About the Session
Comparative Evaluation of Open-source Large Language Models for Automated Extraction of Iliofemoral DVT Data from Venous Duplex Ultrasound Reports
Krishnaveni Parvataneni, Undergraduate Researcher, Massachusetts General Hospital
Felix Dorfner; Christopher Bridge, DPhil; Dania Daye, MD, PhD
Extra Thoughts, Extra Watts? Energy-Accuracy Trade-Offs of Reasoning LLMs for Radiology Report Labeling
Dharmam Savani, MS, Software Engineer, St. Jude Children’s Research Hospital
Aditya Kulkarni, MS; Briana Malik; Vidith Phillips, MD, MS; Ammar Ahmed Pallikonda Latheef, MS; Sayantan Bhadra, PhD; Pritam Mukherjee, PhD; Paul Yi, MD
Large Language Models in Radiologic Numerical Tasks: A Thorough Evaluation and Error Analysis
Ali Nowroozi, MD, Postdoctoral Research Fellow, University of California, San Francisco
Masha Bondarenko; Adrian Serapio; Tician Schnitzler, MD; Sukhmanjit Brar; Jae Ho Sohn, MD, MS
Optimizing AI-generated Radiology Impressions: Stakeholder Preferences Along the Brevity-Completeness Tradeoff
Andrew Del Gaizo, MD, Abdominal Radiologist, Moffitt Cancer Center
Sharang Phadke; Nivedita Suresh, MSE
Systematic Vulnerability of Large Language Models to Adversarial Inputs in Radiology Impression Generation
Briana Malik, AI Researcher, St. Jude Children’s Research Hospital
Pritam Mukherjee, PhD; Paul Yi, MD
Zero-shot and Few-shot Large Language Model Performance in Pediatric Upper Extremity Fractures Radiographic Report Classification
Gustavo Lischinsky Moreno, MD, Radiology Research Scholar, Children’s Hospital of Philadelphia
Vahid Khalkhali, PhD, MSc; Mario Sinti-Ycochea, MD; Michael Welsh; Susan Sotardi, MD, MS
Presented By
