Machine-learning model detects, characterizes rib fractures on CT

Tuesday, November 28 | 9:30 a.m.-9:40 a.m. | T3-SSER01-1 | Room E351

A machine-learning model can detect traumatic rib fractures on CT scans and automatically provide a fracture severity score, according to this scientific presentation.

A group led by presenter Suvrankar Datta, MD, of the All India Institute of Medical Sciences (AIIMS) developed what it calls the AIRib Framework, a model that segments rib fractures and automatically generates a radiology report containing a RibScore radiographic score for fracture severity.

In testing, the ResNet algorithm was able to classify rib fractures at an area under the curve of 0.887, a sensitivity of 81.3%, and an F1 score of 0.793.

“The AIRib Framework has significant management implications for patients, especially for triaging and prognostication, with CT scans having higher RibScore being prioritized for viewing and reporting in the worklist,” the authors wrote. “By incorporating fracture details in a preliminary report, it can assist the trauma radiologist in decreasing the rate of missed rib fractures and significantly reduce their reporting time.”

How did they achieve these results? Stop by this Tuesday morning session to get all of the details.

Page 1 of 2
Next Page