rather than just saying there are cubes in this um in this picture, you know, um and I'll give you an example of cats and dogs. It's a famous example which I like. And then I'll show you something about the giant cell tumor of bone, we miserably failed. Um And I'll show you some object detection in, in nude fields we've done recently, but it's worked quite nicely and some segmentation and then I'll stop. Um So cats and dogs um pixels, as I said, when you look at the picture, actually a picture is three pictures, a red channel, green channel and a blue channel. And all those pixels, they have a number as I just said from 0 to 255. So the 256 numbers because computers start counting at zero. Um So if you have a picture uh of these papers here, then you got the, this, you got a matrix of three matrices of numbers. And here they ra rather than going from 0 to 256, they've just been normalized uh to go from 0 to 1, but they are in a way from 0 to 256, same thing. Yeah, they're just normalized. And then the idea is, is this mystery convolutional neural network, deep learning network whereby you put the picture in, it goes through the convolutional uh network. And then it says it's a cat or it's a dog here, it says it's a cat. Ok. Uh That's how it works. Now, how does this work? Because this again is not a mystery. Um It is just the whole thing. It is difficult to understand and get your head round. But what it is if you have a picture, if you look here at the matrix numbers and you look here at these numbers here, these three by three matrix here at the middle of that picture. And you put a filter or kernel is it also called over it, which is 000100000. Yeah. Now if you lay this over that, then you can multiply. Um So this is the center of the, of the kernel. Yeah. And you multiply them together. So 40 times zero is zero plus 42 times zero is 42 plus zero times 46 is zero. And these are all zero as well. So this pixel here becomes 42. If you then move the kernel one step to the right, then you can see that the next step here will be 46 and the one here. Um It will be 52. So you can see actually this kernel doesn't do very much apart from shifting the whole picture um one pixel down. But you can design other kernel uh the more mo more clever kernels, horizontal, vertical diag diagonal or cross or whatever you want to do. For example, if you design a kernel like this uh with minus ones here and a five. Yeah. And you have the PP uh the same image here. And uh just as an example, if you have a black uh a black 0.22222 with a, a same volume in the middle, then you get a value of two if you do that. But if the middle pixel is darker, if you're only one tint darker, it becomes straight away minus three. Whilst if it's one tint lighter, it becomes seven. So you can see how this, how this uh amplifies the, the, the difference and just picks out um there is one pixel in the middle which is slightly lighter than the rest. So this is like a fea a feature which you can recognize about the computer then normally does doesn't these pictures don't like negative numbers. So they use this value to make the negatives zero. And then uh what they do is slowly and slowly step by step, they, they called, they pool them in and they're called the maximum pooling. So if you got here a quadrant of, of um four numbers to take the largest one out they see to take the largest one out here, take the largest one out here and you got a smaller picture. Um And they do the whole process again and then they do the maximum pulling again and again and again. So you can see a lot laborious task. But, and at the end, you come to a uh uh what they call a fully connected layer and you can decide whether it's a cat or a dog. Um So these, these networks are quite complicated as you can see here. And they here, 12345 maximal pooling um steps in into most of these networks. So they take a long time to do. And up here, this is a picture for example of an um uh chest X ray. And it has been shown to be very useful to uh predict whether somebody has cancer or infection or pneumonia or um uh pericarditis or you know, any, any cardiac disease. Um It can pick that up uh on the chest X ray, uh a say cardic, I mean pulmonary disease emphysema or something. And you can, you can see that you can hear that I'm an orthopedic surgeon, don't look at chest x-rays very much. Um So you can see how this picture with the maximal pooling gets smaller and smaller and form a two dimensional picture slowly and slowly becomes a one dimensional picture. And becomes a number and then you can say it's infected. Yeah. Um, they normally do some preprocessing, uh, steps in, in, in this, um, uh, thing. And for example, if you're going back to our cats and dogs, a cat or a dog here in the middle, then the computer fairly quickly sort of like, just looks at the middle, but to try to enhance the features, what you can do is put a, you know, just change this picture by flipping rotating uh doing whatever with it, magnifying it. You can see here this, you would say, well, this could actually be a cat if you ask me. Uh but this, this same dog just in a different position. Yeah. And so, but you, you, you then, uh you then if you tell the computer that this is AAA dog, you force it to look at other features rather than, than just the face. Uh which is what we we as humans would, would normally do, I think. Um And so that's quite a big training set and they tested this in 2.5 thou uh 12,500 images. So a lot of images, um I'll, I'll skip this. Um So if you look at some of these pictures, you know, this is pictures of this data. So now everybody would agree. This is a cat, you can't even see uh the face of the cat. But because you know, we know that cats are, uh, having their collar on the X chromosome. So, you know that this is a female cat, uh, because it's, um, tortoise shell cat. Uh, but could this be a dog? Hm. Maybe this one looks a bit. But to me this, you know, this cat could maybe look like a dog. And here the cat's actually quite tricky to see. So you can see the, the problems, uh, with these are not easy pictures. Um, this to show you what the computer sees. The computer sees. An average cat like this and an average dog like that. It's even worse than an ultrasound scan if you ask me. And I certainly can't see the, um, the difference uh to that, but that's how the uh computer classify this. And I don't know if you can see it here, but 98% accurate. Um I just move on a little bit because I'm thinking I'm going a little bit late in uh time otherwise, um, just show you the giant cell tumor of bone we've done there. You can probably see here a giant cell tumor of bone. And actually, you can also probably see here there's a fracture through it and we've done this same image classification thing here. Uh But there, I don't have 12.5 1000 giants of tumors of bone because they are very rare uh tumors. So we only had 61 which is actually quite a lot um to train on a nine and 16 to um to test on. And then we did this. Then initially you see here, the blue line is the learning on the training set. Oh, almost 100% accuracy. Uh But then if you look at the red line, which is the validation that the test set, then you see, you actually didn't get more than 75% which is not really very good. Um I'll skip this because of, of, of, of time. Um, so, but if you look at, you think 70% out that is reasonable. But you may recall that the test set of the 16, there were only five giant cell tumors and 11 others now and, and the same percentage was there in the training set. So the computer very quickly will learn. Oh, if I can't, if I don't know what it is, I just call it no G CT because then I'm going to be most likely to be right. And, and that's how a computer thinks. Um, you, you would never, er, dream if you would see 16 pictures like that in a row that you will just, er, close your eyes and say no G CT for every one of them and just gamble. Um, but computers will just do exactly that. Um, and this is because the groups are not balanced here and if the groups were balanced then it would be much more difficult, uh, to diagnose how the computer sees it mystery big black box. There are some methods to try to extract his features. You can hear, hear from the fish, but it's, it's, it's certainly not easy, interesting. And this is the, the counter side of artificial intelligence, which I wanted to show you, uh which I think is quite interesting that is that he had the same giant or tumor of bone. And I have this little teaching session on, on it, which takes only about 10 minutes and I'll tell him it's epis, it goes right into the epiphysis. It's eccentric, not in the middle of the bone. It's expansile. As you can see here, there's usually a thin rim, there's no soap bubbles on this case, but there often are as well and it's obvious it's a big black hole with nothing in it. And then o on this teaching, the first thing is, and this is probably also quite interesting when you do exams. The red line is the people who have no teaching and the green line is the pa the people who do have teaching or had, sorry, teaching. They can see that. Luckily, for me, the people who have been tea who have been taught, they have a much better score than the people who didn't. But also the longer you take to answer these questions, these 16 pictures. Yeah. The longer you take your score actually doesn't really improve. So this is why you're doing exams don't sit there waiting, looking at the, glaring at the picture forever. If you don't know, you don't know. Um, and it's even worse for people who have no teaching because the longer they took it they only got worse and not better. Yeah. So, uh, don't, uh, no, on, on questions forever. Uh, there's no point. So if you look here at the scores with different categories, it is also very interesting as you would expect orthopedic oncologists, you know, so orthopedic surgeons who treat bone cancer, they do very well and score pretty much everything. Correct. Other consultant, orthopedic surgeons don't do quite so well. But yeah, they're not in the, they're not in the branch. So that's, that's understandable. The registrars we have, they do worse than the consultants, which is reassuring. But after teaching, they do, which is the green one here. They do just as good as the orthopedic uh oncologist. Don't forget that the consultant orthopedic surgeons also wear ones there, but they've dropped down again because they forgot that you could easily retrain them back up again to that level if needed. Yeah, no surprise. Radiologists and pathologists who look very regularly at x-rays do very well. Other medical, these were nurses in our, in our oncology clinic and they are very well educated nurses and you can see they, they do very well and they, and they, with a bit more teaching, they do just as well as the registrars, you know. So don't be too derogatory to nurses. Uh They, they can see these things just as well as you can. Um And then this, I find one of the most interesting uh aspects. These last two groups called nonmedical people. These are people, including my brother and a few other uh lots of people I know who are actually very clever but not medical. And you can see, they don't, I teach them but they don't see it. And that is because their mind is closed. Uh, here, the other group is actually I call it other. But these are actually Children, uh, Children and very young adults like yourselves, um, included probably. And they can see they, in, they score very low because they're just gambling. They're guessing they don't know what they're looking at if you don't teach them. But then if you train them, they'll score just as high, um, as the registrars and the trained nurses. You see that, um, because Children have an open mind and they learn and so use your education wisely, I would say and learn the things you can learn now and, and try to see these things because after a while your mind closes, um, I'll skip this because, uh, you're gonna get some time for these kids annoyingly. Uh, but they are actually very smart, uh, at this and then I'll just finish with, with two things to show you what, what that, because I don't want to be negative about a I, or machine learning because I do love playing with computers. Um, you neutrophils is one of the things pathologists hate counting and they're actually quite important in the diagnosis of infection, orthopedic infection. So, if you have more than five, uh, neutrophils by high powered field, then it's, uh, on the, then it's known to be infected. Now, here, the red ones, this is the special stain and that they call it red. They don't normally do that. But if you, you would say here