Daniel A. Donoho
Papers
2
Total Citations
181
H-Index
2
About
Daniel A. Donoho is a pioneering researcher at the intersection of artificial intelligence and surgical science, whose work is fundamentally reshaping how we understand and evaluate surgical expertise. His primary research areas include computer vision, machine learning interpretability, and surgical data science, with a focus on decoding and assessing intraoperative human performance. Donoho’s most impactful contribution came with his 2023 paper on a vision transformer for decoding surgeon activity from surgical videos (135 citations), which introduced a novel machine learning system capable of analyzing the fine-grained details of surgical actions that directly influence patient outcomes. This work addresses a critical gap in surgical research, where the nuances of intraoperative technique have remained poorly understood. In a second highly-cited study (46 citations), Donoho tackled the pressing issue of algorithmic fairness, demonstrating how human visual explanations can mitigate bias in AI-based assessments of surgeon skills. This is particularly vital as such systems move toward high-stakes applications like credentialing surgeons for operating privileges. Through these contributions, Donoho is establishing himself as a key voice in ensuring that AI-assisted surgical evaluation is both powerful and equitable.
Research Focus
Key Achievements
Top Papers
- 1A vision transformer for decoding surgeon activity from surgical videos135 citations · 2023
- 2