As mentioned, I would like to talk about how we have been shaping digital health in the Turkish health care system environment.But also, we have been exporting these solutions for the last five years so that more patients around different countries can also benefit from it. Before going into the solutions, I would like to spend some time on the needs. So why we needed to implement digital health to support the disease health care structures so that the health systems and patients can benefit from it. The need is basically the number of increased chronic diseases. If you see the right hand side chart and how steep the blue line, which is noncommunicable diseases, their burden has been increasing.

Because if you go to the backhand side, the average life expectancy has becoming 80, which is a very positive thing, of course. But it's also a fact that over 65 years of age, chronic diseases prevalence becomes around eighty five percent. It means it's becoming more risky. From another point of view, from my own story, I have been one of the chronic diseases patient this summer and still going through it with good health. But I have been facing regardless of the age, the burden of the disease is impacted on us women very heavily, especially in breast cancer. What it affects is basically health systems and patients. Let's look at how it impacts the health systems. Of course, there are more chronic patients, and more patients means more visit to the hospitals.

You see a steep curve of how number of population visits and hospitals are going. The part is from the 2024 IQVIA global study that one third of physicians are planning to leave profession. It's kind of linked to the box, which is so many extreme number of patients. In cases, we see a hospital physician has around five minutes for a chronic disease patient. It makes the burnout and harder to manage the patient load, the burnout leading to leading the profession, making the case further accelerated, further impacted. The two bags combined kind of generate the one, which is elongated waiting times. It's coming from the very busy hospital floors and very late scheduling in their care, visits.

Many of the patients even drop their treatments because it becoming a hurdle to have a good physical visit in the hospital. Well, this is on the health systems and not to mention the burden on the payers of these chronic chronic patients increase. Let's look at what the effect on patients are. This is driven from some focus groups and, market research we have conducted that due to the availability of many resources and hurdles in reaching the hospital, the patients unfortunately adopt using online resources, which in many of the cases are unfortunately misleading.

and the complex chronic disease treatment requires a quite disciplined regime in terms of managing the doses appointments, which is many of the time not managed well. one, side effects or progression signals is very important for the physician to decide on the regime or maybe change the treatment along the pathway, which is harder to manage considering the workload. one, the supporting treatments like nutrition or physical therapies are usually missed or, let's say, less prioritized by both sides. And one is kind of a result that adherence to treatment is becoming an issue. Patients seem to drop along the way, which is very much impacting their overall survival. So what we did, we didn't just sit in a room and designs, you know, what we can do digitally to resolve that. We gathered many roundtables that are combining health tech companies, physicians, hospitals, patient associations. So we really discussed what can be done to make the journey easier and better.

And some of the local need identification underlined, there is a huge need of digital health tools that support the patient's journey as well as the physician's journey, which you can see, couple of examples of. And while we apply those learnings, we wanted to deliver as much as possible because this is a complex area that design and delivery takes time. So we kind of bundled them in a portfolio, applied design thinking, and delivered solutions with agile methodology, continuously following the feedback and improving them. Now giving you a part of the background about why we needed to do all of these things is then what the solution is, what kind of change portfolio that came out of it. To understand that, we have grouped the solutions in terms of six, which are kind of categorized under three different phases of the patients. These patients are going through a very, long and cumbersome diagnosis process. Then their treatment is decided.

And since this is a chronic patient, there is a long care period where they should track their health very well. For diagnosis, we will go through auto detection and virtual case analysis, for treatment health assistance and decision support, and for care self monitoring and physician portal solutions. So let's start with diagnosis. In diagnosis, it is very crucial to make the diagnosis as fast as possible so the next step treatment can start, but also as accurate as possible. Maybe even accuracy is more critical than the pace. In order to ensure that, we would like to share the solution, which is auto detection. As a matter of fact, due to the workload I mentioned at the beginning, it's reflection on the pathologists and radiologists.

There are many cases when a report of a diagnosis takes fifty, sixty days. Plus, according to research done in couple of hospitals, in breast cancers, there is a false negative ratio of twenty percent, which means one patient, one woman out of five who care to go through all these screening is having a tumor but not diagnosed at the early stage. I am a part of those one in five, actually, because the radiology images has been accumulated for the radiologist to be taken care of for too long, and there is a burnout as I mentioned. So this solution, what it does, by image recognition of artificial intelligence, it transfers the image high resolution to a database where there has been prediagnosis of, breast cancer, compares in a pixel manner as you see on the below photo, and gives a risk score like you see in the green, yellow, and red.

