Postdoc position in ML for surgical AI/computer vision

[Guy Rosman]

Please excuse cross-posting.

Mass. General Hospital’s Surgical Artificial Intelligence and Innovation Laboratory, in collaboration with MIT/CSAIL's Distributed Robotics Lab, is looking for a smart and energetic machine learning/computer vision postdoctoral candidate. Our project is set to revolutionize how surgery works via collaborative surgical AI and surgical video understanding. Candidates should have good knowledge of probability and inference, machine learning, and some experience in computer vision and deep learning frameworks. They should have publications in relevant venues, and be eager to explore how computational methods can improve patient care in the surgical field.

• The project includes both computationally novel aspects, as well as clinical ones. Computational elements include modeling/ML and computer vision aspects aimed at revolutionizing the way we understand surgeries and assist surgeons

• The project is a two-year project with the possibility for extension and has a strong research element as well as possibilities for a commercialization track

Candidates will have the opportunity to work with novel datasets that combine streaming surgical data with datasets of clinical outcomes and procedural quality. The project aims to significantly redefine the delivery and quality of surgical care in both resource-rich and resource-poor settings. 

Candidates should contact either professor Guy Rosman <ros...@csail.mit.edu> (MGH), in collaboration with professors Daniela Rus (MIT) and Ozanan Meireles (MGH). Please attach a CV, as well as a brief (few sentences) statement of interest. 

Skills:

• Ph.D. in an EE/CS field that leverages probabilistic reasoning and machine learning techniques. Publications in either probabilistic inference, computer vision, and relevant ML fields such as knowledge representation and reinforcement learning

• Some experience with computer vision / robotic perception, healthy ability to explore data and ask questions, working knowledge with deep learning frameworks such as Pytorch, no medical knowledge is necessary