Sultan Mujib Dabiry, Yunus Demirtaş, Fuat Türk, Tuğrul Yıldırım, Gökhan Ayık, Gökhan Çakmak
Joint Diseases and Related Surgery - 2026;37(2):543-551
Objectives: This study aims to evaluate the diagnostic performance of a ResNet50-based convolutional neural network (CNN) in detecting osteochondral lesions of the talus (OLTs) on magnetic resonance imaging (MRI) and to compare its efficacy between T1- and T2- weighted sequences. Materials and methods: A total of 219 ankle MRI scans were reviewed retrospectively, including 60 with confirmed OLTs and 159 without lesions. From each study, coronal and sagittal T1- and T2-weighted images were extracted and standardized to 224 x 224 pixels. Augmentation techniques were applied to strengthen model training. Data were divided into training, validation, and test sets in a 60:20:20 split. A ResNet50 model initialized with ImageNet weights was fine-tuned using cross-entropy loss with class weighting. Diagnostic performance was summarized with accuracy, precision, recall, and F1-scores. Results: The model performed better on T1 sequences, achieving an accuracy of 94.1% and an area under the curve [AUC] of 0.93. T2 sequences were less reliable, showing an accuracy of 87.2% and an AUC of 0.91. Conclusion: Even with a relatively modest dataset, the ResNet50 model delivered strong results for T1-weighted MRI. While T2 images proved more challenging, suggesting that deep learning can add value to routine assessment of OLTs.
