 So thank you very much to everyone for joining us today as our first ROP webinar on telemedicine for ROP screening, being held jointly with the Pakistani Ophthalmology Society. And my co-chairs today are Professor Claire Gilbert and Professor Moin, and we're really delighted to have a really exceptional final of global experts for this webinar, who I will introduce and turn soon. First, I just wanted to give some background to ROP Net, though, and what led us to have this webinar today. So ROP Net is a network of professionals which was established in 2017. It was funded at that time by the Diamond Jubilee Trust, and it had the goal of really dealing with the challenge of the looming third epidemic of ROP, which we can see is beginning to occur across Africa and Asia. With the increasing number of neonatal intensive care units and increasing survival of premature babies. So ROP Net formed with this goal of preventing vision loss and premature babies by improving both the coverage, the prevention and treatment, and it's been based around a number of principles. These have included South-South partnerships, so we matched institutions in Africa and Asia with other mentor or training institutions that were also initially primarily based around India and South Africa. And the concept of this was that institutions that were slightly further ahead in the ROP screening and treatment program like their journey and who faced similar challenges would be able to mentor and support those that were just starting their ROP programs or trying to scale up. Another important aspect of the principle of the program was this multidisciplinary aspect which was based around both the neonatology and ophthalmology team working very closely together. And in the first phase of ROP Net, there were six institutions, so Kenya, Tanzania, Nigeria, Ghana, Pakistan and Sri Lanka, and they were supported by training institutions in four different places in India and also in South Africa. That phase came to a close a number of years ago and now currently we are restarting ROP Net phase two where we are planning to expand to new members and we are keen to keep the successful sort of South-South partnerships model and the networks as well as the strong multidisciplinary aspect. But we're also planning now to focus more on telemedicine, on strengthening data collection and monitoring, especially integrating with the neonatal data systems, collaborating with Ministry of Health and others for advocacy of ROP international policies. And then finally also this aspect which is the sort of ongoing education and webinars which we've just started. And what we hope is this can be a forum for ongoing both education of our members but also discussion with each other of the challenges that we're facing in a way of supporting each other and coming up with new ideas. I'm just at this point going to draw your attention to ROP Net website that we've just set up. We'll put the link in the chat in just a minute. So I would encourage you all to go and sign up at that website that will make sure that you're part of our ROP network and you'll hear about further about our webinars and other events and any kind of future opportunities. So without further ado at the moment, I'm going to go straight to introduce our speakers and the way the session is going to run today is that we're going to have approximately 10 minutes each of each speaker that you will have seen. We'll follow that by around five minutes or so questions for each speaker. You can throughout as a speaker is talking, put your questions into the Q&A box afterwards we can put those questions directly to the speakers. If you want to speak directly you put your hand up use the kind of hand function. And then once we've had all the speakers we should have around 30 minutes or so for any general questions to the entire panel. And for any other kind of discussions or questions that you may have. So start, we're going to start today with Professor Paul Chan. He's the head of the Department of Ophthalmology at the University of Illinois. He's got extensive experience in telemedicine screening and AI in a number of low resource settings and that's what he'll be discussing today. We're going to follow that by Dr. Michelle Kichago. He's a consultant vitro retinal surgeon in Nairobi, Kenya. They were one of the countries in our initial RAP net program network and they'll be talking about their experience in Kenya. That will be followed by Professor Linda Visser. He was also involved in that RAP net and she they were actually a mentor and training institution for the Kenya team. She'll be talking about their experiences telemedicine in South Africa. Then we'll have Professor Anand Vineker about talking about his very well known and successful kid rock program, which they established in India and their experience there. And then last but not least we have Professor Moine from King Edward Medical College in University of Lahore, who are also, of course, as I said, running this webinar is also part of their ophthalmology conference that's occurring right now and he'll talk about they were also part of our RAP net and they'll be talking about their experiences and challenges in Pakistan. So as I said, we're going to start with Professor Chan and with his talk for around 10 minutes and then we'll take around five minutes or so of questions straight after that. So please do put any questions or comments as you go along. Thank you. Hello. Thank you so much to the organizers for putting together this symposium. And also, thank you for having me speak to you today on telemedicine screening studies and low resource settings. I'm Paul Chan from Illinois, I in infirmary at the University of Illinois, Chicago. Here my financial disclosures. I am a co founder of some vision, which is a company that is looking to provide resources for screening in resource limit settings. And first and foremost, I want to acknowledge my collaborators from the IROP and general groups, especially my international collaborators and also Michael Chang, Pete Campbell, Richard Kapathikramer, Karen Jones and many others who have helped with the work that I'm going to present to you today. So when we think about telemedicine, I think it's very important to understand that this is not a new idea. What's been exciting is that we can practice telemedicine more efficiently now with the new technology that we're seeing with the resources. And also, as we see economic development, we're seeing that our ability to do telemedicine through imaging and so forth has been facilitated. And so you look at this, this is a slide from Michael Chang, the practice by telephone back in the 1800s. And you can see the doctor talking to the grandmother here and trying to diagnose the child with what's happening just through the simple use of a telephone. And now we have imaging. Now we have surgical mentoring. We have a lot of different tools at our disposal. And we think about ROP. I think that this conversation has evolved tremendously over the past 20 years or so. And historically, when you think back, there was always this question, well, is telemedicine good enough? Is it good enough to make the diagnosis? And studies in the U.S., studies outside the U.S. internationally have shown that it works. It's been shown to be reliable in the diagnosis of ROP, accurate, cost-effective. And there are many live telemedicine programs around the world. And I think India is a really great example and you'll hear from Anand Vinikar later in this webinar about KidRop. And he's done a tremendous job of providing access to care in those communities. We've worked closely with the group at Aravind with Parak Shah, Narendra Venkatesanathapathy, on their ROP SOS program, myself and Pete Campbell. Again, doing an amazing job in providing care to thousands of children using telemedicine methods. And one thing that comes to mind now, and the question is that a lot of people ask, well, how do we do this? And can we deploy telemedicine screening in low resource settings in low mental income countries if there is not enough infrastructure, especially for the internet, to do so? And what I found interesting, at least through the work that I've done through the years, is that internet capacity is being built very early as we see economic development in a lot of these countries. You know, similar to diabetic retinopathy, as we see countries have more infrastructure and more wealth, you're seeing more diabetic retinopathy. I think you're also seeing more retinopathy for maturity. But there are limitations and there are concerns around this, and it's not perfect to model, but it is happening. So as I mentioned, we've seen internet access grow exponentially over the past 20 years. Obviously, in middle income countries, you're seeing this happen more quickly and in high income countries, obviously. In low income countries, there are some limitations and there's less literature in low income countries. But what we're seeing mostly is that as low income countries start to emerge into middle income countries that the development of telemedicine becomes much more feasible. Cost also has been shown to be an issue, not just in the startup costs of telemedicine programs, but overall we found that these have been more cost effective. And time and time again, people have shown that screening for ROP is more cost effective than doing nothing. And we don't want to have blind children if we don't have the screening abilities in place. There's also a lot of opportunities in this space going forward. The ability to create better data management systems, the ability to potentially implement artificial intelligence in these communities and in these programs. And also for education in general. Okay, so can you provide tele mentoring tele education and so forth. To maintain follow up care, but also to follow up with students who are learning in these programs. Okay, so telemedicine is happening. And we're seeing this happen in low middle income countries, but mostly in the middle countries. So, our meeting is a great example of what started to happen. And the sub program that I've worked with Tom Leon, many, many years ago. But you can see here, you know, you start to see NICU development, NICUs start to keep kids alive. You're starting to see more premature babies, which subsequently means that you're going to see children