Artificial intelligence improves the accuracy of residents in the diagnosis of hip fractures: a multicenter study

Abstract

Background: Less experienced clinicians sometimes make misdiagnosis of hip fractures. We developed computer-aided diagnosis (CAD) system for hip fractures on plain X-rays using a deep learning model trained on a large dataset. In this study, we examined whether the accuracy of the diagnosis of hip fracture of the residents could be improved by using this system.

Methods: A deep convolutional neural network approach was used for machine learning. Pytorch 1.3 and Fast.ai 1.0 were applied as frameworks, and an EfficientNet-B4 model (a pre-trained ImageNet model) was used. We handled the 5295 X-rays from the patients with femoral neck fracture or femoral trochanteric fracture from 2009 to 2019. We excluded cases in which the bilateral hips were not included within an image range, and cases of femoral shaft fracture and periprosthetic fracture. Finally, we included 5242 AP pelvic X-rays from 4851 cases. We divided these 5242 images into two images per image, and prepared 5242 images including fracture site and 5242 images without fracture site. Thus, a total of 10,484 images were used for machine learning. The accuracy, sensitivity, specificity, F-value, and area under the curve (AUC) were assessed. Gradient-weighted class activation mapping (Grad-CAM) was used to conceptualize the basis for the diagnosis of the fracture by the deep learning algorithm. Secondly, we conducted a controlled experiment with clinicians. Thirty-one residents;young doctors within 2 years of graduation from medical school who rotate through various specialties, were tested using 300 hip fracture images that were randomly extracted from the dataset. We evaluated the diagnostic accuracy with and without the use of the CAD system for each of the 300 images.

Results: The accuracy, sensitivity, specificity, F-value, and AUC were 96.1, 95.2, 96.9%, 0.961, and 0.99, respectively, with the correct diagnostic basis generated by Grad-CAM. In the controlled experiment, the diagnostic accuracy of the residents significantly improved when they used the CAD system.

Conclusions: We developed a newly CAD system with a deep learning algorithm from a relatively large dataset from multiple institutions. Our system achieved high diagnostic performance. Our system improved the diagnostic accuracy of residents for hip fractures.

Level of evidence: Level III, Foundational evidence, before-after study.

Clinical relevance: high.

Fig. 1

Patient flow

Fig. 2

ROC Curves. This is the ROC curve for the EfficientNet-B4 model, which showed an AUC of 0.992. Class 0 indicates cases without fracture, and Class 1 indicates cases with fracture. Each ROC curve was calculated. The micro-average ROC sums contributions by class, while the macro-average ROC shows the average results for all classes (AUC = 0.992)

Fig. 3

Images that were misdiagnosed by the CAD system. a-c Incorrectly diagnosed by the CAD system (false-negative). a A case that even orthopedic surgeons could not decide. b A case in which a non-orthopedic surgeon could be wrong. c A case in which even non-orthopedic surgeons were not confused by the diagnosis. d-f Images that were incorrectly diagnosed by the CAD system (false-positive). d Normal image. e A case after implant removal. f A case in which deformity healed after conservative treatment

Fig. 4

Visualization of the area of fracture detection using Grad-CAM. a For images diagnosed by the algorithm as “with fracture”, Grad-CAM showed a high-signal region consistent with the fracture site . b For images diagnosed as “no fracture”, Grad-CAM showed high-signal areas in the region other than femoral neck and trochanteric. From red to green, the diagnostic basis of the CAD system was strongly evident