is more than five, but do you wanna start counting them? It's tedious, isn't it? And this is only one field of a whole slide. So it would be tedious to going through that. Um, here is the picture people normally get, this is a piece of bowel as you can see here. Uh with here a blood vessel going through them. Um You could probably see that you can see the erythrocytes there. Now take the computer and you can see the computer picked up 1234 neutrophils and they're in the blood cell. And if you look there, you could probably see yourself maybe a bit small on the screen. You can probably see there proper neutrophils. Um If you take the ground truth, somebody, pain, pain has gone through all these neutrophils and, and marked them. Not me, I must say. And then you can see, um, that the computer missed these 12345 ones. So it missed half, but the ones it did, um, identified, identified correctly. So I think with a bit more training and a bit more work, actually, this will be quite useful that rather than, um, ra rather than just uh counting them yourself, you can get the computer to get an estimate of how much they are. And I don't think it has to be that precise. Similarly, if you have an MRI scan here of the knee, and you can see here the, the, the cartilage of the knee outlined, you could with segmentation, which we done here, you could segment uh the cartilage and you can measure and 3d measure the thickness uh of the cartilage that you can see here. Um It is about 3540 in the weight bearing area. So this is a normal knee. Yeah. So again, very helpful, I forget this. So just to summarize the artificial intelligence, I don't really like the word because I think it's a little bit pretentious. Um It is really more machine learning and model forming, which we are dealing with, with statistics. You really need to understand the model, the problem with machine learning, you really also need to understand the model, but it's difficult to understand. But if you're not careful then, so the computer says this, so it must be true. Um But that is, I think a little bit of the of that, it's a little bit dangerous and, and you always need human supervision. For example, in the cars example, I gave you also in the example with neutrophils because the one in the blood uh in, in, in the in the blood vessels, they actually don't count when you start looking at um infection because as you will well know neutrophils in the bloodstream is quite normal to have them. Uh and you certainly can't assess patients with the computer and patients themselves can't do this either with clicking boxes because this categorization is not as straightforward, um, as I hopefully explain to you. So I'm very excited about um uh machine learning and computer techniques, but, um, don't get too excited. That's all I wanted to say. Thank you very much. Call for that, er, excellent talk. Um, it's a really good overview of uh artificial intelligence in orthopedics. Um, it would be good if we could do questions now, but I think in the interest of time we'll probably move on to the next talk if that's alright. And then we'll do the questions. No, that's fine. I'm sorry. No, that's, that's completely fine. It's a brilliant talk and we have to do it justice, to be honest because it's a very contemporary topic especially now. So, um, what we'll do, um, we'll move on to the next talk and I'll just share my screen briefly. So, uh next talk will be delivered by, er, Doctor Emily Hall. So, Doctor Hall is an academic foundation. You one doctor in the West Midlands, north region, er, with a kidney test in orthopedic surgery and er, medical education. Um, Emily completed her, er, degree at Kiel University and did an integrated degree in medical education and you were talking about how to get involved in surgery, in particular as a medical student was also showing her own, uh, expertise and experiences as well. So Emily feel free to, er, share your screen now and get started. Hello? Can you hear me? Ok. Yup. I can hear you just share my screen now. There we are. Can you see that? Ok. Yeah, we can see that. Fabulous. I'll get started then. Um, so hi, my name is Emily. Um, and thank you for having me. Um, so I'm here to talk about how to get involved in surgery as a medical student. I've only, I was a medical student three months ago, so I've been through it, um, and I'm here to share sort of general tips of my personal experience. Um, so whether you're in first year, fifth year, you've got a kind of interest in surgery or you're very sort of, no, this is what you want to do and you're committed. Um, everyone can benefit from getting experience in surgery as a medical student. It is half of the degree. Um, so whether you want to just explore an interest, a potential interest, learn theory for your finals or for your exams, enhance skills like suturing procedure skills injections, which you can take into other specialties. Um, or if you're really committed and want to start preparing for future applications in surgery. Um, hopefully, today, we can just discuss some tips and ways that you can get that experience early on in your medical career. Um, so what we'll focus on is the course called training application portfolio. Um, you don't necessarily need this, there, there are other routes into surgery such as locally employed doctors. But, um, this is kind of a good way to tick off various ways and especially if you're preparing for a career in surgery in the future, it might be quite helpful. Um, a lot of these aspects on the portfolio you can start to get in medical school. This isn't about saying you need to get them all. I certainly don't have them all. Um, but it's just sort of encouraging things that you can do, um to get from thinking. So this will look at the 2024 application cycle. It does change a little bit every year. So make sure you keep on top of it. Um, and we'll focus on the course of surgical training because that tends to be the first one that you'll apply to post medical school. Um, this isn't to overwhelm anyone, there's no perfect candidate. Um It's just to sort of get you thinking about what you can do and what you can gain and whatever you do in sort of terms of approaching a portfolio, just make sure that you're getting value. You're not just taking points off. How is this going to develop you as a better future doctor and future potential surgeon? Um So the current course surgical training, self assessment, when you apply for course surgical training, you've got a self assessment score, um an M SRA exam and an interview. Um This is one where you can self score and you can get a lot of points behind you. And it's split into five main themes. So commitment to specialty presentations and publications, teaching experience, um quality improvement, and audit and training, qualifications or teaching qualifications and to briefly discuss it just because we don't have enough time to go over everything. So how can you build this as a medical student? There's lots of ways. So approaching the first um points of commitment to specialty, um you can attend various surgical conferences. These could be run by your local surgical society or regional societies or national societies such as a um bomba, things like that. Um And if you attend them, it just shows that from medical school, you're really keen and you're thinking about potentials. You can either just attend this as someone who wants to approach surgery and learn or you could try and present at conferences and get more points further down the line as well. Um From your clinical years, you'll have placements in surgery. It's a great opportunity to get some operative experience. If you go into theaters, ask if you can watch, if you can scrub in, you could get involved in procedures and you can log them on something called an E log book where you can sort of log the operations you've um assisted in. And that also goes towards the portfolio. Um Most medical schools, if not all during the penultimate year or final year of medical school, you do an elective. So this can be sort of four weeks, eight weeks and this is a great opportunity to hone into surgery. So you can use this to get further experience as well as your student selected components during medical school, looking at quality improvement and audits. Um These are really popular at the moment essentially. What you do is you take something, compare it that your hospital or trust is doing, you compare it against the standard. Is it meeting the standard? Could you improve it at all? And then you can basically implement something and then a and see if you've demonstrated a positive change. Um You can do this during medical school. If you're on placement, student selected components, you can ask in the department, are there any audits going along? Could you get involved in collaborating and data collection or could you even lead them and then re audit and close the loop? Um So it's an excellent look into a little bit of sort of research in academia as well. And improving those skills um in terms of presentations. So either from previous projects, audits, you can present these locally regionally, even internationally and you get more points based on how broad that conference is. You can either do this as an oral presentation or a poster. Um And you can also earn prizes as well, which look great publications is something that everybody tends to obsess over. I don't have any publications to my name at the moment. Um And I got into sort of an academic foundation program and stuff. So it's not the be all and end all, but it can really give you a lot of points when you apply for training. So you could write up a project, you could write up an audit, do a case study. Um As long as it's pub med cited, um that's what matters. And then again, you can get any experience during medical school. You can try and advance that finally looking at teaching. Um You can even do this during your preclinical years. So you could teach anatomy or surgical theory to medical students in the years below. In revision sessions, you could look at your local medical education societies and surgical societies to do that. Um If you want to go a bit further, you could set up teaching series yourself. Um And you could organize some spot a gap in maybe the learning at medical school or extracurricular stuff and make a series. And also you can gain teaching qualifications. So postgraduate certificates, diplomas, or even masters in sort of medical education, clinical education. Um and then you can gain further experience in teaching and qualifications. Um If you want to go that extra mile, say you're super keen, if you, you've already started um degrees and qualifications, the only degrees of qualifications that count at the moment are your teaching ones. But don't let this put you off other ones. So a lot of people indicate in um clinical anatomy and essentially what you do, just make sure you actually achieve things from it. So you might do anatomy and it might better your knowledge in that and prepare you for the future. Um So you could do previous degrees that you can bring forward or you can intercalate during medical school. Um Even though points aren't counted as much as in the past, I still think it's a great option if you can and if that's something you're interested in, and then finally you can get awards so you can get local awards by societies up to national awards, these can be through abstracts, um presentations, even artwork sometimes. Um And that always looks at as well. Um Finally, if you're interested in sort of leadership roles, you can join um your local surgical societies as a member or you can go on more sort of leadership societies like the BM. Um it doesn't necessarily have to be surgical. It can still really sort of bolster your applications in the future. Um Essentially whatever you do, make sure you get evidence of what you've done. Um Because then you can log it in the future and again, just make sure that you're actually achieving something from it. And it's better in you as a future doctor, a future potential surgeon, because that's ultimately the main thing rather than just ticking off boxes on a scoring sheet. Um So I'll just briefly go over my own experience. So 1st and 2nd year, I was largely preclinical. Um I joined my local surgical society. I would practice suturing on bananas and sort of pig's trotters as seen in the bottom photo on the right. Um I'd also do teaching. So I teach first year medical students anatomy when I was in second year and I carried that on. Um Second year, I also went to a conference on women in surgery and I met Professor Avril Mansfield, who was the first um British