who are at risk for random optic prematurity. And if you don't have screening programs in place, these children will get discharged and they'll go blind if they are at risk for developing treating heart disease. Okay, so you need to find solutions and one solution is telemedicine. If you don't have enough of a workforce, especially to screen all of these kids at bedside and then also for the follow up care as they come to your outpatient clinics. But using image based diagnosis and care, I think it's very helpful, especially for identifying children at high risk. Again, as I mentioned before, the folks at Aravind in Coimbatore have done a really terrific job at providing care for these children in the Rope SOS. This is a USAID Sponsored Program. You can see here they have a van and they have a red cam shuttle and other imaging devices that they've used over the years. And you can see the large area that they cover out here in India to go out to the communities, screen them, provide diagnosis by physicians in Coimbatore, and then provide appropriate treatment accordingly. Again, this hub and spoke model, I think is very, very effective in many parts of the world. And as I mentioned also, Dr. Vina Carville discussed some of the work that he's done with Kid Rob. Other programs that we've been involved in is a program with Orbis International. So we've we've published on this screening program that we implemented in Mongolia. Again, prior to the work, there were no screening programs in place at that time. But then we started to see children who had developed very severe Rope required treatment. And with Senke, Chilambat, and with Chimja, the folks there have developed really strong screening programs using image-based diagnosis, deploying telemedicine methods, and also telemetering. This is now continuing to the next phase with Orbis International looking at AI and so forth and how feasible that is in these communities. But again, using image-based, not just for screening children, but also for education and training. Okay, so one of the other things that came out of this program is also the need for really good data management systems to standardize data so that we can understand better who are at risk for developing R.P. And treat more current R.P. and to also improve the care overall for the children with Orbis Materially. We've also worked with programs in Nepal at Tekanga Eye Institute in collaboration with Forrest and also Hella Kelly International. Here, different from using the REDCAM, they used a more cost-effective camera using the Forrest. And you can see here, Dr. Samyam Bajamaya from Nepal training and using that Forrest system and the folks here from Helen Keller. And what was great about this program is that they were using data management systems, a system that we developed called ITELGEN to try to not just collect data on the eye findings, but also on the systemic findings of these children. A lot of advocacy was done for these children and promoting better screening programs. And you can see here on the right, the data collection system from the Forrest. And what's neat about this is that over the past decade, they've really done a great job in improving their cloud-based management system, telemarketing reading center. And this was the system that we were using initially. And the left is our reading center ITELGEN, which has been used by the Arabian Eye System for all of the screenings over the past three to five years or so, looking at thousands of children. So I think that this is a very important piece of the discussion. When we think about establishing these telemedicine systems, especially in low-resourced settings, is that not only do we want to make sure that we're doing a good job with the screening and the imaging, but also data collections so that we can learn better who are at risk and to have discussions in these critical conversations with me and about how we can improve the care of these children overall. Obviously, internet speeds matter and can affect the care and also data collection in general. A significant coordination between everyone involved has to be done to make sure that there's good coordination and follow-up. Another topic that comes to mind when we think about telemedicine is the training. So we can implement systems, we can get cameras, we can have the workforce and get people to actually operationalize the telemedicine system. But we can't forget about how important it is to make sure that the people doing the diagnosis are trained well and can understand the critical features of the disease, but also make sure that follow-up and treatment is performed. We've subscribed to the establishment of teleeducation systems that we've built, not just internally, but also in partnership with the American Account of Ophthalmology and others. We've shown that, and as I mentioned, we've established teaching programs through the American Account of Ophthalmology to provide an open access way that people can look at or train on the basic diagnosis. There are a lot of diagnostic changes in ROP. You can also think about using imaging to train people how to treat. We've published papers doing this, looking at can you use a red cam to identify skip areas and so forth. And we found that it can be very useful as compared to just using indirect ophthalmoscopy. Also training people on doing the imaging. This is very, very important. So when we established these telemedicine systems, making sure you have good imagers to go in country and to teach a workforce and how to do appropriate imaging in general. And very importantly, we all know this indirect ophthalmoscopy still matters because we have to use indirect ophthalmoscopy to apply appropriate laser treatment for these patients and also when to do anti-veg F injections and how to follow these patients appropriately. So in summary, telemedicine screening is happening in low resource settings and I think more so in middle income countries. But very importantly, you see here at the top, there's more to ROP telemedicine screening than just reading an image. And I think that's something that we all have to think about. There are a number of challenges in image-based diagnosis. There's some variability in training for ROP care. So focusing on education I think is critical. Telementoring can be done. Data management incredibly important as we look to the future and collaborate together. So making sure that we're using good systems and all agreeing on what should be established, not just for the eye findings, but also for the neonatal findings and how people are doing systemically. And I think importantly, one of the questions that comes up too is who should be responsible for ROP screening. And I still believe that it should be the skilled ophthalmologist who determines the management decisions for ROP, especially in these cases. So I think the future is exciting as we start starting to see better technology or cost-effective technology and really a team approach coming together to care for these children on a global scale. So again, thank you for the opportunity to speak to you today and I look forward to the discussion. Great. Thank you very much. And thank you Paul. And I think Paul has also joined us now. So thank you for being here. Hey Aisha. Thanks a lot. Yeah. And I thank you for that. It was really, I mean, it's an amazing, great presentation, but also like the amazing kind of the experience that you've got. I was, I'm just going to wait until there's any other questions that come up. But one question I had was just about because you've got the experience of seeing things across India, Mongolia and Nepal. Like just thinking about what are the kind of main differences and the challenges that you faced in these sort of different settings in terms of implementation and the telemedicine screening in those kind of different places. I guess with the view of thinking about what are the challenges for people in general when they're first starting that they're more kind of to be aware of and plan for. Is there something that looks like. Yeah, you know, it's interesting. And Aisha, I think you and so many and everyone here in the group have heard this a lot, which is how can they make it work so well in India. Right. And we face all these interesting challenges elsewhere to implement these large scale screening programs. I think even in the US, you know, we face a lot of challenges around, how do we just get this off the ground? How do we get by and I'd love to get a non-stake on this too. I mean, we've spoken to a Prague about this in Aravind, but just getting the folks to do the imaging to set up the programs. You know, in this hub and spoke model, I think is very challenging, right? And you kind of have to get buy in. So the, so there are common themes, right, which is you need to get buy in from everyone involved in the care of the child. Right. So you need name etology 100% on board. You need to have the infrastructure from the hospital to make sure that you have the personnel who are going to do the imaging and help you with the workflows. You need the technology. Right. And I think that that's another sort of bottleneck that we face is that there are so many systems out there, but not everything is perfect. Right. And it's very cost prohibitive. And then you need the platforms. Right. And that's why we kind of focused on the data management telemedicine screening part. But I think it's one thing you learn, and you know this just as well as I do, is that if you've implemented in one place, you've been implemented in one place. And if you've seen one system, you've seen one system. So it's not as easy as it seems. Right. And we know this, but we have to do it. One of the things I will say that I think we're going to have to also figure out as a community are the, and this goes to the question that's in the Q&A. One of the earlier questions around what would you do? Would you do screening first in person? Would you start with telemedicine? Even in the United States, we face the similar challenges and we have, you know, at least in our, in our world, you know, the American account of ophthalmology and OMIC, which is an insurer, you know, comes up with these guidelines around what people should be doing. So it would be really interesting discussion in the global community, sort of recommending best practices as we start to implement