Projects

ppicheck.ai is an AI-powered clinical tool designed to optimize the use of proton pump inhibitors (PPIs) in healthcare settings. It intelligently audits prescriptions, identifies inappropriate use, predicts deprescribing success, and provides real-time, data-driven guidance based on patient risk profiles.
Built on advanced machine learning models and clinical scoring algorithms, ppicheck.ai empowers clinicians to make safer, more informed decisions — promoting rational pharmacotherapy and improving patient outcomes.
Fully open-source and clinician-friendly, it represents the future of intelligent, precision-driven medication management.

ppibot is a smart AI-powered clinical chatbot designed to support healthcare professionals with PPI prescription auditing, indication analysis, and deprescribing decisions. Currently in deployment, ppibot serves as the conversational interface for the intelligence behind ppicheck.ai, offering real-time, evidence-based recommendations through natural language interaction.

Powered by LLMs, clinical scoring logic, and machine learning predictions, ppibot transforms complex guidelines into simple, interactive support for rational drug use.
It’s fully integrated into the MedAI Lab ecosystem, accessible across platforms via a Streamlit + AI hybrid interface.

​PONV Risk Pro is an advanced AI-driven tool for predicting and managing postoperative nausea and vomiting (PONV). It combines hybrid clinical scoring systems (Apfel, Koivuranta, Bellville, SAND) with machine learning algorithms to deliver personalized risk assessments.
The platform adapts dynamically to patient profiles, surgical details, and drug doses, offering real-time, individualized predictions. With interactive visuals and research-grade precision, it supports both clinical care and academic use.
Now in deployment, PONV Risk Pro is designed to enhance perioperative planning and improve patient outcomes.

NephroPredict will be an advanced AI-driven tool aimed at predicting clinical outcomes in patients with nephrotic syndrome.
We plan to develop it using machine learning algorithms integrated with clinical risk profiling to forecast remission likelihood, relapse risk, treatment responsiveness, and complication probabilities.
By processing patient-specific clinical and laboratory data, NephroPredict will deliver personalized, data-informed predictions to assist clinicians in making optimized, evidence-based decisions.
It will become a key component of the MedAI Lab ecosystem, supporting precision nephrology through an open-source, research-oriented platform.