HealthDay News —For fracture detection, artificial intelligence (AI) and clinicians have comparable diagnostic performance, according to a review published in the July issue of Radiology.
Rachel Y. L. Kuo, M.B., B.Chir., from John Radcliffe Hospital in Oxford, England, and colleagues conducted a systematic review and meta-analysis comparing the diagnostic performance in fracture detection between AI and clinicians in peer-reviewed publications and gray literature. A total of 42 studies were included for analysis, with 115 contingency tables extracted from 32 studies (55,061 images). Overall, 37 and five studies identified fractures on radiographs and on computed tomography images, respectively.
The researchers found that for AI and clinicians, the pooled sensitivity was 92 and 91 percent, respectively, and pooled specificity was 91 and 92 percent, respectively, for internal validation test sets. For external validation test sets, for AI and clinicians, the pooled sensitivity was 91 and 94 percent, respectively, and pooled specificity was 91 and 94 percent, respectively. No statistically significant differences were seen between the performance of clinicians and AI. Fifty-two percent of the studies were judged to have a high risk for bias.
“We found that AI performed with a high degree of accuracy, comparable to clinician performance,” Kuo said in a statement. “Importantly, we found this to be the case when AI was validated using independent external datasets, suggesting that the results may be generalizable to the wider population.”