programs. Right. And that's something that people can lean on. Yeah. So Yisha, I don't know if they answer your question, but you know, every program that I've worked with is different and faces similar, but also different challenges. Yeah, absolutely. And I think we're going to hear from Michelle next and then Linda, and I think they will part of what they're going to talk about some of the challenges of trying to do, like doing in their setting and what they've come across. Yeah. So I think then maybe like at the end when we can discuss together, we can have a think more about that kind of deeper. I've got like a kind of another question, which I don't know if you're like just from the audience about what other cadres could be involved and I think you touched on that a bit. But yeah, just what other kind of people could be involved in RFE screening and I know that again you've seen that in different settings, who else could be trained for. Yeah. And I think that this is again, you know, in the US, you know, we are very clear about the ophthalmologist doing the screening and being responsible for the training for the screening programs. Again, that's going to be different everywhere we go around the world. I think it really depends on who we are comfortable with or who is approved for certain types of care for our patients. Right. And, you know, I'm not going to pretend like I know what's happening in every part of the world, but, you know, for example, in Sub-Saharan Africa, you know, non ophthalmologists can care for patients with cataract surgery and so forth. Right. We know this. So, again, this is going to be the discussion around who is competent to do so and who within your environment and within your country. It is accredited appropriately to do so. I think that's another issue is that accreditation will matter and we have to make sure that the ophthalmologists are in that discussion on who should be doing this. Right. Because you are ultimately going to be the one responsible for the treatment and the long term management of, you know, of these children. I'm actually curious, Aisha, what, you know, there are so many people in this room that are really great and have thought about this. I mean, that's a really great question. Right. This is a scope of practice issue. What do you think? Who should be allowed to? I was just going to say, I think this is actually a question for all because Anand is going to talk about, of course, like they've trained other catars in India. But then I think there's been, I think, again, Linda, they've had issues with like legal cases and things like that in South Africa. So all of that, I think you have to balance up. But as you said, ultimately, the responsibility then, who bears the responsibility for a baby, they were to go live from ROP. And I guess that, you know, so that that person who bears that responsibility have to kind of be overseeing whoever else is conducting the screening and all of that as well. I think that's an important, important question, which kind of, as you said, run differently in different countries. Yeah, exactly. And, you know, Linda, I mean, that's, you know, we always consider the medical legal issues around this because, you know, that's ultimately the onus is going to be an ophthalmologist, I think, and also neonatology. Muhammad put in something about AI, that do we have a few hours to talk about that one? I saw that question and I was actually going to subtly put that for the panel to cut to the end. But I think that question about AI with telemedicine. At some point, we maybe we can put that question at the end of the for the, because there's going to be like a kind of around discussion, which we'll do after those speakers. If that's okay, Professor Boyd. And to Dr. Kachago's presentation and he's here and he's also going to be around for some specific questions and I do like I think these will be quite interesting now what we're coming up with next. We're going to have the example from Kenya case studies from Kenya and then South Africa. So maybe if we, if that's okay, I'll move on to that. Thank you, Paul. Thanks for the presentation and I hope you can stay and join us for the end for the panel. Anything else you want to add Moin at a stage. Sorry, thank you very much. I think Paul, very wonderful talk. I think it's commendable that he's doing so much for so much and I don't know whether he gets that energy to do that. So we're trying to start that kind of stuff but I think this is the future and it's he has realized that very, very early and he's getting everybody on to it I think I think this is the way to go now. Yeah, absolutely. Thank you. All right, so we'll go on to the next speaker which is Dr Kataga from Kenya and we'll just run presentation first and then again everyone can just put any questions for him into the chat book. We're actually also going to be running a few other questions as we go along so please just notice if there's any things about whether you're practicing RP screening and whether you're doing treatment. It's just very good to hear a bit more from the people who are joining us as well so just be aware of any questions pop up in your chat to answer those as you go along as well. Thank you. Hi. My name is Dr Mosha Gashago, I'm an anthropologist and retina specialist for Nairobi Kenya. And I'll be talking a little bit about telemedicine in Kenya in the ROP setting. These are some of my affiliations. The ROP journey in Kenya started in earnest at around 2015. This is when we really got down to organize the ROP screening. The pilot program was set up at the Kenyatta National Hospital, which is a tertiary referral hospital in Nairobi Kenya. The pilot at Kenyatta National Hospital in partnership with the University of Nairobi was selected as a best side because it had a large NBU or NICU with a capacity of about 100 patients occupancy at the hospital was at about 150 to 200. With time improvement, NATO care had become available and therefore smaller babies were surviving. Oxygen monitoring however was very poor and therefore we suspected that there would be very high prevalence rates of ROP. The expertise was also present. We had neonatologists which are retinal surgeons pediatric ophthalmologists both from the Kenyatta National Hospital and the University of Nairobi. ROP also fit the criteria for screening. ROP as a screenable disease qualifies much because it is site threatening. Serious and irreversible consequences occur not treated early. Treatment at earlier stages is more effective than after the development of symptoms. High prevalence of the detectable preclinical phase. There are also suitable screening tests available which have high sensitivity, specificity, adequate low cost, easy, safe, have minimal discomfort and generally acceptable. There's also appropriate and available follow-up. At the time of starting our research program, there wasn't much in terms of research that had gone into ROP. There was the one Jala study of 2003 and took up to 2007 where he had assessed 120 babies, 240 eyes in a public urban hospital. ROP prevalence at the time was established to be about 16.7% with 0.8% of babies having the threshold of eyes having threshold ROP. There was a study by Staten in Western Kenya, which was a rural setting and there wasn't any severe. There was no ROP that was detected in the study. A Nyango study of 2015 is what shed quite a bit of light on ROP status in private hospitals in Kenya. There were 102 babies who were studied in a private urban hospital and ROP prevalence was noted to be as high as 42%. They also noted that babies were much smaller in surviving at 1.1 kilos with a minimum gestation of about 29 weeks. The beginnings of the program were difficult and there were a lot of nail sales. We had very poor equipment, the laser had dodgy optics and a broken fiber optic cable. A 28D lens was just not available and I had to buy my own. Pediatric speculums also not available and my colleague and I had to buy some and donate them to the program. We got a donation of a fasting to the university that was able to be used for the ROP babies. The ROP drops are also donated, however, the supply was erratic. We developed some standardized documents that we reduced to capture the information. This was in the form of forms and registers. The first fully of ROP screening was 2016 and that year we managed to screen 183 babies and diagnose the 18 cases of ROP. This was at about 10% of the pregnancy. We then decided to escalate the screening to a national level of this need and some organization and therefore we formed what is called a national premature maturity working group. We looked in some of the somnolatologists to join into the working group and got a stake point of involved with the last meetings. This was the ROP working group that was established in June 2017 and in it we had three pediatric ophthalmologists, two vitroignal surgeons and four neonatologists. And you note that the neonatologists were more than the ophthalmologists because they are the primary care givers to these little babies. So with that program, with the working group set up, we were able to extend training and ROP services to more hospitals. And the public hospital during the screening program are now up to six, like in a hospital, national hospital, normal Lucy hospital, Mwani, Bagadu, Mugwana and Mwani teaching and referral hospitals. We have also increased the screening programs available in private hospitals, and now these are up to nine private hospitals. We have also identified the next few sites that we shall go out to train. We have the Hibakambo level five hospital, F وال hospital and the general mobile wing and referral teaching hospital, and these are very large hospitals with very high volumes of patients including neonates. And we have actually seen cases of ROP coming from some of these setups, and sometimes unfortunately we are out on need. So what is the role of telemedicine in our setup? Because we realize that telemedicine is an ability to deliver health care services remotely. And I think this would apply to ROP because we have the opportunity to gather images and sharing. So telemedicine would be the form of photodocumentation of cases to the photography, peer-to-peer consultation aided by these images, patient-to-doctor consultation, remote imaging and lab works, remote monitoring. In our setup, with