The value of S-Detect for the differential diagnosis of breast masses on ultrasound: a systematic review and pooled meta-analysis

Jun Li, Tian Sang, Wen-Hui Yu, Meng Jiang, Shu-Yan Hunag, Chun-Li Cao, Ming Chen, Yu-Wen Cao, Xin-Wu Cui, Christoph F. Dietrich


Aim: To evaluate the value of S-Detect (a computer aided diagnosis system using deep learning) in breast ultrasound (US) for discriminating benign and malignant breast masses.

Material and methods: A literature search was performed and relevant studies using S-Detect for the differential diagnosis of breast masses were selected. The quality of included studies was assessed using a Quality Assessment of Diagnostic Accuracy Studies (QUADAS) questionnaire. Two review authors independently searched the articles and assessed the eligibility of the reports.

Results: A total of ten studies were included in the meta-analysis. The pooled estimates of sensitivity and specificity were 0.82 (95%CI: 0.77-0.87) and 0.86 (95%CI: 0.76-0.92), respectively. In addition, the diagnostic odds ratios, positive likelihood ratio and negative likelihood ratio were 28 (95%CI: 16- 49), 5.7 (95%CI: 3.4-9.5), and 0.21 (95%CI: 0.16-0.27), respectively. Area under the curve was 0.89 (95%CI: 0.86-0.92). No significant publication bias was observed.

Conclusions: S-Detect exhibited a favourable diagnostic value in assisting physicians discriminating benign and malignant breast masses and it can be considered as a useful complement for conventional US.


artificial intelligence; ultrasonography; diagnosis; meta-analysis; breast

Full Text:




  • There are currently no refbacks.