Good afternoon, good evening. Um And uh good morning to anyone who is in all the time zones. I am excited to be here and have been enjoying the conference since morning.It has been such a great programming that um you know, have learned a lot from a lot of speakers. Mm As Margo mentioned, I ha I am, I work at CS. I'm a lead principal here managing our ecosystem and alliances uh space. And in my role, I come across a lot of um newer technologies as well as uh companies bringing the those new technologies out to market. Um And uh one of my primary jobs is to see how those technologies can be brought out in a consolidated solution to our clients here at Zs. Just very quickly. We are a software. Uh We are a consulting firm and we help um most of the healthcare industry and beyond healthcare um in their business consulting as well as the complex solutions around consulting and um technology. I thought we could spend some time today in um talking about what uh what we see in the space around.

Uh he uh with A I in healthcare and just walk you through a few things that we have learned um as we go uh and help our clients in taking uh you know, uh optimizing A I and building A I applications for use within the healthcare and life sciences space. Um So with that, um wanted to um just get um started around uh where um oh why do we think that this is the right time for A I now we've been talking about A I for quite some time. But um why do we think that this is the right time? And why is why are we at the inflection point? So there are a few things which are a confluence of things happening externally that are all coming together, which is forming like a right, right situation for A I revolution in healthcare. One of them is exploding data volumes, right? We we um uh from right from fitbits and Apple watches all the way to uh uh medical devices that are inserted within our body. We are generating a huge amount of data. So there is a statistic that we are gonna be crossing hundreds of cab bytes of data that we are collecting on human health on ourselves, right?

And then that data is a extremely rich data that can be used uh for creating productive applications, uh programmable biology, right, cell and gene therapies that are coming to market is from uh from the need for programmable bio biology and excuse me and with ultra um high dimensional molecular data which is helping us create uh uh extremely high precision medicine for human body um productivity promise.

Um It is accelerating use of data and A I is giving us insights and showing early promise to inefficiency. So efficiency that comes at a higher cost initially always uh will create to uh will lead to greater effectiveness through uh better applications, better outcomes, better diagnosis and treatment.

So if we look at where we are in current state of A I in healthcare, um healthcare ecosystem is getting ready for what we call the fourth industrial revolution. What do you mean by that? I'm sure you're familiar with that term. But the fourth industrial revolution is essentially an in interconnect is heavy interconnectivity, um uh automation, machine learning and real time data. So all of these things kind of coming together and applications being built across um uh across these uh healthcare ecosystem with these features is what the next revolution is going to be. Um We are looking at um digital caregiving and new understandings of what is health and what is care which is going to pay for better outcomes in health care efficiency will also lead to, as I said, better healthcare prevention as well as diag diagnosis and um treatment uh for healthcare.

So you see some stats here where we, we uh in our work, we've seen 56% of life sciences say that they have the right um management in place to introduce more A I into work. And three times more hospitals in the US uh are going to be implementing um A I uh uh as we conduct our research. So what are some of the A I use cases? Right? W when we talk about um A I uh it's a big field, you can apply it to any aspect of uh health care. But if you zone in um I'm just taking an example of um the survey that we did with Pharma uh executives with pharmaceutical company executives. And we asked them that in the next year, what are some of the areas that you are likely to be using A I? And you see that a lot of it is within clinical trials with the drug discovery with real world evidence. I mean, they always um shoot up to the top because looking for new medications and looking for new molecules is something that is top of mind uh for all of the Pharma executives but not lost upon us is all of the other pieces which around sales and marketing and manufacturing and supply chain, right?

Getting um all these few areas to the top is something that is top of mind in linking these most executives say that linking these areas to our strategy and making it scalable is what um we uh we are looking to do in the coming years. So um uh Furthermore, we asked the same set of executives, what are the areas that can benefit from A I? Right. And we talked about new drug candidates, real world evidence, um faster clinical trials but not lost is personalized customer experiences or customer insights, right?

Or more and more we've already seen within uh areas of retail, financial services. We want more personalized care, uh personalized recommendations or personalized uh um uh digital advertising. Uh it is health care is not not lost on that. More and more patients personalized customer experiences, right?

Everything from personalized medicine, all the way to how I'm treated when I walk into an hospital is all personalized now. And a lot of this has to go back to all of the applications that are created with the data and A I uh applications that can help with this personalization. So enabling this journey from data to impact uh depends on a few things done uh uh done, right? As you move um across the spectrum um having data um uh uh having a rich data foundation which then gets into building the knowledge layer and then helping out with um developing applications for insights which are very much connected into driving impact, which is, which comes through business processes and programmatic approach.