woman appointed to be a professor of surgery. And that was a really inspiring talk and really encouraging. Um going into my clinical years in third year and my surgical placements, I'd get involved in theaters, I'd scrub in, I'd um sort of get operative experience and I got to suture for the first time. Um And this is when I really sort of became keen in orthopedics. Um and that sort of has shaped my aspirations as such. Um going into fourth year I got involved in a bit of data collection for an audit. And then I also led an audit looking at developmental dysplasia of the hip. And it was presented um at the Oswestry Research Day earlier on this year in April. And I got to close the loop in that as well. And I learned to lodge in that process about audits and just about developmental dysplasia of the hip, which was really interesting. Um Going into fifth year, I became a member um or a committee member of my local surgical society at Kiel. And I was also on National Societies for Surgical Mentorship Society. I was a local representative for the British Orthopedic Medical Student Association. And I got to get involved with a lot of things like the key on national orthoplastic workshop um in the bottom left that wasn't even staged. I was just having a good time. So I was very smiley. Um And then I also um and I got to present my project at um a national medical education conference. And so that was to do with sort of education and women and orthopedics. Um And through that, I got to meet a lot of amazing women and mentors. Um So just a few tips to take home from this. Um so essentially just be strategic and be organized and be realistic with your time. Ultimately, as a medical student, your studies come first and anything like this is just a further benefit. So make sure you're prioritizing your studies and if you have time and you think it's feasible, you know, obviously go and get those extra things that would look great on your portfolio and give you experience and help each other out. You can do projects together, you can get involved in multiple because if you share the workload, it makes it a bit easier to squeeze into your placement schedule and revise it for exams, um be open minded. So some projects I've done on in orthopedics, they're in colorectal surgery. I've just done data collection for that with star surgery. Um But ultimately, it, it helps me become a better data collection. It helps me become more interested in surgery and learn more. Um finding mentors is really good. They've been through it, they can give you a more sort of personalized experience and they can help you out and sort of point you in the direction of new projects. Um Like I said, the log book. So once you start scrubbing into operations and you get involved in assist, make sure you log those because the number of operations you assist in will go towards towards your portfolio. And like I said, keep evidence. Um And finally, as a medical student, your university societies are your best friend. So join your surgical societies, they've got loads of events running. Um And you can also, if you notice a gap, you could even make your own University Society, which would look great. Um So that's my talk. Um Thank you very much for listening. I don't know if we'll have time for questions now, but happy to sort of answer any emails or that's my Twitter if you've got anything else. Um And yeah, thank you very much. Thank you very much doctor for that. Er, great talk as well. That was a really good overview of uh how you can you get involved as a medical student and also um meet the different demands of each part of the um portfolio. And it's important to know that the portfolio basically changes like every single year. So even though they don't have interplayer degrees right now, literally a couple of years ago, they did and there's nothing stopping them in like five years or so when we get to that stage or um to for them bringing it back. So it's, it's also a really good opportunity to get that dedicated time to pursue other projects as well. But yeah, we'll leave the questions uh to last. So thank you for uh giving that talk and then we'll move on to the next talk now, which is about how to get involved in research as a medical student. Um So I'll just share my screen. OK. So you should be able to see my screen now. So um I'll be giving this talk as um I'm finding a mo at at, as I mentioned before and I recently completed a mil which is a research degree as an intercalation in big data clinical research. And so what I've done is used um national datasets from the United States of America er for Prognostic Factor research er using the big data and data science methods. Um and these datas commonly have anywhere between like um 5 million to even up to 200 million patients um to analyze. And so it's very, it's very, quite clinical and dry lab for research. And I'm currently on the Star Surge steering committee, which is um, why I'm pri primarily driven to give this talk as well. And I can, I'm gonna be explaining to you um how to different ways you can get involved in research and the different types of research essentially. So it'll be a quicker little summary of all of this stuff. So in terms of what is research, I'm sure a lot of, you know, already, but generally it's the gathering a day to answer a specific question. Um, and it's generally done by people at all stages of their career as well and without it, well as you can tell, nothing, nothing really would actually be advanced in medicine. So it's actually quite important um to get involved in research and especially if you're interested in it even better, essentially, I suppose if you're interested, that's why you're kind of er, attending this talk. So hopefully it's quite useful. So in terms of the types of research. Um So you've got broadly um two different types. You've got the lab base, which can be um divided into the wet lab and dry lab. So dry lab is generally what I um I'm engaged with. Um And you've also got clinical based, which um generally involves um patients um uh most of the time anyway. Um so in terms of getting into this in a bit more detail, so what lab research is obviously done in the lab setting and you're generally working with biological matter. And this can include um cell lining, tissue cultures, small mammals, chemicals and drugs. But experience is usually required for these um for these environments, but it's not necessary. And sometimes you can do um student selected components or uh summer internships when and get lab um experience, but generally, it's less common and um accessible amongst medical students due to the time constraints. Really, I mean, that's why a lot of people can do it during their intercalated years as well if they choose to do that. So in terms of the dry lab research, often it is computer based. So really all you need is a computer and internet connection. Um And generally, it's about using data that's already collected and data that are on like registries and national data that's like I've been using to analyze and look for trends. Um But it can also be analysis of actual papers themselves in the form of systematic reviews and meta analyses. Um and generally, it can, so it even can be done remotely as well. It doesn't always have to be done um Right, in a lab environment, et cetera, which makes it quite accessible for a lot of people. And it's actually quite um low risk research in the sense that all the debt is already collected or published. So you don't necessarily depending on the study, you have to go through like ethical approval or be the main investigator of the study et cetera. But there are some difficulties that can come in in terms of like critical appraisal or systematic reviews and data analysis can be quite complicated to learn and can take quite a lot of time and these um studies are quite, er, should be quite vigorous, er, rigorous as well in terms of their methods. However, if you, if you can do all that stuff well, and you've got a great supervisor and you have a good research question, it's quite hard to go wrong with er dry lab research overall. So in terms of clinical research, as we've mentioned, this can be stuff to do with like conducting surveys with patients at outpatients clinics for, for instance, um consenting patients for involvement in clinical trials or auditing local uh prescribing practices or just doing audits in general and interest in that. Um um Emily mentioned that the colorectal cancer, that's what we did for um our recent star search. Um collaborative project. And these things are a good way to start getting involved, especially in the um clinical environment where you get a dedicated um supervisor um or supervising consultant or registrars or people who are experienced in this to start introducing you to the world of research, essentially. And it's a big world of research to be fair. The whole of academia is because there's so many um different steps from an idea to paper and you can get involved at any stage of this process. But most of the time people get involved in ideas that have already preexisted from um from consultants and supervisor that might not have got around to do these types of studies or you can actually um create your own kind of ideas for research. Um But again, it's probably best if you had a supervisor for that, who's an expert in that field to see if that research is really worth pursuing overall. So why do you, why would you get involved um as an undergraduate? So generally, the main motivation is, well, you want to change clinical practice in order to help your patients. Um and you can do that by improving your understanding of diseases and clinical processes and research acts as a kind of wave as a proxy for you to learn more things more specifically. So when I did my first research projects, I learned more about those specific areas and I hadn't even encountered them in medical school yet. And I think that's quite important. Um, obviously you want to stimulate your academic interest, but also as everyone wants to, er, as everyone knows, they wanna be able to get into their specialties in the future. So let's be um realistic. A lot of people wanna do it because they wanna get involved in a specific speciality and that's also fair enough as well, but you also learn valuable new skills. Um and that can range from anything such as data collection, presenting um analysis skills using different Softwares such as SP SSR Python. And also you learn writing skills if especially if you're writing a, a manuscript up for for publication or you're collaborating on other people's works, et cetera as well. So you really learn a vast um variety of skills and like you said before, the surgical drain jobs are training jobs are quite competitive. So and research does score quite highly on applications. And for another reason, as we always say, the GMC kind of documents in terms of outcomes for graduates. So newly, newly newly qualified doctors must be able to apply scientific methods and approaches to medical research and integrate these with a range of sources of information used to make decisions for care. And again, that can be across that whole of the um translational pathway of research and the different stages of research as well. So in terms of questions to ask before you started, first of all, what are you actually interested in? I find it's actually way better to do a project that you're interested in mainly because you're mainly engaged for longer. You're more likely to be committed to it. You might like to read around it and become more of an expert in that specific topic area. And II agree with them when you're saying that don't be closed minded. It's good to be open minded with this sort of stuff, but make sure you're actually interested in it. It doesn't have to be the exact specialty that you like. But as long as you're interested in the topic matter, that's, I think a very big, um, weighting of importance in terms of your decision making as well. How much time are you willing to actually spend on this? So there's no point doing great research, but you've got no time to actually study for your finals or something. So you have to be, um, wary of maintaining the really good balance as well. And also I think that comes with having a good supervisor who understands that you've got auto balance actually going through medical school or early, um, placements as, um, foundation, you et cetera. And what are you actually want to is what is your, what is your actual goal? Are you aiming for CV applications? Are you deciding on a career in academia again, that's important to discuss with the supervisor as well because