the expanding network of institutions now offering ROP, we feel there could be more of a need of telemedicine now than there ever has been in the past. Red cam in our setting, we have a red cam, but unfortunately it has not been used much in the screening of ROP, but has been used lightly in the documentation of retinoblastoma. So we haven't done much with the red cam in ROP screening. What we have made use of more is a 2.2 or 20-diliter lens, together as a final photograph. Some people have been written about it as you can see, showing that it is an effective way, effective and actually very affordable way of capturing images using a phone. And these images are therefore then very easy to share and learn from. Sometimes the apps on the phone don't work well and you have found that there is an app called the VF Light that works quite well because it is able to have the flashlight running on your phone as you search for them, with images that you want to capture. These are some of the images that we have been able to capture with a 20-dilens and a smartphone. For those are courtesy of Dr. Oscar Nyango, who is a member of the ROP working group and we call him the father of ROP in Kenya. The stage four disease, as you can see in this case, with retinal detachment, very well documented with a 20-dilens. Here you have other stages of ROP, earlier stages where you can see the abnormal vascular and abnormal demarcation line there, couple of younger spots here, small hemorrhaging right there. In this case, you see very aggressive plus disease all the way to the posterior pole. In this case, you see abnormal vascular coming in the periphery of the retina. Here you see very nice images of laser marks that have been delivered to the child and you can also see here a nice features hemorrhage right there. And here you also see abnormal vascular which is the periphery of the retina with an abnormal demarcation line right there. So in summary, with scaling of screening, there will be more organized telemedicine to assist with knowledge sharing, documentation, better demonstration to parents, further consulting on the road cases, monitoring of disease progress and development also of an ROP database that could be assisted with a permission of high quality images that can then be stored. Thank you very much. And I look forward to listening to the rest of your presentations and learning how better we in Kenya can adopt telemedicine in the fight against ROP. Thank you. Thank you very much, Moussai. I think, Moussai, you used to have just gone off. It was actually on here on the panel. So it was a really interesting presentation from Moussai. I don't know if there's any questions or comments for him at the moment. Is there anything, Professor Mouin, from your side and actually the people in back then who are watching as well? Oh, hold on, Moussai's coming back on. Hello. Hi, Moussai, we're just looking for you. Hi, sorry, it just dropped off the call immediately in the presentation. I noticed that. I thought I'd seen him throughout. And then just as soon as his presentation was actually dropped off. So it was a great presentation. Thank you for sharing it with us. I mean, I was thinking, I don't know. I'm just going to have a look and see if there's any other questions. But I had just a couple. Oh, actually, Professor Mouin's asked one. Let me just read out the comments for us, actually, before I go to mine. So Marcia has written, congratulations, Shai, Sarah, Oscar, and the team in Kenya. Wow. You have really moved forward with your network into Kenya and in such a short time. So interesting to see your learning with the smartphone too. Well done. And keep up their amazing work. So that's just a very nice comment from Marcia. And then Professor Mouin is, is phone-based screening as good as a brand? And is it possible to decide where it is if you stop breathing? I think you probably know the answer to those questions. I think at the moment, it's, yeah. It's for the same reason we're all looking at it. We know that, like, at the moment, I think that's not the case, using images from an iPhone. So I think Professor Mouin just wanted to, I guess, more about the smartphone-based, the phone-based screening that you're doing and how you find that working and how you find those images compare and do they help you? Are they really able to help you with your decision-making? Yeah, thank you for that question. The phone-based screening works in terms of, especially when you need to get posterior, you know, posterior photos, it works quite well because it's not as wide a field, of course, as the RETCA. So when you need to get more peripheral photos, then it requires a little maneuvering in terms of turning the baby's eye with a tiny kind of indenter or using a vectis loop. We found that it's quite small and gentle for the baby's eye with a speculum in place. So you have to do that to kind of see more peripheral disease. So in terms of screening, personally, I'll take the photos for photo documentation and in case I feel like I need to share them. But I will always do an indirect fundoscopy along with that so that I have a better view, I have a wider view, and can, therefore, make a better decision. So we'll do the fundus photo kind of to kind of indirect fundoscopy and to enable us to share and also do presentations in the course of teaching or for photo consultation. And how did you learn to do it, actually? How have you taught yourself? Is it teach yourselves? Are you teaching each other? How's that, Karen? I've given it a go at the odd time and it's not easy to think so. Exactly. It is not something you can be taught. It's like indirect fundoscopy. You keep doing it until you get it. That's a trick. So it helps to watch somebody do it because you find the image they share with you is often not as great because once you get the image, you get part of your fingers, you get some of the baby's face. So once you've done your photo, you then have to kind of crop it and what you share with people is a nice clean image that they think you got. So you kind of just have to play around with it. If you had the opportunity of using the monocular indirect, then it's a bit easier because you had that experience of 20D and monocular indirect of thalmoscope. But if you're just used to the indirect, then it takes a little more training knowing where to position the lens and where to position the phone for your best quality image. But basically you replace your phone where your face normally is when you're doing an indirect fundoscopy and then just keep playing with it. An easy way to do it is with a video function because as you do a video function, then you can see what's happening. So try to play around with a video function because the flash is on until you're able to get good images, then pause and crop. You know, like I get a screenshot, that's one way of doing it. But with time, like with a VF light app, it activates a camera even when you're doing a photo. So you have a nice viewfinder, so to speak, until you get the photo that you want and then snap it. So those are, you know, tricks that you can use in the course of learning how to use the indirect. These days, there are some commercially available kind of lens holders, like the, we call it a red cam. What do they call it? Some like a mini red cam or something. It's basically like a bar which holds your phone on one end and the lens on the other end with a preset distance and therefore you get good photos all the time. So that's also a gadget you can use that is quite cost effective, so to speak. And how often do you actually practically use it? Do you have people like, for example, do you have like junior ophthalmologist or whatever in another hospital doing screening? Will they actually send you these for advice? And do you feel like you can give clinical advice based on those images? Well, the challenge right now is what you're talking about. It is more difficult to get a photo with the app than it is to see the retina with your indirect. So most of the consults I will get is people who have actually just looked and then call me and describe what they have found. But on occasion I have found that their description and what I actually see when I finally get to it are quite different. And therefore as we roll out the kind of the screening that's one of the skills we want to incorporate into the training to enable better communication because getting red cams at the peripheral hospitals as of now still looks like a pipe dream because of the restrictive cost. But you know there are cheaper options from India that we could also look into. But for now the most affordable option is try to get images with a smartphone and a 2.2 or 20D. Yeah. And I think, I mean, how far do you think you are from or what do you think else needs to be put into place before you start like I suppose a very kind of telemedicine screening program using like one of these other imaging devices like the forest or the other ones that we know of? Yeah, well at, you see at the Kenyatta Hospital there actually is a red cam but it kind of came through the retina of lastoma program. So it's been used for retina of lastoma all along and not really much for ROP because the ROP there's a little bit more movement with moving each other into the NBU baby to baby. And so it hasn't really caught on, but it is there. We have had one of the Indian cameras for funders camera not for ROP screening, we don't have the hand gadget. So I think it's with more kind of lobbying because right now even the expansion of the ROP program has suffered because of funding. So we're in the process of trying to get some funding to kind of do more training for the peripheral facilities. Because we find that is kind of a lower fruit than investing so much money to one machine in one institution. So the next step might be to get a shared more affordable red cam that can be used to screen a couple of institutions. And so that's the kind of funding we're kind of following up through now with the retinopathy of prematurity working group which runs out nationally. Yeah, and there are more and more lower cost cameras and we can talk a bit about that later on. And I think, but maybe I think the initial thing is that in the beginning there is a kind of an increased time. Like people need to be trained to do telemedicine screening and if you're used to doing it