Advancing Medicine through Machine Learning

Independent Research at the Edge of Medicine

Advancing Medicine through Machine Learning

Independent Research at the Edge of Medicine

About The Tensor Lab

The Tensor Lab for Computational Medicine is launching a research fellowship for talented computer science students who want to work on challenging problems with real stakes. Over the summer, fellows will develop independent machine learning projects aimed at solving active, unsolved challenges in medicine. These are gaps identified by physicians working at the leading edge of their fields.

Our faculty mentors from institutions like UCSF School of Medicine, The National Institutes of Health, and UMD School of Medicine present the biggest open questions they face in practice: diagnostic bottlenecks, high-risk decision points, blind spots in data, failure modes in triage, pain points in radiology, surgery, oncology, and beyond. Fellows listen to these clinical challenges, choose the ones they find most compelling, and begin developing innovative ML-based solutions with close mentorship from medical students and regular input from physician-scientists.

This is a remote, project-based unpaid fellowship to train future innovators to design and lead high-impact research, from idea to execution. And they'll be in a cohort of equally driven, curious, and talented students to learn from. Successful projects may lead to academic publication and even influence how care is delivered to real patients.

The Process

Gap Analysis and Literature Review

The Tensor Lab Summer Fellowship runs from June to August and is built around focused, independent research. At the start of the program, physician-scientists from institutions like UCSF, NIH, and UMD present the most pressing open questions from their fields: real clinical gaps in areas like radiology, surgery, cardiology, oncology, and more. Fellows listen to these presentations, select the challenge that resonates most, and begin designing a machine learning–based approach. In the first two weeks, each fellow conducts a targeted literature review, defines a tractable research question, and develops a concrete plan for execution with feedback from a medical student mentor and a clinical advisor.

Project Implementation

Once a project direction is approved, fellows move into focused development. They clean and explore real datasets, build models, and analyze performance using clinical context as a guide. Each fellow meets weekly with their medical student mentor to troubleshoot, refine ideas, and maintain momentum. Every two weeks, they check in with a faculty mentor for higher-level input. Fellows also participate in weekly full-cohort meetings, where they present progress, test their thinking in front of their peers, and offer feedback on other projects. The optional opportunity to shadow a physician in the specialty they are studying allows fellows to better understand the clinical environment their work addresses.

Communication of Results

At the end of the program, each fellow presents their work at a final symposium attended by faculty, researchers, and invited guests from industry. The most promising projects may continue into the academic year with mentorship for submission as a paper, abstract, or poster. Throughout the program, the emphasis is on ownership, rigor, and impact: each student leads their own project, supported by a sharp, ambitious cohort and experienced mentors, working on real clinical questions with the potential to inform how care is delivered.

Advancing Medicine through Machine Learning

Advancing Medicine through Machine Learning

About The Tensor Lab

The Process

Gap Analysis and Literature Review

Project Implementation

Communication of Results

This website uses cookies.