Keynote by Sangita Reddy Joint Managing Director Apollo Hospitals India AI Impact Summit

India and that your health care should not be defined by the zip code in which you’re born. It’s about sustainable costs and it’s about preventive care and early detection. It’s a new paradigm in collaborative care where I believe India has an advantage. This advantage is because we not only have one of the highest out -of -pocket payment and therefore we’re creating innovation and keeping our costs low, but also we’re growing more doctors, we’re training more nurses, and we have the largest talent pool of over 600 ,000 AI engineers. All this coming together to create something truly significant. But I’m not here to talk to you about technology. I’m here to share our story. And this story is about using the passion and the mission of bringing health care within the reach of people and using every tool possible to enable this to happen.

Dr. Pratap Siredi. I’m the art chairman and I’m honored to say my father. brought to polar hospitals when he returned from the U .S. almost 43 years ago to bring this, to bring healthcare within the reach of people. Today, we’ve tried to embed and imbibe every technology, whether it’s surgical robots, the proton therapy, all kinds of treatment and curative capability. We’ve gone beyond to say we must find a way to not just use these machines, but also to connect with our customer. So Apollo 24 -7, our digital front door, is actually, not only can you buy your medicines, order your diagnostics, store your health record, but also on Apollo Assist, ask queries, questions, get these answered, and then find ways.

And our market has rewarded us with the volumes that we see. Over 45 million users have come into this. and now we have close to a million users on a daily basis coming in to interact on this ecosystem. These records, these capabilities are getting enhanced every day because of the power of the communications that we have. But moving on, I think what is most important is that we’re not just in the big cities. We’re serving multiple PIN codes across the country and over 1 ,100 towns and cities. Moving across divine methodologies, I just wanted to share with you quickly a few of the things that we’re doing in AI because this is the AI summit. And approximately now we have about 3 .5 million API calls on our AI platforms.

These platforms we’ve divided into five areas. Number one is really our clinical intelligence engine so that a new doctor can have the knowledge and the capability of the cumulative data that we’re providing to the patient. And number two is the cumulative doctor workforce of about 20 million records analyzed. So this is our clinical decision support and our clinical intelligence engine. The next one is the disease prediction and the risk score, because we need to know in a population of 1 .4 billion people, where do we focus? What should we do more? So this is the second work stream, and this goes across cardiac, diabetes, multiple others, including hypertension, but we’re also looking at embedded AI. The next and another critical one is taking images and signals, because the body is an amazing piece of machinery that continues to give us this messaging.

How do we pick this up, synthesize it smarter than any one individual can do, and bring this multimodal signaling into a causal interpretation to thereby enable the doctor. We also have acute care augmented pathways. About 2 ,000 of our critical care beds are connected with our early warning symptom, and there we are predicting. The onset of sepsis. 24 to 48 hours before it happens. Imagine if we could take this AI algorithm and put it into a hundred thousand ICU beds. Imagine the number of lives saved. So here I’m sharing these examples because I believe that the power of AI is directly proportionate to the impact that we can have on lives saved, disease prevented, cost reduction, and therefore talking about cost reduction, the final one is really throughput optimization.

How can you be smarter about billing? How can you ensure that your patient has zero waiting time, that the data capture is using ambient systems, therefore the doctor is able to look at the patient and talk to the patient and you’re doing auto -population of your records. Millions of these capabilities are coming together. We’ve collated them. We’re getting MDSAP approval on almost 19 of them, FDA approval for nine, and we’re looking for partnership to build because I believe in this space. a thousand flowers can bloom, and that there is deeper work to be done on the use of our blood bank and our biobank with genetic testing to move further into disease prediction, biomarkers. So these are just new dimensions opening up.

And I’m sharing more of the examples of how we’re working in these areas, but before I go into those, I want to talk about the EASE framework. I’m happy that our EASE framework has been published fairly extensively because it talks about the ethical considerations of the use of AI. It looks at adoption, the suitability of a certain algorithm within the area that it’s being used, and finally the explainability so that every healthcare worker is able to understand what they use in which environment and what the interpretation means. I believe this is a base framework that we need to put into every healthcare environment. Moving on is another area of deep passion, and that is that while we’re doing the highest end of surgeries, curative care, transplants, etc.

How much can we spend our time on health care prevention? Because for every life -saving intervention, for every 1 ,000 people screened, you will have 11 people where you have averted a major crisis. And therefore, the ability to look at proactive preventive care and get a lot more intuitive on the mechanism of biomarkers and early detection in cancer. We are working with the ultrasound company to do an embedded AI into the ultrasound machine so that we can pick up NAFLD, non -alcoholic fatty liver, of which 40 % of the adult population of India is susceptible to. And if you can pick it up early, you can completely prevent a major crisis if you find it late. These are candidates for liver transplant, a lot of pain and suffering, and some of them potential death.

