Fig. 4From: Enhancing prediction of supraspinatus/infraspinatus tendon complex injuries through integration of deep visual features and clinical information: a multicenter two-round assessment studyComparison of the performance of the ensemble CML-DL model with that of physicians. The figure displays the identification of SRCT and NIRCA in the training cohort, internal validation cohort, and external validation cohort using the ensemble CML-DL model and by individual physicians. The performance of our ensemble CML-DL model is compared with each of the five readers and the average readerBack to article page