Wound Care AI for Diabetic Foot Infections. #diabetes #diabetesai #aifordiabetes

Hi, I'm Sonu. We are building an AI system to help detect and understand diabetic foot infection before they lead to amputation. In India alone, thousands of cases progress silently. Six Indians are amputated every day, and that's a huge number. When we talk about diabetes foot infections, they're not only skin problems. They are caused by multiple organisms, and diagnosis completely depends on delayed cultural lab test or subjective judgment. And that delay costs limbs. So, if you look at this app, this is how a clinician will log in with their credentials, right? And they can basically look at everything at the dashboard, like their previous patients, their previous cases. You know, they can add a new patient, or this also has integration capabilities with the EHR and EMRs. I'll show you an example of a previous case. Now, if I click on this, you can find out the already available results. Now, these results are basically showing the organisms which have caused the diabetes foot infections. Now, these are generated on top of the uploaded images that clinician has uploaded. And it also includes different comorbidities. So, this is not a black box that you see. We explicitly show the reasoning signals. So, visual markers, and then we show all the patterns, and then we also include the comorbidities, right? All the correlations comes up, so the clinician can trust the signals. That's what it does. Now, how it works basically is that our agents analyzes it and predicts the likely organisms which are involved, along with a grounded explanation of why, which is really important when it comes to explainability of the outputs in healthcare. Now, while you see more, you know, in the app, under the hood we are fine-tuning vision models on clinically labeled data sets, right? But more importantly, we are aligning them with the real feedback. So, we have created a feedback loop between clinician and the AI. So, if you see it over here, right? The feedback comes up from the podiatrist and diabetologist who are our end users to continuously improve the agent's performance. It's not just an app, you know, we are building an ecosystem for patients, doctors, and all the health care stakeholders who are involved. And this is personal for us. Both my parents are diabetic, and one of my co-founder's father has recently gone through diabetic foot infection. We know how fast it can escalate. As a startup founder, we always want to move fast. That's how we came across Emergent. We used Emergent to design this entire UI that helped us accelerate our design and development phase. We are not just building an AI system here. We are trying to make an impact because we all know that diabetes is already an epidemic, and it's going to become the next pandemic if you don't control it.