Sunday, November 28 | 1:00 p.m.-2:00 p.m. | SSMK03-2 | Room TBA
South Korean researchers will report in this session on a convolutional neural network (CNN) model for automatically diagnosing foot deformity on standing lateral radiographs of the foot. The model appears to work well enough to help in settings with inexperienced readers, according to the group.
A team at the University of Ulsan College of Medicine in Seoul culled a dataset of 1,200 foot lateral radiographs from South Korea's Military Manpower Administration. The x-rays were acquired while patients were standing on a table with their feet in neutral position. The CNN model was trained to identify 25 anatomic foot deformity landmarks per radiograph.
Results suggest readers with less experience had fewer detection errors classifying foot deformities when assisted by the CNN model, the researchers found.
"With the help of AI, inexperienced doctors can identify the anatomic landmarks better than without," according to presenter Dr. Seung Min Ryu, PhD, and colleagues.
Find out the details by attending this session.
