GUANGYI ZHANG

Hello! I am currently a research fellow at Massachusetts General Hospital and Harvard Medical School. My current research focuses on ultrasound foundation models, multimodal learning, and clinician-guided AI to predict liver disease progression.

I am a co-investigator on an NIH R01-funded project led by Prof. Anthony Samir, focused on improving fibrosis assessment in metabolic dysfunction-associated steatotic liver disease (MASLD). My first-author research on precision-constrained evaluation of noninvasive biomarkers was published in Nature Communications and led to a provisional patent application. I also led the deep-learning development of PWI-Net for speed-of-sound reconstruction from ultrasound radio-frequency data, with applications in fatty liver assessment and image-quality improvement. Our team placed second in the IEEE IUS 2026 QUASOS Challenge.

As principal investigator, I have secured the MGB Clinical Research Fellowship (2026-2027), and research computing support from Google Cloud, AMD, and the MIT-Anthropic AI for Science Program.