it's in, uh, in my opinion, it's important to make things nice and clear early on about what you want to do and in terms of asking around. So senior colleagues are actually really useful to help you get um to get these experiences and prevent you making the mistakes that they have. So I've always found talking to, talking to others who are more senior than me. Um I'm less likely to make the same mistakes that they do because they've basically been been there, done it essentially. So generally, when thinking about how to get involved as well, there's you can use the three PS in terms of the success of a project. So generally, um that's person project in place, in terms of person, it becomes about you. What is your previous experience? What are your interests? What is your motivation in terms of projects? It's generally, what are you actually working on? Is it novel interesting? Is it actually achievable in the first place or was it too big of a project or is it specific enough to make a good impact as well? And in terms of the place, I think this is probably one of the more important ones which is your supervisor and your wider research team and their resources, cos a supervisor, in my opinion, can be the make or break of a project because they can be the ones that really drive it alongside you as well. Or it can be a case of if a supervisor doesn't actually respond to you and isn't receptive to you, then that's a major red flag in terms of if you can actually complete a project in the first place. So in terms of the ways to get involved, and this is a generic diagram, but you can do like intra curricular research, which is part of the medical school, like in ca obviously, we do like the A Mr S and the SF ES et cetera. But you can also get involved in extracurricular as well such as a spy project, which I'll get onto um in the future as well and also intercalation, which is technically you could say extracurricular or even um independent research as well. So on the side, essentially of medical school as well, and that's what I've basically outlined here. So your sse intercalation and electors are perfect ways to get involved as intra and extracurricular summer projects. And there's also collaborative research that we've mentioned um already in the form of star surge as well. But the main thing is finding a good supervisor and this generally is a clinician at your university who's experienced and actually has a good track record. And the only way the the accessible way to measure their track record is actually to do with publishing. Um I think that's AAA quick way of saying, OK, this person actually has outputs, but that doesn't tell you if that person is actually very good or not in terms of working with a student. So it's always good to reach out to people that they've worked with before and see what their experiences are. If they're actually supportive, if they were obstructive or if they um open up to new opportunities, et cetera. So I think publishing is more of a, a face value kind of assessment and it's good to actually talk around and also talk to the supervisor as well to see if you get the good vibes from them as well about research projects. And in terms of the daring part, that is the hard part contacting an unknown person. And I think people find it a difficult task, just cold emailing someone. But I think really that's if you don't know anyone, that's the best thing to do and you have to find the right contact details and be be mindful of that. Not everyone will reply and be be at peace with that essentially because it can be that you're sending lots and lots of emails and not getting much replies. A good thing to do is create a template, email, basically outlining who you are. What year you in the reasons you want to undergo research, your current experience, but it's ok to say you have none. But more importantly what you actually want to achieve out of it. And um also what is your motivation for working with them or what have you seen from them? Why do you think that they would be a good fit um to help you start off in the world of academia as well and also draft off a short CV. Um So it's more of a glance for the supervisor to see and what your experiences are and how you can both can collaborate in projects together as well. And I think um it's a really important way to get involved in projects. Just cold calling um co colder emailing clinical academics as well, especially, but also you can approach commissions after a lecture. Um And you can use the points from the lecture as a conversation starter to be fair. Um Or, or you can actually just search them up afterwards and see what kind of um what works they've been involved in. Um Also, it's good to explain your own interest and ask for these opportunities. It's the hardest thing is asking for the opportunities, but you have to do that to get yourself out there to put yourself out there. And then that's the only way that they can actually help you. It might be a case that they might refer you on to someone else who's more receptive to taking it on students or has better opportunities, et cetera. So it's um quite important to just at least start the conversation instead of thinking about, oh, how am I gonna do this? Just at least just get it done in a way. So, in terms of my own experience, I did my first um project um just after second year, um, er, in the summer of second year, which was, er, an aspire project from Kyoto. If you don't know AI is a summer studentship program where you can put up to up to 1250 lbs of re of um funding over a 4 to 8 week period um to do a research project and no one says that that project should be finished in that 4 to 8 weeks. You can extend that for as long as you want, you can work on that project and finish it off well into the next year if you want. But it's good to have that dedicated research time and funding cos it's nice to have some funding with that as well um to actually do a project. So for my first project, it was actually on a cohort study that was local to around the Staffordshire um area, which is about polymyalgia, rheumatica and fragility fractures. And um what we did was we analyzed the cohort study dataset. And thankfully, we were able to achieve a national presentation, a publication and also a national prize from that um dataset as well. So if you're someone that's not interested in academia, but just wants to get something on the board, it's good to be prolific with your um research as well. So make sure that you've got a good project, you're able to submit it for conferences and publication and maybe you'll get like a bunch of domains actually sorted, locked off in one in one go. In second year. I engaged slightly into collaborative research where I was at da uh helping with data collection for uh a urological based um audit of um urology teaching and the curriculum. Um Again, that was a kind of loose involvement. It was more so me dabbling to see what collaborative research was like. Then in aspire in, I did aspire again the next year, mainly because I enjoyed it. The, the year before cos I was someone that was actually against research. I didn't actually like research and the concept of it and getting involved in it, but I thought I should get involved. So I just don't discount it um without trying. And that was the main reason why I actually tried it in the first place. And that got me to do uh even more and try again. And this was where I first did a project um with big data, analyzing a national data. So, again, in patients with polymyalgia, rheumatica, but their outcomes um when they're admitted with acute myocardial infarction or heart attacks. And again, this one was um published in um in a, in a um American journal as well. Um quite gratefully. And since then, I've worked an independent research with my same supervisors and that went on to form the base of my intercalation project, which was um fully funded for the er where we did um, um, more analyses using the national data sets and we created like three or four, papers that have all been, um, published or in publication or under review right now. And then, um, towards the end of my intercation during my ation and start of this year, um, again, we've been, I've still kept that independent research going along cos I think it's good to keep that relationship going and keep pursuing your interests. And my supervisor have been excellent. And that's why I've kind of stuck with them for so many years. I've also engaged with Star Search in, in the form of collaborative research. So I was a regional lead for the Apollo study, which was auditing the management of colorectal cancer patients and their management. And Emily was one of the day collectors for that. And um there are good like nine day collectors that were all medical students from Kiel and hopefully they'll be getting their first um collaborative publications and presentations as well. And again, we've also got elective coming up this year. Um I haven't fully ironed out the details of exactly what I'll be doing, but that's another opportunity where you can um explore um research opportunities alongside doing whatever you wanna do for clinical or it could actually be the main um base of your, of your elective depending on if the medical school that you go to allows it. And another good resource is something that's for free, cos II only like free resources to be honest. And that's er, incept dot ac dot UK. So that's something backed by the Royal College of Surgeons of Edinburgh, where there's a bunch of modules that you can do for free. And here you can see examples of those modules such as how to get involved in research. Um It's more like informational modules about clinical academics and then also about what about knowledge as well, sort of introduction to study design, observation studies, clinical trials and also the statistics behind them in terms of the core concepts and analyses as well. Again, and there's one thing I want to reaffirm as well. You don't have to be great at maths or statistics to get involved in research. It's just about getting involved in the first place and learning what you're doing. I think it's the most important thing is about getting involved. But yeah, I think that's everything I was gonna go through. Um feel free to. You can put questions in the chat, but we'll leave questions to the end and then we'll move on to the next talk and then um at the end, we'll hopefully have all of us together that can answer questions. So if it's alright, we'll move on to the next uh talk now and this will be delivered by er Doctor Nalla Nantha Kumar. So, Doctor Nantha Kumar is a core fellow in Children orthopedics at Adam Brooks Hospital and anatomy demonstrator at Cambridge University. Er, so Naka, er, graduated from the University of Bristol in 2021 and completed the academic foundation program in the West Midlands. And Doctor Nan Kumar will provide an overview of the academic foundation program for those of you that might not have had much information about it yet. So, um Doctor Nant Kumar, you can feel free to share your screen. She, hello, good evening everyone. Thank you, Bela for the very humbling um introduction. And yes, it, we listened to very great talks earlier today regarding how to get involved with research as a medical student. And I guess I'll take you one step further into how to get involved with research as a foundation training. Uh I think Bala uh summarized my journey so far quite well. Essentially I did my academic foundation program. Can everyone see my screen, by the way? Is it working? Yeah, you can. So I did my foundation program where Emily is actually, um, and the academic component was so West Midlands North Deanery and the academic component was at the Robert Jones and Agnes Hunt Orthopedic Hospital in o History. It was quite a good learning experience. And essentially in this talk, I will tell you, um, basically tips and tricks on how to get into academic foundation training and then how to make the most of the program itself. And as Bala said, I'm currently an anatomy demonstrator and fellow um in Cambridge university hospitals presently. When do, when do you do the academic foundation program? Um the NIH R basically set this up essentially to start you off on an academic career. The first question you need to ask is why do you want to do research? Many people do research for many reasons. Um It could be, you know, just to tick the box, which is, which is not wrong. I think it's very important to tick the boxes to, you know, enable career progression and get into the specialty you want to do. But at the same time, sometimes um natural curiosity. So wanting