Every individual carries a unique story written across their molecular profiles, clinical records, and health trajectories. Modern machine learning gives us an unprecedented opportunity to read and understand these stories with depth and precision, to find patterns that were previously invisible, and to turn them into insights that improve lives.

Our lab develops multimodal deep learning methods that integrate electronic health records, multi-omics, and digital biomedical data to uncover disease mechanisms, improve risk stratification, and support clinical decision-making. But powerful algorithms are only as good as their impact. We focus on building robust, trustworthy and generalizable models that are grounded in biomedical knowledge and perform in real-world settings, ensuring future data-driven medicine efficiently reaches everyone.

We believe that the greatest scientific breakthroughs happen at the intersection of disciplines, and we warmly welcome collaborations with clinicians, biologists, and researchers to translate computational insights into meaningful biomedical advances.