Jonathan Tamir
University of Texas at Austin, TX·Engineering·Early-career
Imaging devices such as microscopes and medical scanners often look excellent in controlled lab demonstrations. Their performance then drops in the real world, where objects move, conditions change and hardware is imperfect.
- Fields
Engineering
Biomedical Engineering
Computer Science
- Topics
computational imaging
medical imaging
microscopy
- Grants
- 1 known grant, through Aug 2028
What they work on
- Develop deep generative physical models that pair known imaging physics with image patterns learned from data, so computational imaging systems can adapt to dynamically changing operating conditions.
- Create reconstruction methods that account for motion during image acquisition, enabling reliable imaging of moving objects and shorter scan times in clinical and research settings.
- Design methods that learn directly from noisy, sub-sampled measurements instead of clean training images, with principled theoretical performance guarantees.
- Improve resilience of imaging systems to hardware imperfections by jointly designing data acquisition and processing using signal processing, optimization and machine learning.
More about the research
This project asks how to design imaging systems that stay reliable outside the lab. The approach is to design how data are collected and how they are processed together, as one system, rather than separately.
Why it matters
More robust imaging could make medical and scientific scanning faster and more reliable. Shorter acquisition times, sharp images of moving objects and compensation for hardware flaws would benefit clinicians and patients in medical imaging, microscopy users and automotive imaging. The methods could also guide the use of deep learning in non-imaging systems such as wireless communications.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Mohammed Qazi
Tuskegee University, AL·Mathematics & Statistics
Develop the mathematics of diffusion models for generative AI, which create new data by reversing a noising process, using stochastic modeling with applications such as wireless communications.
Roummel Marcia
University of California-Merced, CA·Mathematics & Statistics
Extend deep unrolling networks for signal recovery and image reconstruction by training them with quasi-Newton optimization methods that exploit previous iterates and gradients.
Spencer L. Bowen
University of Texas Southwestern Medical Center, TX·Biomedical Engineering
Build a device that rapidly and safely moves an external PET source into and out of a PET/MR scanner, so that signal attenuation can be measured directly during patient heart imaging.
The school
University of Texas at Austin: 49 labs on PI Pages, $1.1B a year on research, 871 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.