to answer a particular novel clinical question, something that's going to improve clinical practice for the greater good of our patients. That is also a very good reason to get informed research or some people may just like data, data analysis, statistics or wet lab research. As Bala was saying, so many reasons are why you want to get in research and this is a good integrated clinical pathway where you can get more research. So essentially you go to medical school following that you do an academic foundation program, it's not the be all end all you can still get into a research career without doing an academic foundation program. I think that's an important point to appreciate following the academic foundation program, you aspire to enter into clinical or specialty training. There is either full time clinical training. Again, you can do research as a part of that as well or you could do an academic clinical fellowship, what is an academic clinical fellowship? It is essentially an integrated training pathway where you have designated clinic. So clinical and research time with an academic supervisor in a unit that is well equipped to support you doing the research that you need to do. And essentially that's to help you, you know, secure funding for phd uh eventually becoming clinical lecturer and you know, pursuing a research interest within the domain of whatever whatever you're passionate about essentially. So academic foundation program essentially is to prepare you for that a rigorous academic journey to get into academic care for fellowship, essentially across the UK you have specialized units of application, they call themselves. Now when I applied three years ago, it was, it was um basically academic foundation schools and you're only allowed to apply to two. So the deadline for that this year has passed now, I'm assuming that all medical students here will be applying in the future years to come. And hence, I will be giving you tips on how best to maximize your chances when the time comes. Um So you're only allowed to applications which makes it a bit tricky. So I think research into what programs you are quite interested in is very important to maximize your chances. Uh As Emily was saying, you don't really need to be doing research in a field that you wish to pursue in the future, you could be interested in cardiology research, but later on decide you want to pursue trauma orthopedics and it's all transferrable skills. So I guess the main important question you need to ask yourselves at the outset as a medical student is why are you applying for the SNP? I'll tell you my reasons very shortly of why I applied for it. The UK FP have published an application handbook and the uh every year they publish a new one. It has details of the timeline, the interview process, what schools are advertising posts, academic foundation posts, how many posts they are? So obviously, the school only has two posts and you know, it's a competitive school um perhaps look into other options as well because you only have two shots. It's a bit like applying to medical school, you're only allowed four options. And if my top option was Oxford second was Cambridge UCL and Imperial. I'm less likely to, you know, secure, I mean, basically putting all eggs in one basket and hoping to get a very good B mat score to secure a seat. So be very strategic with the application process. And one new thing that has happened with the academic foundation program is that they've introduced streams so you could do an academic foundation program targeted towards research. So when you do get your academic time, which is a placement in um doing a foundation training, foundation training is two years and you have four month placements in each year. So six placements in total. And the way it works is that it varies again from school to school. But in the West Midlands north, in the first year, you don't get any research time at all. However, in the second year, you have one research block where in that four months, 80% is clinical, uh sorry research time and 20% is clinical, then it swaps over where you do 80% clinical time and you still get 20% research time. So it works out to be like one day a week for instance. So that's the research um stream. You could also have education and training. So there are programs even in the east of England at the moment. So the education program where you get involved in teaching at the university, they fund you to pursue APG cert in medical education, which is again on the course surgical training assessment criteria. But then again, um obviously, you're doing the PG cert to enable yourself to become a good medical educator, understanding the principles underneath why you teach the way you teach and how better to improve yourself. Because at the end of the day, as medical students now you are being role models to the junior medical students. Uh people who have yet to enter medical school as you become a foundation trainee medical students would be looking up to you for teaching, for training. And that keeps on going, that keeps going on as you progress in your clinical period. There's a third stream leadership and management as well. So it's important to think of why you wanna do these choices because that's what you're gonna be putting in your application. Each foundation school has a designated website that tells you about their program, the potential opportunities, the research that they're doing, the supervisors and the specific streams of research that they have. So for instance, in Cambridge, it's, it's um they have 24 posts. There is medicine, neurocritical care, Neurosciences, surgery. And um on the website, it, it clearly states who are the supervisors that will be uh will be given to you or assigned to you when you do successfully get into the program. And the application process is what I wished. Every time when I am asked to give a talk like this, I often ask myself what I wish I had known at the stage of application, you know, five years back or 10 years back. And one thing that I didn't really get at the time or II hope to sort of share some insight on is the white space questions. So essentially the application happens on a program called Oral, which is online, you can all sort of create a login and log in to it. And when you open up your application, it asks you for your, obviously your qualifications, which medical school you went to? Uh what your grades are what your decile is. That's, that's quite important as well. Uh Whether you take it or not, what your current outputs are. So, have you published any papers again? Not uh essential but desirable? Uh What presentations have you been involved with any sort of research methods that you're involved with? Then it comes to a section uh called the White Space Questions. It's a bit like the personal statement that you need to write for medical school. So one thing that you sort of need to start thinking about is that it will ask you about uh what research experience and achievements you have had. As Bala me was saying systematic reviews, great way to publish a easy meth methodology to get involved with because number one, you don't need to be an expert. Everyone is a beginner at some point, you can learn how to do it. However, after publishing 10 systematic reviews, for instance, I haven't published 10 but giving you an example, the answer that you'll be able to give for this question would be quite limited. So whilst your goal would be to get involved with one, I think it's important to diversify your portfolio as well. So get involved in things like co studies or um survey analysis, qualitative analysis and diversify the types of research methodologies that you're getting that you can get involved with. And again, obviously submit to as many conferences, go and present education and teaching as Emily was saying, get involved with societies. That's the best way to collaborate with your teammates and basically organize teaching events and um sort of be able to demonstrate that you're committed towards becoming a good clinical educator in the future leadership, university societies again, are the best way to get this done. Um It's again, not the only way I think being the leader of your football team at university or any form of extracurricular activities, it's all transferrable quality. So I think teamwork and leadership is something that can go beyond something medical, it doesn't have to stay medical. If you initiated a volunteer sort of scheme of some sort A B scheme in your university, that's all very important. So these are the sort of four domains that you sort of need to start thinking about because when the time comes to apply, you basically need to write a personal statement detailing how you can demonstrate that you are going to excel in these four domains. The uh points of preparation is that at the interview, the interview is gonna be divided into two points uh with the academic component and the clinical components, clinical component. I think getting to grips with your A two E assessment, knowing how to assess a patient on the ward. Again, that is why clinical placements in medical school is very important. The more time you spend on placement, the better you are, you're going to be at it. Start thinking like a foundation doctor. So don't say I'd ask my colleague, what would you do as a foundation doctor? Because essentially as an SFP, they want to make sure you're safe on the wards as well as be able to commit to an academic career. One of the main things that I think is going to be very useful for you to prepare for the academic component is obviously understanding that each study has AP O component to it and that will help you understand the study design. So what are the patients involved? What's the intervention? Are they comparing it to anything? And what the outcome is ca is a very good website because critical analysis or critical appraisal of papers is something that you will be tested on. I was given the abstract on the day during my interview. However, some schools sort of do send the abstract in advance and cas basically tells you what particular things you need to look for to critically appraise a paper. And I think that's very important because you familiarize yourself with those questions, you'd be able to demonstrate that you are actually thinking about the abstract as you are explaining the abstract to the interviewers and again getting to grips with basic statistics. So what is AP value? What's a type one and type two, a sensitivity specificity types of studies? What's an RCT? What's a co study? I think that's all very important, James Lind Alliance is important because it essentially tells you what the priorities of research are. So, if your department is, you know, involved in XYZ research and that's listed on James Lind Alliance, it means that that research is very highly publishable or again going to basically impact patient care. It's very important and um groundbreaking. So I think that's something to pay attention to pros and cons. So the academic foundation program, obviously you get academic time as a medical student, I didn't realize it at the time, but I had a lot of time on my hands. Um maybe could have spent it a bit more wisely as well. However, as a foundation doctor, you'd be on the wards and again, some wards can be really full on, you'd be working full time and you might not have enough time to do academic work. That is why EFPS or SFP. Now, as they call it is very important academic supervision. As Bala Met said, I think finding a good supervisor, I was very fortunate to have a very supportive supervisor in a history. However, um not, it's important to be able to have a good working relationship with your supervisor. And I find that when your interest aligns with your supervisor, that is when there is going to be maximum output. So when you apply for these programs, look up your supervisor, put them into PUBMED their names and look into what research that they've been publishing and see if that research aligns with your interests, see what their department is doing as well, what their phd candidates are studying. And I think that will really help the other benefit of doing an AFP is that you would have centralized teaching. So at Kiel, we got given the opportunity to pursue a research methods, masters level course, which I think was very useful. So basic statistics, understanding research methodologies. And I think that has stood me in good stead to pursue further research moving forward. It's also self l so you can really determine what you wanna do with the academic time and uh provided your academic supervisor agrees to it. So