Deep learning has been increasingly used in image reconstruction, artifact correction, and noise reduction in medical image formation. The objective of this workshop is to update the SPIE Physics of Medical Imaging community on the latest development of deep learning applications in medical imaging formation. At this workshop, experts in the field will provide an overview of the current status and future directions in this area.

Welcome and Introduction

Session Chairs: Lifeng Yu, Mayo Clinic (United States) and Rebecca Fahrig, Siemens Healthineers (Germany)

Part I: Academic and regulatory updates

---

Deep learning in CT reconstruction and artifact correction
Guang-Hong Chen, Univ. of Wisconsin at Madison (United States)

Deep learning reconstruction with known operators with applications in MRI, CT, and hybrid systems
Andreas Maier, Univ. of Erlangen-Nuremberg (Germany)

Uncertainties of deep learning in image noise reduction
Vegard Antun, Univ. of Oslo (Norway)

Part II: Industry updates

---

Deep learning reconstruction from denoising to super resolution
Kirsten Boedeker, Canon Medical Systems (United States)

Deep learning in CT image formation
Brian Nett, GE Healthcare (United States)

AI in image formation
Christian Wülker, Philips Healthcare (Germany)

Improving MRI workflow using Deep Resolve
Fei Han, Siemens Medical Solutions (United States)

Vendor-agnostic DL-based CT denoising clarity engine: an internal quality improvement procedure
Jonghyo Kim, ClariPi (Republic of Korea) and Seoul National Univ. (Republic of Korea)

Part III: Regulatory perspective

---

Generalizability performance of deep learning-based image reconstruction and denoising methods in CT
Rongping Zeng, U.S. Food and Drug Administration (United States)

Part IV: Panel Discussion on Deep Learning in Medical Image Formation

---

Enjoy a discussion on the topics with a panel assembled from the speakers, moderated by the session chairs.

---

This special event is part of the Physics of Medical Imaging conference.