On the red score, it gives an intention like a double eye principle, to be reviewed by the machine, and then the risk score is transferred to the radiologist or the pathologist. Another thing, if there is a high risk, the waiting line is narrowed, which means this case of high risk tumor or lesion or any other chronic disease is prioritized on their workload pages. We have seen that, there has been 96% accuracy in the test group that we have applied compared to the 80% accuracy in the university work that has been conducted. So earlier diagnosis and earlier treatment is possible. one is what we call a virtual case analysis. This case analysis is regarding creating a multidisciplinary, multidomain board to analyze a complex case. Chronic cases are getting more and more complex, and it requires many different disciplines like pathology, radiology, oncology, surgeons, so on and so And these centers don't always have the specific expertise in their systems, or we have even seen the supported countries that don't have that experience at all.

The solution, virtual case analysis, allows multiple domain physicians join a virtual board. As you see on the video now, they co direct images or videos collectively. They define the right diagnosis and even link it to the right treatment via these boards and capture them as nodes in terms of next best action. These tools, when they are applied, they block the they they remove the blockage of not having the right expertise in the center or in the city. They also accelerate the process and impact the overall survival. There has been measurements that claim seventy percent survival improvement compared to a patient going through a virtual board, a collective board, and a control group. So this, we believe, is very critical to make the right case analysis. Having look at the diagnosis, we can move to treatment.

Treatment is the core of the physician applying the medicine, so the tools are more supporting the physician at the back. The one is health assistance. To give you a bit of a background and medical information with our current setting of medical clinical studies, new drugs coming into place, six months make medical information obsolete. So it is a very hard and time consuming work to be on top. On top, due to complexity of treatments, there are 10 to 15 guidelines to watch to decide the treatment as well. Also, physicians face through many daily activities calculating doses or side effects. So the tool you see on the left hand side, it's a web page, and on the right hand side, there is a mobile application that has 15,000 users.

These users have their profiles, and 5,000 medical content is personalized with large language model models and works like a CRM. It shows what the physician would be most interested in in their fraud pages. So the physicians save a lot of operational time in their daily work. one is a decision support system. Now consider you give all the information, but the patient has very many digital biomarkers that change the treatment due to the, again, the complexity and number of treatments available. This is a rules engine based algorithm that matches to the best guidelines to the treatment guidelines. And according to the situation of the patient where the patient's subgroup belong to, it suggests the ideal next best action, of course, to be given by the physician themselves.

We measure only in the pilot, but around fifteen days turnaround time drop while using the system. And considering extremely late stage metastatic patients, these fifteen days really means, survival. Now looking into treatment, we move through care because, okay, the treatment is given, but patient is in a lifelong journey of good health and good support and adherence to the continuation of the treatments. So what we proposed and delivered here is two solutions, but the same interlinked. One is for the patient, which is an app. They can control their disease journey and a representing web portal for the position where they trace their patients. have a look at the patient's application. This works for differently for other chronic diseases as well, but as a portfolio, it allows digital health tracking of the patients. They can enter their side effects, appointments, unexpected situations to the system. Also, even has an online tracking system of things like walking performance to show their disability rating.

On top, by, again, language models, they can use speech to text voice assistance if they have, disabilities to type, to ask their questions from a verified physician delivered information, book their telemedicine, which means you see on the right hand side, thirteen percent to two percent drop of hospital visits.

So it kind of eliminates less necessary hospital visits to release again the workload and ability for physician to

Patient is now asked patient is now asked to give permission to share their data with their health care professional. Given Given the permission, we're not able to ACP to monitor the data of the patient. On the main screen, patients are offered to track their BMI score. They can also set reminders for BMI score to track it later.

Are very critical for the

Now we see

manage in terms of taking the right action. So what happens when the patients fill this information is the solution. This is the physician portal where the digital information, the digital biomarker of those patients fall in two. This way, the physician can set alerts. For instance, if this person's BMI becomes below 10, just give me a warning. So alerts can be put in to call the patients for an early treatment, But, also, it works as a real world registry for r and d purposes, for physicians to generate their papers and academic research, not applying to a lengthy process of specific registries, but using their own databases to create research as well, which brings us to the realized impact.

There are over 5,000 users with 97 satisfaction ratio, And users, almost half of them use this symptom journal or telemedicine appointments very actively, which means they're adherence to the system as well. Now I have mentioned the impact underlining each specific solutions around my presentation, but I wanna end with the most critical and most general impact, which is impact on patients. Due to accurate information, health assistance, and treatment adherence acceleration, we can see impact on overall survival. The study you see on the right hand side is only done with digital patients monitoring. And I can tell you nine months survival impact is beyond any available innovative treatment today. So having the patient on board on the driving seat and having a close coordination with their physician has a great impact on the overall survival.

Just imagine the six portfolio of solutions I have mentioned that generate the collaborative effect. The impact would be unmeasurable. I hope that these solutions enlighten and suggested with others, create a genuine portfolio for digital health to improve and, hopefully, health care systems to be improving patient journeys and more sustainable. Thank you for listening. You can contact me from both links.