DOI 10.1109/embc58623.2025.11254878
3D Texture Analysis in Knee MRI for an Early Diagnosis of Osteoarthritis: Identifying Biomarkers of Cartilage Degeneration.
Аннотация
Osteoarthritis (OA) is the most prevalent rheumatic disease, affecting an increasing number of people. This work proposes an innovative methodology based on texture analysis on femorotibial cartilage and T2 mapping on femoropatellar cartilage in T2-weighted magnetic resonance imaging (MRI) to identify early cartilage degeneration biomarkers, imperceptible to the human eye. Two distinct analyses were conducted: (1) On the one hand, in the femorotibial cartilage, 3D manual segmentation was performed and a radiomic approach was applied extracting 43 textural features; (2) On the other hand, in the femoropatellar cartilage through the elaboration of cartigrams, 27 features derived from T2 relaxation times were extracted to evaluate the deep, intermediate and superficial layers of the cartilage. The database included 100 subjects, distributed between controls and patients with different stages of OA according to the Kellgren & Lawrence scale (I-IV). In both analyses the KNN model is positioned as the most effective model followed by the SVM, demonstrating its potential in the early diagnosis of OA using advanced medical imaging techniques and machine learning methods.Clinical Relevance- The diagnosis of osteoarthritis is usually made in advanced stages, at which time treatment options are reduced. MRI makes it possible to detect joint alterations even before there are indications in other radiological tests or clinical symptoms suggesting joint deterioration. This study explores the correlation of MRI findings with clinical features, promoting the integration of advanced technologies into clinical practice to improve the accuracy and speed of OA diagnosis, thus contributing to more effective and personalized interventions.
Предметные рубрики MeSH
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.