James S Duncan
Yale University, CT·Biomedical Engineering
Medical images hold far more information than a doctor's eye can pull out. The research builds machine learning and image analysis methods that turn brain and body scans into numbers that measure disease and predict how a patient will respond to treatment.
- Fields
Biomedical Engineering
Computer Science
Medicine & Clinical Research
- Topics
medical image analysis
machine learning
deep learning
- Grants
- 2 known grants, through Aug 2027
What they work on
- Develop image-analysis methods that learn jointly from repeated MRI and CT scans of liver cancer patients. The goal is to track the tumor's surroundings during combined chemoembolization and immunotherapy and to predict outcomes.
- Build a higher-resolution version of BIRDS, a magnetic resonance method that measures tissue acidity, combined with machine learning. It will map acidity changes around tumors after combined therapy in a rabbit VX2 liver cancer model.
- Combine long short-term memory networks, which capture time patterns, with graph neural networks, which capture brain connectivity, into one model of fMRI data. The model will predict how children with autism respond to Pivotal Response Treatment.
- Expand autism fMRI analysis to children with co-occurring anxiety, ADHD or depression. Domain adaptation will combine different scan types, and federated learning will pool data across institutions privately.
More about the research
The two projects apply this idea to two very different problems. One is liver cancer treated with combined therapies. The other is autism, tracked with brain scans before and during behavioral therapy.
Why it matters
For people with liver cancer, imaging markers that show early whether combined catheter therapy and immunotherapy is working could help doctors choose and adjust treatment for each patient. For children with autism, brain-scan markers that predict response to behavioral therapy could help guide personalized treatment. They could also clarify how conditions like anxiety and ADHD change the picture.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Kevin Staveley-O'Carroll
University of Connecticut, CT·Microbiology & Immunology
Dissect how Bacteroides thetaiotaomicron changes dendritic cells and macrophages through the KLF2 and TLR9 pathways to strengthen anti-HCC immunity, using a mouse liver cancer model with a trackable tumor antigen.
Rebecca Schwarzlose
Washington University in St Louis, MO·Neuroscience·Early-career
Test whether neural suppression, the reduced brain response to expected sensations, is weaker in children with sensory over-responsivity, using existing brain imaging data.
Timothy Ziemlewicz
University of Wisconsin-Madison, WI·Biomedical Engineering
Compare strategies for limiting breathing-related liver motion during histotripsy, testing high-frequency jet ventilation against cone-beam CT-based motion modeling that adjusts the treatment prescription.
The school
Yale University: 73 labs on PI Pages, $1.5B a year on research, 438 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.