So the interventions at the appropriate time using technology open up an entire… realm of what we can do differently in this world. I’m sharing now this aspect of how lifestyle changes risk reduction. All of you on Instagram are getting thousands of messages a day on what to eat, how to exercise, what to do better. But is it quantified? Is there a risk scoring? Do you understand the difference between a high -risk group and what they need to do to a low -risk group? But every single group, by understanding the risk profiling and the modifiable risk factors of these non -communicable disease can move into a healthy pattern. This has been studied in partnership with Solventum, the company with 3M, with definitive proof on the power of doing something like this.

We also have a significant product on AI prediabetes which I think we’ve used for a long time. We’ve used it for a long time. We’ve used it for a long time. We’ve used this algorithm over 450 ,000 people. But I would love to see the 85 million diabetics in our country using this to predict and to handle their diabetes better. I also want to move on to the fact that in radiology, because of the years of data and the teleradiology services that we do across the world, we are able to take these images, and here we’ve worked with Google on prediction of tuberculosis in a simple x -ray. We’re working with various other companies, whether it’s an early detection of a brain bleed.

So once somebody goes into the emergency room, you’re quickly able to diagnose these. Each one of these are amazing new factors which are coming in. This is a quick example of the clinician co -pilot. Because I’m running out of time, I’m not going to share this video, but basically… Okay, they are playing the video. Can we have some volume on this? Or I’ll click through, because we’re really running out of time. But basically what the clinician co -pilot does is it’s synthesizing the record so that you’re summarizing. We’re approximately saving… We’re saving one to one and a half hours per day of doctor time in the records. We’re now doing the nurse pilot. I’m moving now to reimagining the way patients are monitored, whether it’s the challenge of a misdiagnosis, the integrated solution, which is looking at Care Console, and the technology stack around this, which is connecting the command station with the ICUs, with home, and with connected wards.

And because of this, we’ve not only saved millions of lives, we’ve saved time for doctors, and this is connected even to external nursing homes in small rural areas. I believe this is a powerful solution where the current AI algorithm has multiple factors from antibiotic usage to early warning symptoms of sepsis, but there are potentially another hundred algorithms that we could add on to this to enhance the quality of decision -making. And share this further, enabling a safer patient care and also less burnout in our staff. I’ve been sharing lots of hospital -based examples, but I do want to say that many of the solutions are applicable in rural India. We’re running mobile vans, we’re doing non -communicable disease screening in small rural environments, we’re finding ways to do cancer screening, tele -ophthalmology screening, and sharing this data and enabling either the ASHA worker or the district health authorities or even the government hospitals to diagnose faster, better, cheaper, and earlier.

And this is really the power of what can be done through early screening. I also do want to say, because for those who are listening from research organizations, from pharmaceuticals, from manufacturing, that we are among the people doing the largest number of validations. So innovation happens from multiple quarters, but validation is what moves a pilot into a mainstream activity. And that is what is critical for our country because you’ve been hearing this over the last two days about the number of pilots happening, but we’re not finding ways to continue this. I believe the hospital of the future is interconnected in multiple ways, from the theatres to the ICUs to using drone delivery. But then as we were drawing and designing this, we actually said, no, our thinking is too small and narrow.

We need to think bigger because the world is more connected. And primary care, preventive care, out there in the market, home care, these are the important redefinition factors of the future of healthcare. And so now I talk not about hospitals of the future, but about health systems of the future. This is what we need to redefine, and we have to do this together. These health systems of the future connect public and private, connect primary care with advanced care, connect research institutions, universities, innovators, health tech startups, all together to build new solutions for the betterment of healthcare. And I believe that this is a flywheel which will drive not just positive health productivity and the economics of the healthcare environment, but this data will enhance into new algorithms.

And these algorithms can be predictive and preventive, and if you find disease earlier, you’re actually saving so many aspects. So let us remove skill gaps. Let us push through regulatory gaps. Let us bring companies, organizations, and people together to build a new healthcare world, which is predictive, preventive, personalized, participatory, and place agnostic. Let every village in any part of the world, or every city, or every apartment building, wherever you are, be able to access good clinical care. Let’s come together to build a healthier world. And definitely, let’s say that this is the time for us to… to dream of finding cures for cancer, of enabling the world to be healthier, and finding a methodology for us to say that we brought our next generation into a healthier world.

Thank you so much, and namaste. Thank you. Thank you.