News

UCSF Imaging Symposium 2026

Members of the Sohn Lab presented various works on topics ranging from post-deployment reliability frameworks to RV dysfunction to subsolid nodule growth modeling.UCSF 23rd Annual Imaging Research Symposium

Multiple Papers Accepted to NeurIPS and COLM!

We are excited to announce that six research contributions from the Sohn Lab have been accepted for presentation at workshops at the 2026 Conference on Neural Information Processing Systems (NeurIPS) and the 2026 Conference on Language Modeling (COLM)!

Large Language Models Improve Clinical History Summarization for Radiology

A new study from our lab, published in Radiology, evaluated whether large language models can transform electronic health record notes into concise, clinically useful indications for diagnostic imaging. The project was led by Adrian Serapio, BS, and Timothy Chen, MD, combining expertise in machine…

Meet Our Summer Students and New Lab Members!

This summer, our lab is pleased to welcome a talented group of new and returning students and team members. After contributing to the lab throughout the academic year, UC Berkeley URAP students Justin Kim and Brian Kunzang received Rose Hills Foundation scholarships to continue their research with…

AI-Generated Summaries Improve Patient Understanding of Radiology Reports

New research from the Sohn Lab, recently featured by RSNA News, found that large language model-generated, patient-friendly summaries of lung cancer screening CT reports improved comprehension, reduced anxiety, and increased participants’ willingness to wait for a scheduled follow-up appointment.

Rose Hills Summer Research Scholarship Awarded to Sohn Lab Undergraduate Researchers

This summer, our lab is pleased to announce undergraduate researchers Brian Kunzang and Justin Kim, both University of California, Berkeley students and members of our lab through the Undergraduate Research Apprentice Program (URAP), as recipients of the Rose Hills Scholarship.

Using LLMs to Improve Understanding of Cardiac MRI Image Quality

A recent publication from our lab in the Journal of Magnetic Resonance Imaging highlights how large language models can be used to improve understanding of cardiac MRI image quality before scans are even performed.

Advancing Patient-Centered Radiology Reporting with LLMs

A recent publication from our lab represents an important contribution to patient-centered radiology AI and highlights a central direction of the lab.

Demonstrating the Reliability of Reasoning-Capable Large Language Models in Radiologic Numerical Tasks

In our latest open-access study, “Large Language Models in Radiologic Numerical Tasks: A Thorough Evaluation and Error Analysis,” the Sohn Lab demonstrates that modern reasoning-capable large language models can reliably perform clinically meaningful numerical tasks directly from radiology reports.

Newly Accepted Study: Understanding Patient and Physiologic Factors Influencing Image Quality in 0.55T Lung MRI

We are pleased to share that a recent study from the Sohn Lab has been accepted for publication in Radiology Advances. This work reflects our lab’s ongoing interest in better understanding how emerging MRI technologies perform in real clinical settings. As mid field MRI at 0.55T gains attention for…