a certain way within direct then again there's a sort of a time of learning and then training others and all of that. So there's quite a lot of initial investment not just in finances. So also in time of the people who are actually doing it. So that can also be quite a challenge. Is there any other questions from the panelists or from the team in Pakistan? I can see actually Anand has put something in the chat. He said, Anand has written, the fixed-stands advice that Dr. Mishai spoke about, there are two in India. The MIRECAM, as a MWIRECAM and C3, and they use your native phone in a fixed distance for the 20D at the end. They're good for photo documentation but not good for peripheral field. And for these also answer Dr. Moyne's question, images good enough to discharge the baby? Yes, if you can image the aura. And he said, this is possible in the RECAM in the Forest Neo and we have publications, images on the aura in the Neo. Yeah, and I think Anand will talk about that in his talk as well. So yeah, if you can obviously, if you can image up to the aura, then you're able to make clinical decisions. Okay, thank you very much, Mishai. It's really great to have you. Please do stay at the end for our kind of general discussion and panel discussion. And I will, yeah, we're gonna move on to now just hearing from Linda. So thank you, Mishai, and we'll speak to you again soon. Is there anything else, Professor Moyne and Gilbert, that you want to add and I'll ask just now? Okay, great, thank you. So we've now next got Professor Linda Visser and she's going to be talking about their experience in South Africa. Thank you for the opportunity to speak to you on ROP teleautomology screening with our African experience. We all know that improved neonatal care in middle income countries that led to a third epidemic of ethnopathy or prematureity and South Africa has not been spared. Without concomitant improvement in ROP screening or a peer-associated visual loss will remain a challenge. Teleautomology screening programs and artificial intelligence technology might be the key to address unmet demand in ROP screening. But the question is, has there been a move towards teleautomology screening for ROP in South Africa? I've been chatting to my colleagues about this question and the short answer is no. Reasons for this need further research but there may be different reasons depending on where one practices, whether it be in the private sector or insured sector or the public sector which is uninsured. And if you're on the private sector whether you are working in a city or a rural setting and if you are in the state whether you're in an academic or non-academic setting. South Africa has a binary type of health system private South African health care as well as academic public health care equal that of most developed economy health care but the non-academic public health care system tends to lag behind. In South Africa the current screening guidelines are those of the National Guidelines of 2013. These mandates screening for infants with a gestational age of less than 32 weeks or birth weight of less than 1500 grams. These are absolute criteria. But infants with birth weight between 1500 and 2000 grams should also be screened at the discretion of a neonatologist based on other facilities and especially if it's unbladed oxygen supplementation with administered to these infants. These guidelines on our 10 years old and under revision and new guidelines should be available from next year. And we plan to decrease the birth weight in the absolute criteria to 1250 grams. Some provinces have already lowered this screening criteria to birth weight of less than 250 grams and in the private sector this has been ongoing for a while. So how does it differ between the public and the private sector when it comes to screening? Well in the public sector the numbers are large babies are generally admitted to either regional or tertiary facilities in large centers. This is different from about 20 years ago where decentralization had taken place and the healthy bigger babies had been sent to the district area units and had been put on 100% oxygen by nasal prongs and not screened at all for retinopsis prematurety. At that point we had a large number of big babies developing RFP and going blind and was followed by a lot of medical legal action and that has meant now that all premature babies are kept either at regional or tertiary level until they've been screened at least once before they are decentralized. And we are definitely seeing a lot less RFP since this has happened. Screening in the state facility or public sector is done by medical officers or registrars. These are junior staff, not consultant pediatric ophthalmologists, but they're really good at doing the screening they usually do about 30 babies in the morning with using indirect ophthalmoscopy. Most academic units or at least five of them in the country have red cam facilities or equivalent contact wide field digital retinal imaging. This is however not used to do screening. These machines are typically used to monitor progress in type 2 ROP or to document changes that occur pre and post treatment for medical legal purposes. In the private sector, the number of babies are a lot smaller, but the number of hospitals we're screening these in are quite much more. These are units that are sometimes in larger private hospitals, but sometimes in small private hospitals where any one or two children need to be screened a week. The screening is done by pediatric ophthalmologists, so much more senior people. And again, most of them use indirect ophthalmoscopy to do the screening. The problem is that the number of doctors prepared to do the screening, the number of pediatric ophthalmologists prepared to screen is fast dwindling for two main reasons. One is that they have to pay an increased malpractice insurance levy. So those people who are doing screening are given a higher levy to pay. And this is largely because of previous medical legal claims. This is however unfair because the people doing the screening are generally not the ones who are sued as rather than neonatologists, but still the medical protection society have decided in South Africa that the screeners will also be penalized. The other factor is that the screening is actually quite time consuming. The ophthalmologist needs to drive to various units. It's time in the car, fuel. The process is a long process to get to various hospitals. Sometimes the babies are not dilated and they waste their time. And the remuneration for ROP screening in South Africa in the private sector is actually quite atrocious. There's almost nothing. So most pediatric ophthalmologists doing ROP screening in South Africa in the private sector are actually not getting paid for it. In fact, they are paying to do the screening. And for this reason, a lot have now stopped doing screening. And the dwindling number is a problem because some of the new neonates are now having to purchase cameras. And that's happened recently in Cape Town. The first private unit to get a camera in South Africa and start doing teleophthalmology ROP screening. I spoke to the specific doctor who's involved and she's quite unhappy. She says at least 50% of the images are actually quite poor quality and she still needs to actually go and see these babies most of the time. So this is just to point out where the facilities are that we are doing screening. So the provinces that are coloured in red, Quasulina-Tel, Carpentine province and the Western Cape are the provinces that have currently probably the better healthcare, the more private care and most of the academic units are in those three provinces. The other provinces are still going through the third epidemic. Some cases are still coming from those provinces but they definitely also have learned from the province of Quasulina-Tel which in the past was where most of the stage five ROP was seen and most have learned from that and are managing with babies a lot better. But the new NATO services in the other nine and six provinces are still not quite up to scratch. So the diamonds denote where the red cams are, the large diamonds and the smaller ones where the ROP screening can take place in secondary units as well. Generally in South Africa, Celia ophthalmology is commonly utilised especially in Cape Town where I am but also other parts of South Africa. Most referrals come to us via GP optometrists or our colleagues in the hospital via a app called RULA and patients are referred with a short history and usually a photograph of sorts. Diagnostic procedures for eye diseases rely on clinical assessment and image capturing with devices like fields and funnest photography. And these we know are well suited to deep learning techniques and because of that, a lot of the glaucoma and diabetic retinopathy screening currently is performed by Celia ophthalmology with more and more commonly artificial intelligence grading becoming available. Even with other Celia ophthalmology screenings such as glaucoma and diabetic retinopathy, there's still some barriers such as poor quality images and these can be patient related, these can be technician related. But there's also other problems in developing countries like internet speeds, bandwidth. If you want a really good quality funder's photo, you need to have quite a large bandwidth to send big files. The equipment is expensive, specifically in the case of ROP screening, retcams are socially expensive in South Africa. But even OCT machines, funder's cameras, et cetera, these can be quite expensive. Then employing and training your additional staff to be technicians to do good quality images. In South Africa, we have something called load shedding where electricity is cut off through certain areas to decrease the burden on the grid. And therefore one needs to install uninterrupted power supplies and that can also be costly. And then obviously as with everywhere else in the world, security and confidentiality of patient data will always remain a concern. But largely, many of these barriers are cost related. In the rest of the world, there are currently quite a number of success for ROP teleautomology models. This is from an article by Taneb from last year. And the well-known successful models include Art Rock in New Zealand, in Auckland, where specialist ROP nurses capture the images and the grading is