Diagnostic accuracy of deep learning in orthopaedic fractures: a systematic review and meta-analysis. Issue 9 (September 2020)

• Record Type: Journal Article
• Authors: Yang, S.; Yin, B.; Cao, W.; Feng, C.; Fan, G.; He, S.
• Is Part Of: Clinical radiology. Volume 75:Issue 9(2020)
• Page Start: 713.e17
• Page End: 713.e28
• Publication Date: 2020-09
• Subjects: Medical radiology -- Periodicals; Radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiology -- Periodicals; Societies, Medical -- Periodicals; Medical radiology; Radiotherapy; Electronic journals; Periodicals; 616.0757
• DOI: 10.1016/j.crad.2020.05.021
• Languages: English
• ISSNs: 0009-9260

Abstract

Abstract : Aim: To gather and compare related clinical studies, and to investigate the accuracy and reliability of deep learning in detecting orthopaedic fractures. Materials and methods: This study is a retrospective combination and interpretation of prospectively acquired data. Articles from PubMed, EMBASE, the Cochrane library databases, and reference lists of the qualified articles were retrieved. Heterogeneity between studies was assessed using a random effective model. Pooled sensitivity, specificity, diagnostic odds ratio, and area under the receiver operating characteristic curve (AUC) were obtained by a random model. This work was managed from October 2018 to March 2020. Results: Fourteen studies were included in this systematic review and nine were synthesized in the meta-analysis. The pooled sensitivity and specificity for the whole group (17 trials, 5, 434 images) were 0.87 and 0.91, respectively. The AUC was 0.95. Eight trials (1, 574 images) were included in the long-bone group, which contained seven studies. The pooled sensitivity was 0.96 and specificity was 0.94. The AUC was 0.99. Heterogeneity existed in the four pooled results of the whole group and the pooled specificity of the long-bone group. Conclusions: Deep learning is reliable in fracture diagnosis and has high diagnostic accuracy, which is similar to that of general physicians and is unlikely to produce a large number of false diagnoses; however, the ability of deep learning to localize the fracture needs more attention and testing. Deep learning can be extremely helpful with pre-classification of clinical diagnoses. Highlights: The accuracy and reliability of deep learning in detecting orthopedic fractures. Pool sensitivity, specificity and AUC of whole group are 0.87, 0.91 and 0.95. Pool sensitivity, specificity and AUC of long bone group are 0.96, 0.94 and 0.99. Heterogeneity exists in the four pooled results of the whole group. Heterogeneity exists in the pooled specificity of the long bone group.