Overview
Kirti Magudia, MD, PhD is an Assistant Professor of Radiology at Duke University School of Medicine. She completed fellowship in abdominal imaging and ultrasound at the University of California, San Francisco and Diagnostic Radiology Residency at Brigham & Women's Hospital. Her research centers on high-level applications of machine learning in radiology, including CT-based body composition analysis and prostate MR, which was facilitated by 7 dedicated months at the MGH/BWH Center for Clinical Data Science and a year long T32 research fellowship in the Biomedical Imaging for Clinical Scientists Program at UCSF. She was the founding resident chair of the Brigham and Women’s Hospital Women in the Radiology Program and has extensively advocated for family-friendly trainee policies. Dr. Magudia is a graduate of the Tri-institutional MD/PhD program of Weill Cornell, Sloan-Kettering, and Rockefeller University, where she completed her Ph.D. in cell and cancer biology in the laboratory of Alan Hall developing a novel 3D cell culture model of colon tumorigenesis.
Current Appointments & Affiliations
Assistant Professor in the Department of Radiology
·
2021 - Present
Radiology, Abdominal Imaging,
Radiology
Recent Publications
The State of Paid Family and Medical Leave Policies: An ACR, AAWR, SWRO Member Survey.
Journal Article J Am Coll Radiol · March 31, 2025 Full text Link to item CiteOpen-Weight Language Models and Retrieval Augmented Generation for Automated Structured Data Extraction from Diagnostic Reports: Assessment of Approaches and Parameters.
Journal Article Radiol Artif Intell · March 12, 2025 "Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note th ... Full text Link to item CiteRSNA 2023 Abdominal Trauma AI Challenge: Review and Outcomes.
Journal Article Radiol Artif Intell · January 2025 Purpose To evaluate the performance of the winning machine learning models from the 2023 RSNA Abdominal Trauma Detection AI Challenge. Materials and Methods The competition was hosted on Kaggle and took place between July 26 and October 15, 2023. The multi ... Full text Link to item CiteRecent Grants
Beyond Race: Examining the Association Between CT-based Body Composition and Socioeconomic Factors to Better Understand Observed Differences in Body Composition by Race.
ResearchPrincipal Investigator · Awarded by Radiological Society of North America · 2024 - 2025Optimizing Preoperative Assessment and Postoperative Management of Bariatric Surgery Patients with Fully Automated, High Throughput and Normalized CT-Based Body Composition Analysis
ResearchPrincipal Investigator · Awarded by Radiological Society of North America · 2022 - 2024Prediction of clinically significant prostate cancer and prostate¿cancer related outcomes from only T2 weighted imaging using machine learning
ResearchPrincipal Investigator · Awarded by Society of Abdominal Radiology · 2021 - 2023View All Grants