done by pediatric ophthalmologists. Other well-known one is at the Sun Drop screening, which is done at Stanford, where NICU nurses capture images and they grade it by ROP specialist. In India, there are quite a number of these. The kid drops the most well-known, and I'm sure Anand will be talking about this. But in the case of kid drop, technicians actually capture and grade the images. And then there are other well-known ones in Germany, Chile, et cetera, that were discussed in this article. They do however say in the article that there's full limitations, even in these successful models within adequate quality of images, up to 21% of cases. And they found especially in infants with dark fungi and small purple fishes. Also, there's a significant initial capital outlay to purchase the equipment and to train the clinical staff for doing the screening and doing the imaging. And it's estimated to be up to 125,000 dollars, which is significant for a middle income country where the biggest burden of disease is. It's helpful that some of the newer cameras, like the three-in-five NIO, will be more affordable and will lead to more of these machines being available. Also, when you look at the AI artificial intelligence screening, certain algorithms are only validated for use with particular brands and models of imaging, which are devices. So even if you get the cheaper model, then you may not be able to use AI. AI, some of the models that currently are available and have been validated, show some variation in the classification of the disease. So, ROP AI and IROP DL focus on the detection of plus disease in particular. So it's one of the more simpler ones, but there are others that look specifically at criteria on the e-crop staging. Deep ROP uses criteria that are not strictly according to the e-crop staging, but they divide it into minor and severe. And the minor follow the e-crop staging one, two and zone one, two. So probably the disease that's not necessarily needs treatment. And then severe, which corresponds to type one and two or aggressive ROP or stage four or five ROP. These all demonstrate comparable sensitivity and specificity to other implemented screening techniques. But despite this, these systems have so far been seen limited implementation in their world settings. So from the same article by Tarnet O, in the Saudi Journal of Ophthalmology of last year, they come up with this flow-to-art of what the ideal ROP screening should be and that you start off with the infant that's identified according to screening criteria in your country. This infant is then taken for regional imaging and the images are uploaded and shared on a database. If you have artificial intelligence, that's great because the first screening is done by artificial intelligence. And if ROP is suspected, the images are then verified or looked at or seen by an expert ophthalmology. If there is no artificial intelligence available, the images are directly sent to an expert ophthalmologist who then obviously looks at the images, decides that there is ROP that needs treatment according to the guidelines or there is no ROP in the child can be discharged. So would South Africa ever get to this ideal scenario? I think so, yes. I think both in the public and private sector in the next few years, we will start seeing more teleophthalmology ROP screening. But certain criteria probably first need to be fulfilled. In the public sector, I think it will happen when imaging systems become less expensive, more mobile, and when enough technicians can be trained and employed to capture images on a full-time basis. So you have one technician that can then travel to multiple sites and do a good job, upload the images to either then be graded by artificial intelligence first, if possible, but these models will then also have to be more refined and become useful in all instances, even those with darker funding and with all imaging systems, even those that are much cheaper. But I do believe that in the coming years, this will probably start being rolled out in South Africa. In the private sector, I believe it might actually be coming sooner. And the reason I think it might be coming sooner is specifically for the reason where the pediatric ophthalmologists have decided that this is not worth very long to do screening. It's actually costing them. And it's a stressful thing for them to do and not being in a lunarative for it and paying very high insurance level. So because of the shortage of screening, there's no in the private sector, various NICUs will in the near future be purchasing these imaging equipment themselves and training the theater of the ICU nurses to do the imaging and probably then send the images to a pediatric ophthalmologist to grade. So I'm looking forward to the future where we will hopefully be doing tele-ophthalmology screening for all of you. Thank you. Thank you very much. And thank you, Linda, for sharing that experience. It was good to hear. It's really comprehensive. I was thinking of the questions after your talk, but you actually answered it. You went through everything so well and answered actually everything so well, like about what will be needed for South Africa to kind of take the next step and to sort of tele-mitage screening and all of that. How close, I mean, you mentioned a bit that the private sector might be closer than the public, but you know that now, of course, we're getting to that stage and I think Paul knows more and I know more about this than me, even is that, you know, we've got lower and lower cost cameras coming out all the time. And then we've also got these more and more of the AI algorithms being tested and shown to be working in different settings. So do you think maybe like that kind of public sector could also become in quite closer or do you think there are, though, there's the motivation there to kind of drive or advocacy for that to happen? As I say, I'm hopeful it will be if we have cheaper machines. The problem in the public sector is it's very much cost driven. So if the equipment costs a lot and the yield that you get from it is not great, then you're probably not gonna get those machines. And you don't wanna take a very expensive machine on the road to 10 different hospitals every day where it's liable or at risk of breaking down. So you, but I've looked at some of the machines like the forest, which is a much smaller, very mobile machine. You can put it in a suitcase and go and it's well protected in the suitcase, not gonna break on the way. So I think if one can get those at a reasonably very cheaper, much cheaper cost and you get someone in a car that drives from one hospital to the next, they can do 10 NICUs per day and look at all those babies, then I think it will become, it will become worth it, it will become beneficial. Currently, it's so easy for the doctors to, with indirect ophthalmoscopes, they do it really quickly nowadays to screen 30 babies in the morning, one doctor, 30 babies. You're not gonna get that with photography at this point. It's gonna take one technician much longer to screen 30 babies. And that's only part of it. Then it still needs to be uploaded, still needs to be looked at by someone else. So there's a lot more hours and a lot more people used to get to the same answer. And if your photos are not great and 20% of them you can't actually see, you still have to go and see those 20% of babies in the hospital face to face. So at the moment, I don't think it's cost effective and worth doing in the state. But yes, if we get really good technicians who can do a lot on a day, it will become more cost effective. But I think as I've just said in the private sector, I can see there's already possibly a move towards some of these NICUs buying the machines because they cannot actually find trained private ophthalmologist willing to do it because of the medical legal issue. But what Paul said earlier in South Africa, I think we'll never get the grading done only by AI or by non-medical people because of the medical legal issue. Someone has to take responsibility of making that diagnosis. And that's the person who's gonna be taken to court if it's a wrong diagnosis. So it's not gonna be AI. So AI can be a quick screen to if it's 100% sure this is definitely not gonna, not a problem baby. But if they can say, yes, it's a problem, look at it or no, it's not a problem, you don't have to look at it. But they're still gonna be human screening looking at the photography or the photos. But the AI can help to make the numbers smaller. Yeah, yeah, and I think that is the first step, isn't it? Is the AI experience there to do screen out the ones that are clearly normal and reduce that time of the ophthalmologist going to see babies which are kind of normal? I mean, it's a funny, I think in South Africa you're in that kind of balance between the sort of high income and the middle income sort of thing. Cause like somewhere where there's more ophthalmologists like in the UK as well, it's like seeing babies as quick to see babies by indirect, for us to then move to a camera is takes longer as well as the cost of that factor. But then I think we're gonna hear from Anand and I think that's the model that you were talking about. Oh, how can we take them like the forest in our suitcase into lots of different testing that's exactly what I was gonna talk about right now how they've made that work and made that cost effective. So, yeah, I mean, I don't know if you wanted to add anything just now about that or you've gone into screen sharing. Yeah, but is there any other questions for Linda at this stage? Moine, this is something you wanted to add right now. I think the answers have been rightly answered. I think let's hear from Anand. We're running out of time because we have to run another session and see it. Okay, let's get straight to Anand. Thank you, Linda. That was great. And I really enjoyed hearing about it. That's a great experience. Thank you, Ayesha and thank you, Dr. Moine for this opportunity. I'll just delve directly into what we've been doing for the past 15, 16 years in telemedicine and ROP. I think the necessity is a mother of invention. We have three and a half million babies born preterm, less than 150 ROP specialists. It's estimated that 200,000 babies require treatment every year. So clearly it's a demand, a supply issue. And I think the best way to bridge that was try to