Artificial Intelligence (AI) is an important product of the rapid development of computer technology today. It has a far-reaching impact on the development of medical diagnostic technology especially in combination with medical imaging. The aim of this study was to analyze the diagnostic accuracy of AI-assisted diagnosis technology for classification of breast cancer in histopathological images.

A meta-analysis was conducted of published research articles on diagnostic accuracy of AI-assisted diagnosis technology for breast cancer classification between January 2010 and September 2019 in the databases of PubMed, EMBASE, Cochrane Library, China National Knowledge Infrastructure, Wanfang Data Knowledge Service Platform and China Bio-medicine Database. Statistical analysis was performed with software Meta-Disc 1.4 and Stata 12.0, and the summary receiver operating characteristic (SROC) curve was drawn to evaluate accuracy of the method.

A total of 18 studies with 13,573 breast histopathological images were considered for the analysis. The pooled sensitivity, specificity, diagnostic odds ratio and the area under the curve of the SROC for AI-assisted diagnosis technology for classification of breast cancer respectively, were 0.94 (95% confidence interval [CI]: 0.93–0.85), 0.84 (95% CI: 0.93–0.94), 255.47 (95% CI: 168.33–387.73) and 0.98 (95%CI: 0.96–0.99).

Several limitations should be considered when interpreting the findings of this meta-analysis. First, despite the extensive literature search, the number of included studies was small; however, the number of images enrolled was satisfactory, thereby decreasing type II error. Second, data acquisition is not comprehensive enough because the language of literature search was limited to Chinese and English. Furthermore, the heterogeneity caused due to different sources of data affected the study results. Despite these limitations, our study suggests AI-assisted diagnosis technology for breast cancer classification in histopathological images is a highly accurate and reliable diagnostic method for clinical application.