essentially you could basically do a lot with that time and no one is going to stop you from doing more. There are cons so there will be less clinical exposure because you would have academic time. So it means that when you, you do need to pass foundation training, it's not just um turn up to work and, and, and you, you know, pass over the certificate, you didn't need to prove you're competent. And um that means you have less time to be able to prove these through work based assessments and, and all sorts of things which you will hear about during your foundation um induction. So how to work towards the uh S FP as a medical student? I think as uh Bala and Emily talked about this, I won't touch on it too much in the interest of time, essentially whatever SSE S you do, no data is not worthy of publication. Again, negative results may be a bit difficult, however, definitely aspired to pursue SSE S that are research focused, literature reviews, systematic reviews for instance, and aspire to publish them. Try as hard as you can. Dissemination is key go to as many conferences as you can. Um And I have this philosophy of you throw 10 darts and at least one is gonna hit the bull's eye. So definitely submit to as many conferences and you will get accepted in at least, you know, some of them teaching, organized educational events. It's very important to get them accredited as well. And I think Royal College of Edinburgh are very supportive in this and each region has something called the Foundation trainee Surgical Society. And that's basically uh foundation trainees who organize surgical surgical educational events. And I think that's a good way for medical students to get involved with their um sort of seniors and organize good quality educational events um work experience again at Russell Group Universities and any big university, there's always a lab, there's always research going on. So I think during these SSS period or summer holidays, you could, what I did was I just basically wrote to the rheumatology lab and said, I'm interested in musculoskeletal science. Can I come observe one or two wet lab research techniques? And that is what I did and it worked quite well, because it gave me insight into how these methods worked when I was just reading them on a piece of paper and it didn't really make sense. Um, examinations cause are very important. As Emily said, your examinations are the most important thing at the moment. So focus on your medical education because you need to be good doctors at the end of it all. However, when you do have free time, it's good to utilize these free time in the aforementioned methods, audit and research, make sure you know the difference between what an audit is and what a research is. Both domains will carry you far and both domains are different. So that's the main important thing to appreciate. Audits are very easy to get involved with. Basically look up a guidance or a guideline that is published nationally or internationally and see if that guideline is going to help your local hospital provide better gold standard care. And that's essentially how you get involved with and contact a supervisor. Each department has an audit lead. So message or sorry, not message. Um email, the audit lead say I'm interested in this particular project. This is the guidelines, the guidelines says we need to be um you know, documenting XYZ in our operative notes. However, um on my placement, I noticed that this was not being done. Do you think there would be scope for this to be to be audited in the department and being the audit lead, they would be quite supportive and they put you in touch with good supervisors or they might supervise you themselves essentially. So, tips and ticks this book, I found it very useful for my interview preparation at any stage, really. Um Definitely get it and start sort of, you know, familiarizing yourself with why you or uh why this particular foundation training or um what makes you interested in research? I think these are important questions to start thinking about. Uh one thing that I did is get to know what the research, uh what research is going on in the department by basically contacting the people in that department. Each department will have a list of people working there. You can always email them as bar was saying, sometimes you may not get replies, don't be disheartened. These are people who get probably 100 emails a day as professor K might tell you, but um you can email them again, do not uh obviously do not harass. But uh if they do not respond, there's always other people that you can email. One thing that did help me was I basically called up the ward and said, uh do you have XYZ trainee on there? And, and the person happened to be working that day and I got to speak to them regarding the program and knowing the because they are going to ask you, why do you want to pursue this particular foundation training? And I think be to be able to say that I've already spoken to people in your program, they enjoy XYZ um you know, aspects of your uh foundation training. That's one good way to demonstrate you're actually committed and you're going to be a, a good resource to the department. Um And preparation is key. I think preparation is the most important thing when you do enter your academic foundation training program, it is basically like an elective, a four month long elective where you have a supervisor and it is very fluid in the sense that you can literally pursue whatever you want. So planning is key once you know what's going on in the department and you get in touch with the supervisor early. When the four months come, you would not be applying for ethical approval or you would not be planning a project or, you know, searching up the literature, you'd basically hit the ground running and I think that's very important. Uh Thank you very much. That's my email more than happy to answer any questions or if anyone wants to get in touch uh happy to help them. Thank you. Uh Thank you very much for that er great overview, Doctor Nantha Kumar. I think it's very important the specialized foundation program because especially at Kel, the, we only get like one lecture that's half an hour in like the middle of fourth year. So um it's quite late on that you even get any insight into the specialized foundation program. And even then it's a very brief um overview in the first place. So thank you very much for providing that information and providing your contact for people to get um in um involved in as well and um ask you questions, we'll like I said, we'll leave the question to the end. Um And then we'll move on to the next talk first. Um If that's OK, which is, will be the final talk actually. Um So I'll just share my screen one more time to introduce our final speaker. Hello. So um Mr Shahin Hayan, there is a higher specialty trainee general surgeon in the West Midlands Deanery. Um main interest in um general and Hepatobiliary and pancreatic surgery. Um She has authored over 100 and 50 peer reviewed surgical publications with particular expertise in systematic reviews and meta analyses. And this talk, er, Miss Habra will explain how he balances a productive um academic life with the busy demands of surgical training. And I can say that as well because I actually approached him, Miss had the last year about how to actually do meta analysis and he took the time out of his busy day to actually show me how to um do a meta analysis and kind of ran a project with this. So this should be a really good er, talk overall as well. So uh feel free to share your screen. Um Miss Taj Banda And then Pete came on. Hi, everyone. Thanks following me. Thanks for the opportunity. Let me just er, share my screen. All right, can you see my screen? Yeah, we can see a screen and it's now full screen. Yeah, great. Thank you very much. Uh, thanks for the opportunity again. My name is uh Shine A, I'm one of the ST eight in uh Hepatobiliary surgery and general surgery in West Midland. Um, I'm going to talk about surgical training and academia and I really hope uh that uh some of you will find the information that I give you reassuring, particularly those that haven't done anything in research uh so far. So of course, I would, I don't have any conflict of interest to uh declare uh I'm just going to start with something very important. So my first name in uh Persian means Falcon and uh trust me, if I take my hat off, uh you will see me very similar to this picture. And interestingly, uh my interest in surgery was stimulated when I heard uh and story about Falcon. So the story was about a group of healthcare professionals who were asked to provide uh an opinion about a flying falcon. So as usual as to respect primary care, the uh G an experienced GP uh was asked to provide an opinion whether this flying bird is a falcon or not. So the GP said, OK, I think uh I, he refused to provide an opinion. He said we need to refer this bird to a birth specialist to see uh whether it's the falcon or not. And suddenly the medical uh consultant uh comes in and he says, uh oh, there's no point the specialist is already here. So I think we need to uh take a full medical history, social history, medication history, take some blood. And then we investigate to see whether it's a falcon or not. Then the radiology consultant comes and says, no, I think that that is just a waste of time. Just give me to put this bird into a scanner. I tell you whether this is a falcon or not. And suddenly a psychiatrist starts laughing and says, uh I think there is no point for further discussion. We all know that this is a falcon but does he know himself that it is a falcon? And suddenly a surgeon shoots the bird and ask for a pathologist to go and find out whether it's the falcon or not. So this uh uh decision making skills made me interested in surgery. And uh of course, like on the other specialities, you uh heard about this, you hear about the symptoms of the patients and then uh you read about them and then later on, you will deal with something more complicated surgeries that uh when you start the surgical carrier, you, you will never think you're going to achieve this. And then uh uh everything will come together at the end and you will be able to do these procedures for any subspeciality. I just brought an example in general surgery. So you'd see the different types of surgeries, open surgeries, laparoscopic surgeries, and robotic surgery. In general surgery. I'm sure other specialist has got their own minimally invasive way of treating uh others, other uh different pathologies as well. But the point I want to highlight that uh in any specialty that you want to go, uh There is a diagram that tell you what is the next stage, what are, what are the phases that you need to go through? And uh there are some uh dotted line. There are those uh phase, for example, a transition from phase one to phase two and phase two, phase three. There are some barriers and those barriers are, for example, if you call them uh interviews, there are some checklist that you always need to uh go through and make sure that you are a competitive person. And uh first of all, it's uh in terms of the research, it is not, it's never late, it's never early. So you can do it any time that you want. So, uh I'm telling you as a uh uh probably an a senior trainee that uh most of my colleagues, not only in my speciality uh at the same level as myself uh and in other specialities and even a lot of consultants have not published uh more than a couple of papers. OK. So therefore, uh you can easily go through uh all of these stages without having a publication or have a great portfolio. So bearing that in mind that uh you are going to be interviewed and in those interview, answering one question, right, uh or two questions, right can make up 10 publications for you. So they can be equivalent of 10 or even 20 publications. Nevertheless, uh I do not hesitate to encourage you to get yourself familiarized with uh research. Um uh Because some you will need that at the end of the day. So yes, uh uh it's not compulsory for you to have a research when you go to uh to know about research, when you go to phase one and phase two and phase three. But we are going to exit uh a training, you may need to have some publications very easy to achieve. It's not difficult at all. Others have done it. And uh I'm sure you will be able to do for any specialty that you are going to be involved in. Uh uh knowing about the research especially uh and the later stage of your training can make your life easier, can make you to understand your specialty very well. So uh uh I the take home message from this slide is uh it's never