سامانه همایش ها

صفحه اصلی / سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران

Improving Generalization in MRI-Based Deep Learning Models for Total Knee Replacement Prediction

نویسندگان :

Ehsan Karami¹ Hamid Soltanian-Zadeh²

1- School of Electrical and Computer Engineering, College of Engineering, University of Tehran Tehran, Iran 2- School of Electrical and Computer Engineering, College of Engineering, University of Tehran Tehran, Iran

کلمات کلیدی :

knee osteoarthritis،deep learning،medical image analysis،MRI،total knee replacement prediction،model generalization

چکیده :

Knee osteoarthritis (KOA) is a common joint disease that causes pain and mobility issues. While MRI-based deep learning models have demonstrated superior performance in predicting total knee replacement (TKR) and disease progression, their generalizability remains challenging, particularly when applied to imaging data from different sources. In this study, we show that replacing batch normalization with instance normalization, using data augmentation, and applying contrastive loss improves generalization. For training and evaluation, we used MRI data from the Osteoarthritis Initiative (OAI) database, considering sagittal fat-suppressed intermediate-weighted turbo spin-echo (FS-IW-TSE) images as the source domain and sagittal fat-suppressed three-dimensional (3D) dual-echo in steady state (DESS) images as the target domain. The results demonstrated a statistically significant improvement in classification metrics across both domains by replacing batch normalization with instance normalization in the baseline model, generating augmented input views using the Global Intensity Non-linear (GIN) augmentation method, and incorporating a supervised contrastive loss alongside the classification loss to align representations of samples with the same label. In the source domain, this approach achieved an accuracy of 74.12 ± 2.90, an F1 score of 74.57 ± 3.33, and a ROC AUC of 80.65 ± 2.83, outperforming the baseline model, which scored 71.29 ± 4.43, 69.76 ± 4.58, and 77.79 ± 4.66, respectively. In the target domain, the method achieved an accuracy of 70.04 ± 2.49, F1 score of 67.30 ± 3.57, and ROC AUC of 78.12 ± 1.97, compared to the baseline’s 52.87 ± 3.17, 18.98 ± 16.89, and 59.33 ± 6.20. The GIN method with contrastive loss performed better than all evaluated single-source domain generalization methods when using 3D instance normalization. Comparing GIN with and without contrastive loss (for both normalization types) showed that adding contrastive loss consistently led to better performance.