Conference on Artificial Intelligence in Medical Imaging

A Schema-Constrained Hybrid Regex–Large Language Model Pipeline for Reliable Structured Extraction of Interventional Radiology Reports (opens in new tab) 

• Seema Singh Saharan, PhD, UCSF 
• Jhaydine Bandola; Rachel Johansson, MD; Pete Lokken, MD, MPH; Beck Olson; Rajesh Shah, MD; Alexandra So; Harsh Tripathi; Xiao Wu, MD 

 

Large Language Model-Based Automated Detection and Turnaround Time Monitoring of Critical Results in a Pediatric Radiology Workflow (opens in new tab) 

• Pooja Pathak, MS, Lucile Packard Children's Hospital Stanford 
• Christina Vernazza; Linn Dee Doyle; Zachery Seow, MBA, CRA, R.T.(R)(CT)(MR)(ARRT); Paul Fong; Timothy Mak, MS; Richard Barth, MD; Ali Syed, MD; Shreyas Vasanawala, MD, PhD; Sergios Gatidis, MD 

 

Personalizing Radiology Impression Generation to Radiologist Style and Subspecialty with Auditable, Auto-Generated Playbooks(opens in new tab) 

• Ushanandini Raghavan, PhD, DeepHealth 
• Madhu Jahagirdar 

 

Residents as a Proxy for AI Draft Reporting: LLM-based Analysis(opens in new tab) 

• Steven Rothenberg, MD, Thomas Jefferson University 
• Shahir Monsuruddin, MD; Paras Lakhani, MD; Trevor Vent, PhD; Baskaran Sundaram, MD; Adam Flanders, MD, FSIIM 

 

SpineGraph: Vertebral-Level-Aware Evaluation for Automated Cervical Spine Report Generation (opens in new tab)

• Sydney A. Levin, Tulane University 
• Matheus Ferreira; Eduardo Farina, MD; Ronald Peshock, MD; Paulo Kuriki, MD 

 

What Automated Radiology-Report Metrics Can and Cannot Detect: A Controlled Simulation Study of Single-Word Clinical Changes (opens in new tab)

• Mohammadreza Chavoshi, MD; Emory University School of Medicine 
• Bardia Khosravi, MD, MPH, MHPE; Frank Li, PhD; Theo Dapamede, MD, PhD; Janice Newsome, MD; Hari Trivedi, MD; Judy Gichoya, MD, MS, FSIIM