Programming of deep learning software with sufficient diagnostic accuracy can only be done with several tens of thousands of radiographs, or by using transfer learning. Our results are not good enough for our algorithm to be used in current clinical practice. The specificity was 67% and 69%, the positive predictive value was 55% and 56%, while the negative predictive value was 74% and 78%, respectively. The sensitivity of the algorithm was 61% for femoral neck fractures and 67% for trochanteric fractures. In all, 9425 annotated radiographs (90%) were used for the training phase and 1074 (10%) for the test phase. The deep learning algorithm was programmed on Tensorflow™ software (Google Brain, Santa Clara, Ca, USA, ). Fractures were classified according to the AO system. Each radiograph was evaluated by an orthopedic surgeon using RectLabel™ software (), by differentiating between healthy and fractured zones. The sample size was increased 8-fold (resulting in 10,472 radiographs) using a validated technique. Our database consisted of 1309 radiographs: 963 had a PFF, while 346 did not. It is possible to program an effective deep learning software to help make the diagnosis of PFF based on a limited number of radiographs. This led us to conduct a prospective study with the goals of: 1) programming deep learning software to help make the diagnosis of PFF on radiographs and 2) to evaluate its performance. Thus, an effective, low-cost software that helps with making radiographic diagnoses would be a relevant tool for current clinical practice, although this concept has rarely been evaluated in orthopedics for proximal femur fractures (PFF). Using AI as a radiological diagnostic aid may be beneficial in the emergency room. Diagnostic errors in traumatology are rare but can have serious functional consequences. Radiology is one of the domains where artificial intelligence (AI) yields encouraging results, with diagnostic accuracy that approaches that of experienced radiologists and physicians.
0 kommentar(er)