replace the so-called gold standard of using indirect ophthalmoscopy which very few ROP specialists are doing with wide field imaging. And we took that one step further and we said, can we have technicians both being trained to image and grade? And I'll just tell you the evolution. So this has been around now from 2007. We worked with public and private hospitals. We have committed to do free ROP screening and treatment in all public hospitals in the state of Karnataka which houses roughly about 65 million. That's the population of the state. This video may be a little slower on your screen but this is a 15-year-old slide. That's when we used to take the red cam shuttles. So yes, it may be looking like a challenge to taking relatively expensive equipment down on the roads. But believe me, we have not very comfortable roads in rural areas in Karnataka, even 15 years back. But we've been taking this very successfully and I think it's fairly robust. Of course, the newer camera is even better. So these go into the NICU, these trained technicians, images, take these images. We've taught them how to do the initial triage grading and then upload it onto a system. And that on the downright was the very first iPhone. So we've had this ability to look at it on the go. And well, yes, it was the first telemedicine program anywhere in the world to use non-physicians to screen and triage. It took us a long time to actually validate the program and validate the process. But I think we have it night now. The turnaround time is less than 20 minutes. So even though we have, and this paper in Jama, looks at the low internet speeds in some of the rural areas and how we are still able to get back with a good turnaround time. Fortunately, and this was not the first ROP screening guidelines. There was one in 2010, but this was the one, the National Operational Guidelines, which I think is probably the first country to allow non-Oftalmologists, non-ROP specialists. And you can see that there, trained technicians to use the camera. And I think this was a very landmark step. This has subsequently been followed in other neighboring countries of India as well. There are a few caveats on that screening guideline at the one, two, three, four rule. At least in India, we say two kilos or less than 34 weeks. And certainly the first screening before 30 days of life. And then of where it is possible, we can even do the treatment through the incubator wall. This video is to show you that it can be done through a scratched NICO wall. So I think, you know, convincing the neonatologists that it should not delay the screening is really important. These are the cards that we give out to the parents. And of course, we maintain all of the data online because it's imaging-based. This is roughly to show you how we have developed the process of training anyone, any technician or anyone who's gonna take the images to go through this process. So when they look at an image, they either decide it's green, orange or red based on their findings. And you can see the aura, serata and the lower left, where of course, these are images on the red cam subsequently on the neo as well. So subtler diagnosis of, let's say, a very subtle stage one in zone three, which you and I may miss on indirect, is definitely picked up on imaging devices. And then as you can see here, their ability to decide whether a baby can be discharged, which is green, needs follow-up, which is orange or possibly needs treatment or at least the specialist should see a would be marked as red. And this is the system that we've validated and we've trained scores of people across several countries now. And I think it seems to work with of course, a lot of safeguards that we've thrown in. As rightly said, the problem of scaling this up was the cost of the red cam. So we went into a sort of an academic partnership with Forrest, they are a Bangalore based company in the same city where I am in. And this is, of course, you've seen it. This has now got an FDA approval. It has plenty of installations in several, many countries, but this is wide field and they have now an HD model, which has the angiogram unit. There are some limitations in the resolution there, but it still does a very good field. This was our publication comparing the red cam and the Neo. Now, we've tried to introduce a few new systems in trying to expand and sort of scale up this program. For one, we have a online training program partly started before the COVID pandemic. This is how the traditional training happens when they come and work on mannequins, then hand-holding on live babies and then subsequently in the NICU itself. But we've been able to cut short this training program anywhere between 10 and 15 days and the rest of the training can happen with our resource people sitting here. And these are two male nurses in a peripheral center where they've trained a little bit, but their technique is getting refined by our online web portals and video calls. And I think this has allowed us to sort of screen in an expanded modality. Of course, a camera can do so much more than just taking ROP photos, as Dr. Mochai said, that red low blastoma. So here was a publication where we picked up so many other conditions. And now in fact, we are using cameras to do universal screening. And this is one of the mandates of the Indian government. This is slower to pick up than the ROP program, but it's still in a work in progress. And this is one of our publications where we found that even through the lockdowns through COVID pandemic, when literally there was no traffic on the road, we were able to continue the screening program and prevented blindness all around all those 18 months when we had different forms of lockdowns. Very rightly, as Dr. Linda said, medical legal aspects are a problem. And they're so are here as well. This is a very large, the largest compensation ever in 2015. There are 11 cases now all against the neonatologists, one of them against the ophthalmologists as well. And if more than ever, imaging now has become a cornerstone, if it's not running away from the program, it's actually increasing us, it's helping us to actually overcome that program. And we highlighted that in that publication. Where do we see the road going forward? And I think this was a recent Lancet paper and two of neonatologists from Australia are the other authors. What we looked at is there's a very successful polio eradication program in India, probably our most successful infectious disease prevention program. And what lessons can we learn from there? So if you look at the stuff on the right, the stuff written in black is what we've kind of already set out to do. And the red font is where we need to still work on. You need a band ambassador. You need to have adoption of these guidelines. We now have a guidelines, not really adopted. How do we expand to other states? And how do you have every case that is screened or misreported? And how do you have an independent monetary body? These are some of the things that we really need to look at. The exciting field of ROD, and these are two projects that I work on here in India. One is that we are able to now take the tiers of these babies and in our labs, able to predict some of these are, these are the markers. Of course, they're pro angiogenic, but there's also the family of inflammatory molecules and adhesion molecules. This was our first publication. We now use, we now have the ability to look at the, what we find in the tiers in the very first screening and then fairly accurately predict whether these babies are gonna have disease going forward. These are subsequent publications. In fact, both these came out this year. One looks at the inflammatory aspects and the other one actually looks at vitamin D and then its role in possibly preventing or modulating the disease very early on. AI, yes, we've been working on this now for over a decade, but initially it was just image processing, but now we've gone a step further. As some of our publications have shown, we now have the ability to sort of automatically do a binary classification. It goes through the algorithm that we've developed and the algorithm tells you whether there's disease or no disease. This is just an example of the algorithm telling you that there is stage two, whether it's a VS stage three or no ROP at all. Currently, the accuracy is about in the early 90s. It's good, but not awesome. And I think there's a lot of more work to be done. This is our most recent publication this year where we looked at it and how we can actually detect it in sort of a semi-lives scenario and also use AI to train people. The impact and I think this is so important when you're talking to policy makers and the governments and other hospitals. So we started off with three hospitals. We are 152 right now, more than 250,000 sessions, about 78,000 babies, unique babies and over 4,000 babies treated. We have almost proliferated across the country in several states and several countries have come here observed and sort of modeled it partially and for all of them have the local challenges that you spoke about. But I think the impact, the return on investment is the sort of the blind person years that you can prevent. We ourselves have prevented, and that's a slide here, we have prevented around 360 million US dollars of blindness and this is an older slide when we had treated about 3,000 odd babies and just expanding in a few states in India, we can prevent 100 million US dollars of blindness annually and that's really the impact that you can make. I'd like to end with thanking all our collaborators beginning with Government of India and all of the universities, both in India and abroad. I think I stuck to my 10 minute time. Thank you, Aishwarya. You did brilliantly, Anand. Thank you very much. Thank you so much. It's like absolutely fantastic. Thank you. I would like to play you to listen to Anand. He's done, I think, the most work out of after all and you're both head to head with what you do. One is in the west and one is in the east. Yeah, absolutely. Absolutely. Alain, go ahead. Yes, ma'am. So, mine is going to be shorter talks. The problem might be ending slightly in lesser time. It's telemedicine. I think it's over the few talks which we've learned, which we've listened to. We've found out that telemedicine can be problematic. It can be slower in places that the