Data-Efficient Language Model for Assessing Pulmonary Embolism Diagnostic Certainty From Radiology Reports: Model Development and Validation Study

Feifan Liu   ^{1 *} , PhD ;   Ruofan Hu   ^{2 *} , MS ;   Donghyuk Kim   ^{1, 2} , MS ;   Hao Lo   ¹ , MD, MBA ;   Yevgeniy Kharonov   ³ , DO ;   Ansh Johri   ¹ , MD ;   Ben S Gerber   ¹ , MPH, MD ;   Elke Rundensteiner   ^{1, 2} , PhD ;   Lauren M Westafer   ⁴ , DO ;   Jeroan Allison   ¹ , MD ;   Catarina Kiefe   ¹ , MD, PhD ;   Alexander Bankier   ¹ , MD, PhD ;   Max P Rosen   ¹ , MPH, MD

¹ Chan Medical School, University of Massachusetts, Worcester, MA, United States

² Data Science, Worcester Polytechnic Institute, Worcester, MA, United States

³ Department of Radiology, University of Miami Miller School of Medicine, Miami, FL, United States

⁴ Chan Medical School, University of Massachusetts, Baystate, MA, United States

*these authors contributed equally