Daniel Rubin

The QIAI lab focuses on cutting‐edge research at the intersection of imaging science and biomedical informatics, developing and applying AI methods to large amounts of medical data for biomedical discovery, precision medicine, and precision health (early detection and prediction of future disease). The lab develops novel methods in text and image analysis and AI, including multi-modal and multi-task learning, weak supervision, knowledge representation, natural language processing, and decision theory to tackle the challenges of leveraging medical Big Data. Our exciting work is bridging a spectrum of biomedical domains with multidisciplinary collaborations with top scientists at Stanford as well as with other institutions internationally. The QIAI lab provides a unique multidisciplinary environment for conducing innovative AI-based healthcare research with a strong record of scholarly publication and achievement. Core research topics in the laboratory include: (1) automated image annotation using unsupervised methods of processing associated radiology reports using word embeddings and related methods; (2) developing methods of analyzing longitudinal EMR data to predict clinical outcomes and best treatments, (3) creating multi-modal deep learning models integrating multi-dimensional EMR and other data to discover electronic phenotypes of disease, (4) developing AI models with noisy or sparse labels (weak supervision), and cross-modal, multi-task learning, and observational AI approaches, and (5) developing and implementing algorithms for distributed computation for training deep learning models that leverage multi-institutional data while avoiding the barriers to data sharing.