training of the person who's doing the screening is not as good as a person who's doing an indirect ophthalmoscopy. So, in Pakistan, telemedicine experience, we haven't had telemedicine being practiced a lot in Pakistan. So, we was working with Aisha and we're working with Paul Chen and we're working on how to get this telemedicine experience started. So, we're working as a timeline. So, this is the time when we finally were able to start the telemedicine screening. So, it is our initial experience in doing it and we'll get the results of what we have and how we got there later on. But we'll just experience why we were starting to on telemedicine is because of the burden of the disease. Our population premature birth rate is fourth in the world. There's six million births per year and the prematurity rate is 20% fourth in the world. And screening in 2012 was in two hospitals. And now we expanded on to more than 20 or 25 hospitals. So, out of 5% of all premature children born 35 weeks or less can go blind. And so, if that much is the percent of the blindness, technically this much amount of babies can go blind. But in the real sense, probably I think it's less compared to what it is and we'll find that out by getting our multi-centered data from across Pakistan. But the thing is if you've got this much of a burden, you cannot probably run it directly using the indirect screening in all those centers. You need more centers like Anand was mentioning to get them on telemedicine, getting help from either ophthalmic technicians or nurses to get the system started. So, we started off the awareness of ROP and the human resource development and training by Propera Pakistan, ROP and Educational Alliance. And I think we have to thank Claire Gilbert and Graham Quinn for being our mentors supporting this program and running it successfully. I think we've forgot the picture of Omar Mia. He's from Montefiore Medical Center in New York in Albert Einstein University. So, it was his idea along with the mentors that we need to have multiple NICUs. We need to have a city-wide coordinator collect the data centrally and then sort of publish data and get monitor the data what we are getting. Just an overview of how many people will, I was initially at the Law General Hospital for the last 10 years and over the last five years, we were able to screen about 1500 babies out of 3,500 babies and we were getting a mortality rate of 45% to the Law General Hospital and why we want ROP and then the screening criteria is a big issue for us. So for that, we sort of did a study on the number of blind babies which we're seeing and we were able to see that the screening criteria needs to be 35 weeks and two kg. And we're going to start the Lahore Mobile ROP project based on the different studies which, so for that we need a camera and then a photographer. So our main aim is to get outreach centers, ROP team and a reading center and this is the area which we are starting first. This is Sherpur Hospital, this is our screening team who went there yesterday for the first time and we were able to do one last week but this was the next visit we did. We've got a telemedicine reading facility in our university and they go a few kilometers to outside the city and this is, we were using the Zoom platform to actually see how the screening was being done and seeing the pictures. This is one test we were trying to do with the Zoom but the other thing as other people was do is just get the standard pictures. So you can see the camera fits in and we're using the forest camera for this thing. It gives and we need to, if you're taking the pictures you need to take five standard pictures and those will give you that the patient has disease. I think the most important thing is to pick up if somebody's got plus disease and if somebody's got plus disease obviously that patient needs to be seen pretty early. And this is the province of Punjab. Probably we're going to be in a very small area around Lahore and where we're going to start in three or four centers in our programs and the system of health centers in Pakistan is from district is the biggest center then you go to Tessil and a basic health unit. So we'll go to start off with various districts and for that we need to collect data. So we already have our database through which we can collect data and analyze data. So whatever is being done we can get the referrals and get them entered and run them over there. You've developed patient education material we are going to give over there. We are actually pasting charts on the hospital's walls telling about ROP what it is and why you need to be screened for that. So in 2015 we had this many centers across Pakistan and now we have these many centers which have increased dramatically and I think it's not one person's job but everybody's commitment and people who are sort of joining in and sort of joining for the cause to save the babies going blind across Pakistan. And we were able to publish a handbook for training and teaching of the human resource in our country. So it's easier to do I'd like to thank Claire Gilbert she's sitting with us that she's and Umar and Tayyiba and Lubna and Shai the Sand for working on this is a big task but eventually we were able to do it. And in conclusion, significant progress has been made more programs are involved they're better screening, improve neonatal care but telemedicine is the future to handle the burden of disease which is going to be tremendous in the few years to come. Thank you. Okay, that's it. Thank you very much for the morning and all and all our panelists today. Thank you all for joining us and for sorry for we've been running a bit late behind today but I think all the talks were incredible so I'm just delighted that you had everyone come together. I don't know if there's any other final comments or questions from anybody. Professor Mooyen and Claire Gilbert. Thank you very much Ayesha for arranging this and I think it went wonderful and I think my battery is running low anytime so you might just get disconnected. Yes, yes, sure. Congratulations for such an excellent webinar we learned a lot. Now we all know that this telemedicine is the chocolate but the important thing for the province is that how to get the right care? Because we are doing RV screening in three, four hospitals and they are all very dedicated for genetic ophthalmologist and we also need some support from because at least I tried hard with the different politicians and health ministers and we now seriously need some support from some donor agents. I think Ayesha can share details on how to write a project and how to pitch for that. I think... Yes, absolutely. I think also we've got Anand and Paul here and they're both very experienced with cameras and different lower cost cameras so I think of course now with the RECAM there's so many other options now and Anand's been using the four holes developing I mean you can speak a bit Paul about what you're currently developing on a lower cost camera as well. So I think now there's many, many more options coming onto the market which will make it more affordable and that will be... I've known this for a long time that that is going to be key like there's no way that telemedicine and RV screening is ever going to be able to be scaled up with the RECAM. It has to have those lower cost cameras and mobile cameras. So Paul or Anand, I'll let you speak because you do know more about it. Just a quick comment. I think so one, Anand's doing really great work with deforestation. Sorry, my kids are in the background and I think the forest I think is a really great option at this point. They've done a lot of good work I think over the past decade or so in trying to get better quality and really addressing the need and the ROP space. There are going to be other systems that come out in the next, I think five years or so. And I think that there's a lot of hope in that space. I put in the comments box. I think ultimately what will be really interesting is that if I could predict the future there probably will be some non-madriatic options which will be extraordinarily game-changing and I'm sure Anand can comment on that as well. But some of the things that Pete Campbell and so forth, he's been developing this one of his engineering partners, Yifan and I've also been developing some more cost systems with my partner, Shenyang. It's just gonna happen because we have to do it to care for this community. But right now, I would argue that and I'm sure curious to what Anand thinks that the forest is a terrific option at this point in terms of the compromise between costs and field of view and all the other things that we need to do. Yeah, I certainly agree. I think as of now, forest NIO is probably the best bet. But I know of another camera that should be ready before the next World R&D Congress, which I think is in 2024, I've been told. And I think that's also coming out of India and it's gonna be a much smaller one the size of like a hairdryer. And it's said to be about a half or one-third the cost of the current NIO. So that's one, and yes, I definitely agree with you about the non-materialistic, a smartphone-based non-materialistic camera is where we're going forward. And that's also, I think work in progress is not gonna be very far off down the line where that's gonna be a reality. Yeah, thank you, Anand. Brilliant. I think... I think they just dropped off. They just dropped off the entire conference, which I don't know if that was running on a battery or a laptop, but there we go. Maybe it's a sign. I think I'll probably just thank everyone. They say, thank you so much for your time today. Paul said early in the morning, I think with your family around, Linda from South Africa, Anand. And of course, Michelle, who's there from Kenya, I think has in general also dropped off as well. They've kind of epitomised them, the challenges we sometimes face. But thank you all. I've put, we've put again into the chat just like the ROP net website for everyone to sign up. And we will hope to hold more of these in the future and hopefully next time it will be in February and we will, through our emailing list, let everyone know. And yeah, watch this space for what's next. But look forward, great to connect with everyone today and thank you again for all our panelists. So thank you. Good to see you. Take care. Bye, Linda. Bye. Bye.