Adam Yala, PhD

Adam is building AI that helps radiology predict cancer years before it surfaces. Renowned for developing the deep-learning models Mirai and Sybil, he created Mirai to generate a personalized five-year breast cancer risk score from a mammogram. His nominator reports it has been deployed in 115 hospitals across 33 countries and is the subject of multiple clinical trials. Meanwhile his Sybil model predicts lung cancer risk up to six years in advance from low-dose chest CT. His Mirai research ranked among Radiology’s top 10 papers of 2019.

Adam earned his bachelor’s, master of engineering and PhD in computer science at MIT, and received an Empirical Methods in Natural Language Processing Best Paper Award and a National Science Foundation graduate fellowship. In 2022, he joined the inaugural faculty of UC Berkeley’s computational precision health program and helped establish its PhD program.

He then founded Voio, a venture-backed frontier AI lab for healthcare, with UCSF radiologist Maggie Chung, MD. Its first vision-language model, Pillar-0, interprets medical images and generates radiology reports; the nomination says it outperformed models from Google, Microsoft and Alibaba by 10 to 17 points across hundreds of findings. The team released it open source for independent academic validation. His recent work also includes prospective AI risk stratification to speed mammography in a safety-net setting.