Dr. Sohn is a clinical cardiothoracic radiologist whose research focuses on the clinical translation of artificial intelligence in medical imaging. His work centers on developing AI systems that improve radiology workflow efficiency, image interpretation, and clinical communication.

His research team brings together physicians, engineers, and imaging scientists to develop and evaluate clinically deployable AI and quantitative imaging methods. The group focuses on solving practical imaging challenges while establishing rigorous frameworks for validation, safety, and responsible clinical implementation. His work seeks to bridge advances in artificial intelligence with practical clinical implementation, with a focus on safety, evaluation rigor, and measurable impact on patient care and radiology operations.

He practices at UCSF, Zuckerberg San Francisco General Hospital, and the San Francisco VA Medical Center. His clinical interests include lung cancer, interstitial lung disease, pulmonary embolism, aortic aneurysm, and trauma imaging. These clinical domains inform his translational research in artificial intelligence, quantitative imaging, and emerging imaging techniques such as 0.55T mid-field lung MRI and privacy-preserving medical imaging data encryption technologies.

His research interests include multimodal imaging database development, automated imaging protocol selection, clinically contextualized AI-assisted image interpretation, intelligent radiology reporting tools, and AI-assisted medical education for residents and fellows.

He serves as Education Co-Chair of the UCSF Center for Intelligent Imaging. His team welcomes collaboration with students, clinicians, scientists, engineers, and postdoctoral scholars through a range of research and training programs.