Knee effusion, a common and important indicator of joint diseases such as osteoarthritis, is typically more discernible on magnetic resonance imaging (MRI) scans compared to radiographs. However, the use of radiographs for the early detection of knee effusion remains promising due to their cost-effectiveness and accessibility. This multi-center prospective study collected a total of 1413 radiographs from four hospitals between February 2022 to March 2023, of which 1281 were analyzed after exclusions. To automatically detect knee effusion on radiographs, we utilized a state-of-the-art (SOTA) deep learning-based classification model with a novel preprocessing technique to optimize images for diagnosing knee effusion. The diagnostic performance of the proposed method was significantly higher than that of the baseline model, achieving an area under the receiver operating characteristic curve (AUC) of 0.892, accuracy of 0.803, sensitivity of 0.820, and specificity of 0.785. Moreover, the proposed method significantly outperformed two non-orthopedic physicians. Coupled with an explainable artificial intelligence method for visualization, this approach not only improved diagnostic performance but also interpretability, highlighting areas of effusion. These results demonstrate that the proposed method enables the early and accurate classification of knee effusions on radiographs, thereby reducing healthcare costs and improving patient outcomes through timely interventions.

Stroke causes muscle dysfunction in lower limb depending on the brain damage. Therefore, it is important to identify the degree of muscle damaged and perform rehabilitation training for appropriate time of treatment. However, there is a limitation in that analysis using existing imaging techniques and artificial intelligence cannot analyze disease mechanisms. This paper aims to develop an AI GUI system using SRI to acquire damaged muscle regions, segment them into fiber and space areas, and classify them. For the segmentation, Attention U-Net performed best accuracy 95.32%. For the classification, ResNet50 with Attention U-Net performed best accuracy 99.07%. As a result of this, we designated the best performing network as suitable for stroke animal models. As an auxiliary tool for diagnosing the degree of stroke muscle damage in clinical practice, we constructed a system to analyze the degree of stroke fiber distribution on SRI images using pixel intensity values to show the results. Through this study, it is system that uses deep learning in the stroke animal model can be applied as a basic study for objective muscle tissue evaluation.