A decision support tool for early detection of knee OsteoArthritis using X-ray imaging and machine learning: Data from the OsteoArthritis Initiative. (April 2019)

• Record Type: Journal Article
• Authors: Brahim, Abdelbasset; Jennane, Rachid; Riad, Rabia; Janvier, Thomas; Khedher, Laila; Toumi, Hechmi; Lespessailles, Eric
• Is Part Of: Computerized medical imaging and graphics. Volume 73(2019)
• Page Start: 11
• Page End: 18
• Publication Date: 2019-04
• Subjects: Intensity normalization -- Computer aided diagnosis system -- Classification -- OsteoArthritis; Diagnostic imaging -- Periodicals; Imaging systems in medicine -- Periodicals; Diagnosis, Radioscopic -- Data processing -- Periodicals; Diagnostic Imaging -- Periodicals; Imagerie pour le diagnostic -- Périodiques; Diagnostic imaging; Periodicals; Electronic journals; Electronic journals; 616.0754
• DOI: 10.1016/j.compmedimag.2019.01.007
• Languages: English
• ISSNs: 0895-6111

Abstract

Highlights: A novel preprocessing step, which is based on a predictive modeling using Multivariate linear regression is employed to reduce the inter-subject variability. Only pixel intensities are used as features to train the classifiers. The proposed aided diagnosis method is validated on a bigger multi-centric dataset. Compared to state-of-the-art, our approach has a higher classification rate for OA detection. Abstract: This paper presents a fully developed computer aided diagnosis (CAD) system for early knee OsteoArthritis (OA) detection using knee X-ray imaging and machine learning algorithms. The X-ray images are first preprocessed in the Fourier domain using a circular Fourier filter. Then, a novel normalization method based on predictive modeling using multivariate linear regression (MLR) is applied to the data in order to reduce the variability between OA and healthy subjects. At the feature selection/extraction stage, an independent component analysis (ICA) approach is used in order to reduce the dimensionality. Finally, Naive Bayes and random forest classifiers are used for the classification task. This novel image-based approach is applied on 1024 knee X-ray images from the public database OsteoArthritis Initiative (OAI). The results show that the proposed system has a good predictive classification rate for OA detection (82.98% for accuracy, 87.15% for sensitivity and up to 80.65% for specificity).