late, it's never early. And uh you don't have to have publications to go through everything. So the competition ratio uh ca can be overcome by a lot of other factors and portfolio. But portfolio uh and enhanced portfolio can help you to uh have a uh a more, more, more confidence to go through the whole process. So the icon of COVID I brought it here as a challenge because uh as we go through, we will have a lot of challenges uh in our uh uh next in the current stage of our training, next stage of our training. And also with the challenges that you have faced in the previous stage of training. So in terms of the uh you know, I like at the beginning of your, for example, second year of medical school, third year, fourth year, and I'm sure it's gonna be the same for your fifth year if you are fourth year or this first day of your foundation training. So there are a lot of uncertainties for you that yeah, I need to do this, I need to do this. I need to do this. There are 100s of uh uh items are uh in front of you that you're going to achieve. But at the end of the year, you will see that everything have come together and um and uh at this point, the challenges that you have experienced that you managed to go through them. So I think in terms of the surgical training, yes, you have got uh operative skills to develop. Uh you have got non operative skills to develop. And in order to um um stimulate your surgical appetite, I'm going to replace that picture of a stethoscope with something else that you may find it more appealing if you're uh you want to do surgery. And also the one of the factors together with all non operative and operative uh uh uh object is that you have is a research. So there is no doubt and I'm sure most of you are in the same, in the same situation at the moment. So uh when uh for example, others ex talk, talk about the great achievements. For example, uh I have published this, for example, yes, if you uh if you, if you haven't done anything and you hear about uh pa talk about his uh CV. So yeah, you, you, you, you, you, you think you're well behind. So and uh and yeah, and you see, I, my, my, my objective is to go through the next stage of my training. I want to, for example, become a consultant. I want to become a core surgical trainee. I want to become a surgical registrar uh or uh any other objective that you have. And sometimes you see the research as a barrier. OK. And uh uh that uh uh can be in your mind, it can bother you. I want to spend some time on this diagram because I think it is the most important slide of uh uh the day for me. And I'm sure it's gonna be for you as well. So any uh research that uh you are going to be involved or you have been involved, uh uh you uh the the study designs that you are going to have is going to be within this diagram. So um this diagram is known to have a level of evidence diagram where there is a hierarchy in the improvement in the quality of the study design. OK. So uh I see myself in a position to provide this advice because uh uh I'm uh I am uh probably very experienced academically. So I publish uh a lot of papers probably more than uh uh uh a large number of uh uh the trainees consultants in UK in different specialties. So, so I would, I would see myself in the, I'm, I'm the review of a lot of journals. So I see myself in position to provide this advice. And I think uh uh I really hope that uh uh that can affect your decision making or at least you find the information reassuring. I believe that you can enter the research ward in a controlled way or uncontrolled way. So let, let me talk about the uncontrolled way first. So there's no right or wrong, but uh uh I would call it controlled and uncontrolled way. So the uncontrolled way of entering the research is getting yourself involved in any of these studies. And I'm going to bring some examples for you. So I'm sure it's a very common sentence that you hear in your hospital from your supervisors, from the consultants, from the registrars. And a lot of people, let's write this case up. OK. So for example, they are, they, you are encouraged to do a case report. You, you are encouraged to, to write paper on a, a case and that involves talking to the patient uh getting a consent form, uh taking some pictures, making things interesting. And at the end of the day, you are going to uh uh provide a uh paper that is going to uh have a very low level of evidence. And most importantly, uh they are not going to uh uh uh you know, add anything for your future applications. So interestingly, case series, case case report, not necessarily case series, I have nothing to add to your uh applications. OK. Yes. Yes. From the medical school, the foundation training, which is not necessarily the most important barrier that you have it. Maybe you get some points after that, you will not get any points. So because it is uh it is not known to have a high level evidence. Although there is a lot of effort associated with that you need to and the leading journals in, in uh medicine and surgery do not, they do not accept case reports anymore and you need to pay that to be published. And at the end of the day you're dealing with a case that uh for example, at then you have no clue about it. So because it could be something very, very rare, it could be something very interesting or you may argue randomized controlled trial. So from now till you become a consultant, you are never ever going to be staying in one hospital more than a year. Ok. Even if you are a senior registrar. So if you're a foundation training, so you are going to be in a rotation for four months, you cannot do a random controlled trial, you cannot do a case control study, you cannot do uh a long term studies. OK. So, and uh and uh don't get me wrong. You are asked to uh collect data uh for a lot of collaborative uh groups. Uh You, you may ask to collect data for a uh hospital project. But uh uh and you are going to be as in the authorship among 100s of authors and you never ever going to be a primary investigator of those projects. And you're not going to have a uh uh dis that you're looking for. And most importantly, you are lost in that project because you have just collected data, which is not necessarily something difficult to do all of you can do it. So, uh and uh and this is something that I think you can go through that. But yes, you will have some publications via that. But uh uh unless you, you want a specific or projective research for you want to do an academic job. Uh If you want to do a uh uh you know, there are some reasons that, for example, you may get yourself involved in this uh uh trials. But uh you need to have, you need to allocate a specific time for them. For example, you go out of the training, do a phd for 23 years or, or uh uh going through the academic route can be helpful, but that depends on your interest. But uh for some myself, I've never done an academic uh degree, I've never done an academic job. Uh II, my job has always been clinical but is this produced a lot what I would call a, a control way of going to the research is uh going to, to start with a uh the highest level of evidence. So uh a a meta analysis or systematic review. So, meta analysis uh specifically, I'm going to talk about because uh uh they don't need any e approval at that was as, as it was mentioned before. And uh it is a highest level of evidence. So it is evidence that if you do for some metal of randomized controlled trials, the evidence that you provide is better than a randomized controlled trial which has been done uh has been taken for five years to be carried out. So uh it doesn't necessarily down stage that, but it is not going to down stage itself as well. It is going to provide a better level of evidence, meaning that guard plans are going to be made based on the findings of meta analysis, not necessarily from a single randomized controlled trial. Yes. Without randomized control trial or cohort studies, you will never have a meta analysis. But uh this is someone else doing this uh this uh uh projects and I'm going to tell you who would be a successful person to do these projects. When you do a meta analysis, uh uh you do a research that doesn't use ethical approval, you do uh a, a big team work. So therefore, you've got a uh you can be a leader of your own project. You uh write a paper that is going to be most likely published if it is done appropriately with appropriate title and with appropriate supervision. And it's going to be a uh you can use it as the evidence of uh leadership, those, those uh uh and uh the items that Emily highlighted, you are going to fulfill all of those with by doing one meta analysis. So you will, you will have a level one, a research, you can show, you can demonstrate uh evidence of leadership, you can publish it, you can present the outcomes in uh uh conferences very easily. And also uh after doing couple, you can teach others, you can use it as evidence of teaching as well. And interestingly, you can teach people who are more senior than you in any other area that go, it's very easy to learn. But trust me, a lot of people don't know how to do them. And when you teach them that's, you can demonstrate, you have done teaching for people even or more senior than you. When you do a meta analysis, you include all those studies that are in lower down for some random conor and other comparative studies, which means that you get yourself familiarized with what are the important elements in the randomized contro trials and cohort studies. After publishing some uh papers on Methanol, a systematic review and learning how to do scientific writing and the way and uh getting to know their study designs later on when you become established researcher, uh let say when you're ST six or so, you can do your own studies, you can do your cohort studies. There's, you know, uh you can, you should not necessarily do uh other uh do work for others. Uh You can do your own work. OK. So uh you can, you can establish your personality from the beginning. Yes, I want to learn a specific type of research like meta analysis. I'm happy to collaborate with others as opposed to doing jobs of others. And uh and I'm happy to get others involved to have their expert opinion. And later on, I have got enough expertise to do my own research. OK. So you don't do works for others. OK? For uh uh for uh for different motivations. So, uh and uh and then I would call this a control way of going into the research. And uh yes, uh for example, II, I've, I've done a lot of meta analysis. I think uh uh the previous speaker mentioned about the balanced CV. That's a very valid point. You need to have a balanced CV. But uh you will have balanced CV if you go through that pathway anyway. So II, II started the metanol, I published a lot of co studies I have created uh uh the formula that the for, for, for lots of academic people haven't created. So these are something that I've achieved by learning evidence, synthesis myself going through a meta analysis. OK? And then, and then I learned through uh the process and then I got myself uh uh uh experience in other aspects of the uh uh you know, research as well. So you can easily achieve these things by doing a uh uh a meta analysis. And uh and, and if I want to choose a study design that changes my practice on daily, I would choose a meta analysis. OK. It's easier to do. Uh But uh for example, I'm going to bring an example if someone does a uh appendicectomy, which takes uh 40 minutes, uh it's much easier compared to doing a uh a histological assessment of the appendix which may take two or three hours. But does it make my job uh less valid? The answer is no. So it metal is exactly like that. So, uh it's, it's not his fault that it is easier. It is better. So it is better a randomized control trial. Ok. So, and uh learning that is something important and then it helps you to, to, to, to determine uh uh uh uh uh you know what you need to know about the research. And also you can become a experienced researchers after publishing several of those meta analysis. So uh going through the process is easy. So I think uh uh it was uh uh uh brought to attention and it is important to have a good supervisor. So there will be supervised in all hospitals, we will find some and then they will go through to discuss with you individually and then there might be some small group teachings and then you can be allocated to have some your own research project. OK? And then you develop that yourself. I really hope that uh you know, the information I told you uh in your mind, at least it is not going to be necessarily applicable to your uh real life. Uh But uh in your, in your uh