About MedVIC Lab
MedVIC — the Medical Vision and Intelligent Computing Lab — was founded around a single frustration: the gap between what AI can do in benchmark papers and what it does reliably in real medical settings is too wide, and the methods community wasn't taking that gap seriously enough.
We work on methods that close that gap. Probabilistic, geometric, generative — we design AI systems in which structure is treated as a first-class object of modeling rather than an emergent byproduct of scale. In medical imaging, data are complex and heterogeneous, supervision is sparse, and reliability is not a nice-to-have. We believe you can't paper over those challenges with more data or bigger models. You have to address them in how you design representations, how you model uncertainty, and how you build the systems that clinicians and researchers will actually use.
Over more than a decade, this philosophy has produced two things we're proud of in equal measure: methodological advances that the AI research community has cited over 2,500 times, and tools that researchers at over 100 institutions worldwide run on their own data without ever emailing us. ShapeWorks is used by orthopedic surgeons, cardiologists, and craniofacial specialists on six continents. CranioRate is helping surgeons make objective decisions about children with skull deformities — across seven active clinical sites, processing over 2,900 real scans. Our code is on GitHub. It works.
We're also a training ground. Our PhD graduates are at Meta, Amazon, Kitware, and Children's National Hospital. Our MS alumni are at Goldman Sachs, Pfizer, Adobe, Microsoft, and nView Medical. Our postdocs have gone on to tenure-track faculty positions and NIH-funded careers. We hosted the flagship medical imaging AI conference — MIDL 2025 — right here in Salt Lake City. We're still growing.
Institutional Home
MedVIC is housed at the Scientific Computing and Imaging (SCI) Institute — one of the world's leading research centers for scientific computing, visualization, and image analysis — and affiliated with the Kahlert School of Computing at the University of Utah. This dual home gives us something rare: deep roots in both computational foundations and real biomedical applications. SCI's collaborative culture and infrastructure for high-performance computing, data visualization, and clinical partnerships directly enables the scale and ambition of our research program.
Our Culture
The lab has a culture of rigorous thinking and genuine collaboration. We hold weekly lab meetings that are dense with ideas, not status updates. We expect people to read the literature broadly, present honestly, and disagree productively. We write code that runs on real data, not just benchmark datasets. We value students who are self-directed but know when to ask for help.
MedVIC is a place where a PhD student can go from implementing a baseline to co-designing a clinical deployment in the same degree. We don't manufacture urgency — but when something is important, we move fast. Our alumni track record reflects an environment where people grow.
Principal Investigator
Shireen Y. Elhabian, Ph.D.
Associate Professor of Computer Science
Kahlert School of Computing | Scientific Computing and Imaging Institute
University of Utah
Shireen Elhabian's research spans probabilistic machine learning, geometric deep learning, and generative modeling, with a focus on building AI systems for medical image analysis that are not just accurate but reliable, interpretable, and deployable in real clinical environments. She received her Ph.D. in Electrical and Computer Engineering from the University of Louisville in 2012 and joined Utah as a postdoctoral researcher in 2013, rising to her current tenured position in 2022.
She is the PI or co-PI on more than $52M in competitive research funding from NIH, NSF, ARPA-H, the VA, and the DoD, with over $11M under her direct leadership. She serves as a standing member of the NIH Clinical Data Management and Analysis (CDMA) Study Section, General Chair of MIDL 2025, recurring organizer of the ShapeMI workshop at MICCAI, and Associate Editor of the Machine Learning for Biomedical Imaging (MELBA) journal. She has mentored five PhD graduates, five MS thesis students, and five postdoctoral fellows.
📧 shireen@sci.utah.edu | 🏢 Room WEB 3608, 72 South Central Campus Drive, Salt Lake City, UT 84112
Google Scholar ↗ | GitHub ↗ | LinkedIn ↗
Alumni Destinations
Our graduates and postdocs have gone on to positions at:
Industry: Meta · Amazon · Kitware · Adobe · Microsoft · Goldman Sachs · Pfizer · J&J Innovative Medicine Research · nView Medical · Galileo Financial Technologies · Snap Finance
Academia & Medicine: Rhodes College (tenure-track faculty) · University of Utah BME (tenure-track faculty) · Children's National Hospital · One Utah Data Science Hub (Associate Director)
These destinations reflect an environment where people develop real research skills, build real systems, and leave prepared for whatever comes next.