Artificial intelligence can detect hip fractures accurately on pelvic x rays
Singapore: In a research paper published in iScience journal, a deep-learning model was developed to predict hip fractures from pelvic X-rays, even when metallic implants are present.
Yen Yan, Changi General Hospital, and his colleagues from the Duke-NUS Medical School in Singapore developed an AI algorithm (artificial intelligent) using over fortythousand radiographs of the pelvis taken by a single institution. When applied to a set of radiographs from an emergency department (ED), the model showed high specificity, sensitivity and accuracy.
This study approximates a real-world application of a fracture detection model based on deep learning by including radiographs that have suboptimal image qualities, nonhip fractures, and meta llic implant, which had been excluded in previous published work. The team of researchers also investigated the impact of ethnicity on the model’s performance and the accuracy in fracture localization algorithm.
Source:
https://medicaldialogues.in/radiology/news/artificial-intelligence-can-accurately-detect-hip-fractures-on-pelvic-x-rays-114452
