Exploring the Efficacy of U-Net-CNN Architecture for Automated Detection of Psoriatic Arthritis from Medical Imaging Data

Psoriatic arthritis (PsA) is a chronic inflammatory joint disease that often exists with psoriasis, causing substantial morbidity and disability if left untreated. Timely intervention and better patient outcomes depend on the early and precise identification of PsA. This work proposes an innovative method for PsA detection from medical imaging data that uses the U-Net-CNN architecture. Precise localization and characterization of arthritis-associated features are made possible by the U-Net-CNN model, which combines the U-Net architecture for precise segmentation with convolutional neural networks (CNNs) for feature extraction and classification. Using a dataset of medical images from Radiopaedia, the effectiveness of the proposed method is evaluated. It shows that, compared with traditional architectures, it achieves a high accuracy of $\mathbf{0. 9 7 4 6}$, precision of 0.9335, recall of 0.9035, Dice coefficient of $\mathbf{0. 8 9 0 5}$, and Jaccard Index of 0.9092.