So the journey from data to impact is pretty well documented and in industry is on this journey of starting um to deliver impact um to the healthcare um healthcare stakeholders. Some however, some typical stumbling blocks that we see within our research is there's six things that most companies uh trip on, some of them are good in some areas versus um others. But here are some typical stumbling blocks that we see. Some uh it's organizational mindset.

Um some struggle around analytical sophistication. And what we mean is we're not used to uh not knowing which business processes should we uh be using um A I for or uh what are the right use cases to bring uh forward which will drive the highest impact we also see of course talent shortage across the board in terms of having rights skills.

Um and having the um right efficiency within those skills operating model stands out as another one as well as you know, as much uh as well as data. So very critical is to make sure that um the data frag uh data is not fragmented and it's not hard to work with. So having that data foundation um is uh a place where most companies also find themselves working hard on getting that together in health care. The other piece to um note is uh the data privacy and security is another aspect that most companies are also worried about and hence kind of want to go a little slow before they jump on this bandwagon. Of course, technical infrastructure with technology changing so rapidly is another space where many companies um are also um trying to uh scale up. Um So the industry has made good progress in spite of all these stumbling blocks and we've moved from being in the basic A I space to more foundational to advanced A I space. Within the last I would say, 2 to 3 years, right? Where we need to be uh where the healthcare industry needs to be is more around transformational, which is where um it is going to realize the true potential of uh of A I applications. Um uh People, let's talk a little bit about people.

So it's not it takes a village to build these applications, right? We heard a lot about data scientists that we need a lot more data scientists, but it's not just data scientists that are needed. You need folks who can do experimentation in model development. So the domain experts and data scientists and there is a handshake and collaboration that needs to happen between the data scientists and domain experts to make sure that you bring the best um use cases to bear through these applications. Um The other end of the spectrum, you need model development and life cycle management. Folks who can scale up um uh the applications that are developed and actually um the drive the impact that these applications are supposed to um deliver. So um uh uh our research further and in, you know, it's interesting, I've been listening to some of the um some of the talks earlier on the stage and everyone has been saying uh talking about diversity in uh our research points exactly to the same spot where we find that diverse teams build a much better A I applications.

So um in, in the reason there are some critical reasons for that, that A I learns from data generated by human actions and mimics our biases. So um the more diverse the team is in the more um integrated, the team is with the kinds of applications that are built. Um The better applications come to market, right? You need a diverse uh you need teams with creative thinking, you need teams that are um uh that are built such that they avoid this bias in programming. And uh uh the uh the teams that are diverse in terms of race, gender, um age, economic conditions, all of these kind of start becoming more uh start taking out biases and start building applications that are more creative. Um So, um it's something that I heard said before. Uh also uh we reiterate that the, you know, our research points to and what we do here at CS is always build teams uh for developing A I applications that resemble the people who are going to adopt it. You cannot always do it 1 to 1, but the more biases you can take out um the better applications that come into market. So here at ESV approach, um uh A I um with trust in mind, right?

The big thing that uh we are focused on is making sure that we build tru uh build trust in the A I um applications that come out in the market. Um We do that by um responsibility. And what I mean by responsibility is um you know, there are some problems that A I should not solve and making sure that you are picking problems that the A I applications can benefit most of and not using. Uh And that there is no irresponsible management of data or algorithms that can instill wise, right? Competence, innovations have to work. You are uh dealing with um human life. You are uh these A I applications directly impact with. If you're billing them for healthcare, they directly impact human life. So your innovations have to work. The healthcare ecosystem will also need to come to terms with what defines as acceptable error of margin when you're in within these innovations. And once they do these innovations have to work transparency, right? Being upfront about limitations of what digital and data um in A A I can do um can help maintain this trust. So we are we always look for um transparency as we um as we build our applications um in net net.

Um wanted to leave you all with three takeaways that as uh industry uh starts harnessing the data that is available uh as industry starts becoming a more efficient in A I applications bring that efficiency. We are gonna see better healthcare, prevention, diagnosis and treatment.

So the more we can help people from becoming patients, which is where the prevention will play a role. Um It is going to um uh create a healthier society. And once you uh and if you are a patient, if A I can help you diagnose it early and treat it early, you're gonna have a much healthier outcome um across the board. The fourth industrial revolution of digital data and A I is unleashing healthcare's new future. And this means that we are going to need a lot more skill sets, a diverse set of talent pool um that is uh ready to uh build these applications as well as scale these applications uh into um the market. And last but not the least for A I to be successful. We need to build diverse teams and create diverse solutions that look like us, feel like us and are are created by teams who uh for people who would be using them. So with that, thank you for having me. And it is wonderful to um share some of the research that we have done.