From Data to Diagnosis - Advancing Medical Imaging with Curated Datasets and AI Algorithms - MedAI #136

In this 58-minute Stanford University lecture, Dr. Debesh Jha presents his research on advancing medical imaging through curated datasets and AI algorithms. Explore how his team addresses challenges in radiology and gastrointestinal endoscopy diagnostics by developing comprehensive multinational datasets like CirrMRI600+, PolypDB, PolypGen, and Kvasir-SEG. Learn about innovative medical segmentation architectures including ResUNet++, DoubleUNet, ColonSegNet, and Transformer-based models that have established benchmarks in medical image segmentation. Dr. Jha, recognized among the Top 2% of Scientists Worldwide in AI and Biomedical Engineering, shares how integrating meticulously curated data with cutting-edge deep learning methodologies significantly improves diagnostic precision and creates robust, clinically reliable AI solutions for healthcare. This session is part of Stanford's MedAI Group Exchange Sessions, a weekly platform for examining key topics at the intersection of AI and medicine.