in your mind, make this very vague word of research uh more uh uh clear for you without research, without an excellent portfolio, you should be able to go through everything. OK? So because there are other important factors that uh can determine the outcome of your interviews. OK. So uh ask uh you go and if you don't want to ask if you go to a hospital, just say uh uh 10 of your uh uh some senior people in that hospital see how many publications they have. So especially if they are not working teaching hospital. So uh you will be surprised. So, but they have achieved what they wanted to achieve. So this is not the main determining factor, but uh I'm going to encourage you to uh to learn about the research uh because of different reasons is because you, you, you, you, you need to uh practice based on the best and level evidence and learning around research helps you to be a better uh practitioner. OK. So I don't want you to be disappointed and I hope my, my, my talk or others have made this uh more clear for you so that uh by developing skills in research in next few years, you can have favorable outcomes in any specialty that you want to do if it's surgery, I'm sure uh uh that is a very a as well. Thank you very much. Thank you very much for that talk. That was a, that was a really good overview of what you've done. Um up until this point on a really nice story at the start as well to um really er bring home er, all of your points. As well. So, thank you very much for that. Um And yeah, it's definitely important that research isn't the be all and end all of getting into like specialties, et cetera. It's a nice adjunct, but it's definitely important for um um evidence based medicine and practicing and treating your patients, et cetera as well, which is a really good point to, to make for everyone here as well. So, um what we'll do now we'll have questions in the chat if that's alright and then we can all um, use the floor and just um um, answer these questions if anyone's got any questions, put it in the chat. Um, if not, then um we could wrap up. But I, yeah, um, feel free to put it in the chat if you've got any questions. Yeah, b me, I thought fanta fantastic meeting. Really good talks. I really, um, enjoyed the evening. It was really good. Um, I just had a few comments, er, which I sort of heard through the things which, which may people may find useful. I think when you try to contact a, a, an, a consultant, often of especially the older guys, they often don't have email and the secretary will deal with it. Um, so don't rely on the email and if you don't get an answer, don't, don't get annoyed or you know, there are different media uh to use the best thing is always to see somebody in person and surgeons is always easy. They're always in a good mood when they're in theater. So phone up the secretary find out when they're in theater and see if you can go and see them there and they'll be delighted and they will best to get somebody in a good mood if you want something from them. And if you got email it so easy to just ignore it and a lot of my colleagues uh do that. Um you know, um I try not to but, you know, the amount of emails you get is just so phenomenal. You just can't answer them all, you know. So that was the um comment I had on that and I totally agree that um meta analysis is a really good way um of, of, of, of starting a, a research career. Um But um you may, you may not know, but I'm also a statistical adviser to the hand journal, uh the European Journal of hand surgery. And especially after COVID absolutely overwhelmed with me analysis because not the first person to have the idea that if you put the uh talks together, you can do this all from your computer at home, then you can uh do this. So just the, the, the guidelines on the statistics are very, very tough and they, they do actually kick a lot of them out now. And I think to do that on your own um would almost be doomed to fail you. You have to be able to have some support from particularly good statistician. Um you know, to um to make sure that you've got the appropriate back up because people will look through that, you know, I will know that you, that, that the statistics are not right, there has to be right up to right up to scratch the statistics because they are really tough on that. Um I thought um on, on just giving you suggestions on from the research projects I had in the, in the time audits are very good, very good to publish. I think observational trials when you do something very good and you can get involved with, um I wouldn't um uh I've had some really good medical students who write a case reports with me and you may say so, just a case report but there are journals who do publish them and OK, it's not top notch but it is a nice way to start, you know, and to write a paper and to read on it and it is better than nothing when you, when you apply for a job, you know, so I wouldn't discourage it. And then uh the preclinical papers are, are really good. I think for when you're starting, for example, one of my, er, registrars did a thing on cement, er, you know, bone cement when it sets and it takes 10 minutes and when they put the cement in um how viscous it is at different times. So let the cement set at different times, you can do it on the lab on your own. It's really easy and the implant companies will give you the cement for free. Um I've had registrars doing things on anatomy and looking at new blood vessels in something. Um You know, that the, the consultants will be really keen on that you could do your dissection. It's a beautiful anatomy lap and queer you could use to do that. Um, so that these all are like suggestions. So hanging on to a project, other people do. So just suggestions, you may say that easy way to start research rather than, er, because I agree a randomized control trial is absolute. I mean, just, you won't get it started. I don't think, you know, you may participate in a bit of it but um you wouldn't be able to complete at all and then you only get part of it. I don't think it's as satisfying then. Yeah, I agree. I think that's a really good, a good viewpoint to have as well, especially mentioning the anatomy labs at Kiel. Um I know a lot of people that are inter anatomy, get the opportunity to do certain research projects related to dissections or even go through prosection, et cetera as well. And they've actually been able to publish some of their works as well from it and they've got an area of interest that they can demonstrate. Um, they look into a particular organ system of interest as well for, for instance, some of them. Um and I think that's a really good way to also get involved um and get dedicated time cos especially in those settings, it's hard to get the time to actually, er, and the longitudinal nature of having to do dissections over like 2030 bodies, et cetera. And so in some projects, it's hard to get that time and inter inter degree is a, is a good way of doing that. But also having analysis and analyzing data that's already been collected is also a very convenient way to familiarize yourself with how to use um statistical software as well, which is a big skin in itself. A lot of people are getting involved in like R and Python, et cetera and more um more open access software and also engaging in open research practices as well is really important. So it's good that there's a lot of different ways that you can actually get involved. And we've got kind of like the whole span here. People have been into chal, people who have done meta analysis this moment, people have done lab based research and also dry lab research as well. So um hopefully this provided like a good um basis for everyone to start um kind of pursuing what they want to pursue in terms of research. But also thinking about just academia in general as a viable option for them to um, apply to in the future. I don't think there's any more questions in the chat. What I'm gonna do, uh, briefly is share my screen for the, um, for the feedback form. We'll give it a, like a few more seconds in, in a case anyone else, um, of our, of our speakers have anything more to say, uh, about what we've discussed form essentially. Um, so, yeah, I mean, I would agree with what you were saying about a bit about the um about the er dry lap thing. Uh Also obviously one of my interest. But first of all, there are lots of data sets available and all if you have the skills of analyzing R or Python and it, this is a really good skill to learn. And II would say when you are young, like you guys are, then this is the time to learn it. Um And OK, uh Python may be overtaken by Julia or Rust or another program, but it doesn't matter because the, the, the, the thinking behind it on how you do the coding doesn't change. And, and, and you know, people who use data analysis never use graphical user interfaces or S BSS because I mean, one it costs shit, shit loads of money and, and, and, and, and you don't know what you're doing and, and R and Python are much better and there is a bit of a learning curve. But if you have those skills, you you just can cash back on them all the time because you can do all the data analysis for people and they put you on their paper, you know, so it's a really good way of um getting involved spending the time when you're young, playing with computers, you guys have got the brains to absorb this very quickly. So I would definitely recommend recommend that. I think, I think briefly on that mentioning S PSS, it does cost a lot of money but will actually have an institutional access. So if you do find that you are involved in a project, and let's just say so for me personally, I use S BSS because my first ever project, one of my supervisors like is basically the best PS as you could say. Um But so it was a good way for me to learn that kind of stuff. But since I've been able to um expand the repertoire skills by using things like um STA and um R et cetera. And obviously Miss Vander showed me through um revman and such as well. So I think having the knowing the statistical principles, but also knowing just how to do basic good methodology and making sure you're grounded in that to have a good methods um is really important as well. Um But yeah, I mean, do comment on that. Uh I mean, the the first thing is obviously with the graphical user interface, if you have reproducible uh research if you want to reproduce what you've done, it's very difficult. A nice thing with R and Python. If you have a script, you run the script and you know exactly what you've done. So if your data changes a bit, it's not a problem with a graphical user interface where you have to know where you've clicked and what variables and parameters you set is more difficult. The other thing with regards to cost, you're right, obviously, Kiel um have, have got an institutional payment but I mean, if you go in Cambridge is not a problem, but if you go to a for example, they don't have institutional thing. So what are you gonna do if you only know S BSS? Are you going to say that you are a thief by using S BSS and publishing with it? Because that is quite dangerous, isn't it? Because it will be published? And if you say that you've used S BSS in the paper and actually don't have a license of it, then you say in, in public that you're a thief. Um which is um not a good thing. I think. So you have to be a bit careful with that. You don't have to definitely make sure that the institute you're, you're working with has got a license. Yep. Fully agree with that. I don't wanna, I don't wanna get caught up or anything like that for sure. Um I don't believe we've got any more questions So, if we're all happy, I think we'll draw that to a close. Thank you everyone for, um, attending. But, and also thank you very much to our speakers, prof Miss Tavana, m er, Doctor Nantha Kumar and Doctor Hall for, um, taking your time out of your valuable evenings to actually, um, deliver these talks and I'm sure this will be really useful for students and, um, we will put the feedback on there as well. So please get everyone um fill out the feedback form cos obviously, it's good for all of us to know exactly how these talks are going and if it was actually useful for you guys, and maybe it's uh it can be used as an argument to create more of these kind of resources. Um and maybe go into more depth in some of these areas as well. Um So you've got, so you've got more information to actually draw off to get involved in different projects and such. But thank you very much for attending and thank you to our speakers as well and feel free to leave. Thank you very much for. Thanks all of us. Thank you all the best. See you, bye.