


UPDATE: Thanks to everyone for making this a successful workshop. We greatly appreciate all of the speakers contributing their time to come and give talks in-person in Baltimore. Thank you to Johns Hopkins and the Department of Biomedical Engineering for making this workshop possible. And thank you to the over 500 people that registered including those that came and visited us in Baltimore as well as the virtual attendees from all over the world – representing 35 countries and over 200 different institutions (universities, companies, hospitals, etc.)

Introduction
The Biomedical Engineering Department at Johns Hopkins University was pleased to host the 2025 Deep Reconstruction Workshop. This workshop provided a multi-disciplinary forum to exchange ideas of image reconstruction using deep learning methods for major medical imaging modalities, including CT, PET, SPECT, MRI, Ultrasound and optical imaging. Presentations were given by invitation to summarize state-of-the-art and brainstorm future directions.
The workshop was free for both in-person and online attendance.
Previous Meetings
- Deep Reconstruction Workshop 2017 (Rensselaer Polytechnic Institute: Ge Wang, Hongming Shan)
- Deep Reconstruction Workshop 2020 (GE Research, Bruno De Man, Amanda Youmans)
- Deep Reconstruction Workshop 2021 (Massachusetts General Hospital, Quanzheng Li)
- Deep Reconstruction Workshop 2023 (Yale University, Chi Liu) Online presentations
Agenda
The following was the scientific program (All times are Eastern US time zone):
Recordings of the talks are available on YouTube: Playlist with all recordings
Saturday March 22, 2025
12:15-12:30pm | Welcome and Opening Remarks |
12:30-1:00pm | “Putting medical imaging in context: Deep reconstruction and connected scanners of the future,” Daniel Sodickson, Grossman School of Medicine – Radiology (New York University) VIDEO |
1:00-1:30pm | “Provable probabilistic imaging using score-based diffusion models,” Yu Sun, Sun Lab (Johns Hopkins University) VIDEO |
1:30-2:00pm | “Medical multimodal multitask foundation model and LLM for lung cancer screening,” Chuang Niu, Wang-AXIS Lab (Rensselaer Polytechnic Institute) VIDEO |
2:00-2:30pm | “Intelligent automatic treatment planning in radiotherapy,” Xun Jia, Jia Lab (Johns Hopkins University) VIDEO |
2:30-3:00pm | BREAK |
3:00-3:30pm | “MR Spatiospectral Reconstruction Integrating Physics-Driven Subspace and Deep Learning,” Fan Lam, Quantitative Multiscale Imaging Group (University of Illinois – Urbana-Champaign) VIDEO |
3:30-4:00pm | “Computational imaging: Restoration deep networks as implicit priors,” Ulugbek Kamilov, Computational Imaging Group (Washington University) VIDEO |
4:00-4:30pm | “SELFIE – Self-supervised learning for fast dynamic golden-angle radial MRI,” Ricardo Otazo, Otazo Lab (Memorial Sloan Kettering Cancer Center) VIDEO |
4:30-5:00pm | “Advancing photoacoustic computed tomography via generalizable deep learning methods,” Mark Anastasio, Computational Imaging Science Laboratory (University of Illinois – Urbana-Champaign) VIDEO |
5:00-5:30pm | “Solving hard tomography problems with diffusion posterior sampling,” Web Stayman, AIAI Lab (Johns Hopkins University) VIDEO |
Sunday March 23, 2025
8:00-8:30am | “Diffusion Models for Large-Scale Imaging Reconstruction Problems,” Jason Hu, Jeff Fessler Lab (University of Michigan) VIDEO |
8:30-9:00am | “Advancing computational wave imaging through deep learning and wave physics integration,” Youzuo Lin, SMILE Lab (University of North Carolina at Chapel Hill) VIDEO |
9:00-9:30am | “Communicating uncertainty in deep learning reconstructions to radiologists,” Grace Gang, IMPACT Laboratory (University of Pennsylvania) VIDEO |
9:30-10:00am | “Imaging and modeling large-scale brain dynamics,” Adam Charles, The Neural Signals & Computation Laboratory (Johns Hopkins University) VIDEO |
10:00-10:30am | BREAK |
10:30-11:00am | “Deep learning reconstruction and processing for PET and SPECT,” Chi Liu, Radiology and Biomedical Imaging (Yale University) VIDEO |
11:00-11:30am | “Uncertainty quantification with conformal guarantees for inverse problems in CT,” Jacopo Tennegi, Sulam Group (Johns Hopkins University) VIDEO |
11:30-12:00pm | “Multi-scale energy (MuSE) models for imaging with guarantees,” Mathews Jacob, Computational Biomedical Imaging Group (University of Virginia) VIDEO |
12:00-12:30pm | “Image generation under intervention,” Quanzheng Li, Center for Advanced Medical Computing and Analysis (Massachusetts General Hospital) VIDEO |
12:30-1:00pm | “Ultrasound and photoacoustic beamforming in the deep learning age,” Muyinatu Bell, PULSE Laboratory (Johns Hopkins University) VIDEO |
1:00-1:15pm | Closing Remarks |