Welcome to the CMIT Lab!
Collaborate Innovate Translate
Mission: We focus on developing, evaluating, and translating computational medical imaging methods for optimized performance in diagnostic and therapeutic tasks. To develop these methods, we innovate in the fields of imaging science, artificial intelligence, and statistical signal processing, developing methods for image reconstruction, enhancement, segmentation, quantification, and image-quality evaluation.
Core theme: In biomedical imaging, images are acquired for particular tasks such as detection or estimation. The core theme of our lab is to design imaging methods such that they yield optimized performance in these tasks. This spurs innovation in myriad ways, such as reconstruction-less quantification for ultra-low-count therapeutic imaging, clinical task-specific AI algorithm design for reducing radiation dose/acquisition time in diagnostic imaging, physics-guided deep-learning to reduce the requirement of training data, estimation-based segmentation methods that integrate deep learning to delineate structures in the brain that are even difficult to see, and no-gold-standard evaluation to evaluate quantitative imaging methods without ground truth!
Check out our lab youtube channel for video presentations of our work.
The CMIT lab was set up as part of the Imaging Sciences Initiative at Washington University with support from the Department of Biomedical Engineering at the McKelvey School of Engineering and the Mallinckrodt Institute of Radiology within the School of Medicine.
Open Positions: We have multiple NIH-funded open positions in the lab for graduate students and post-doctoral fellows. Please see Open Positions page.
We acknowledge financial support from NIBIB (R01 awards, Trailblazer award, R56 award), NSF (CAREER award),NIH Office of Data Science Strategy, NVIDIA, SNMMI, WMIS, and other sources.