Journals / International Scientific and Vocational Studies Journal / 2019 / Cilt: 3 - Sayı: 2
Automated Diagnosis of Meniscus Tears from MRI of the Knee
- Pages
- 92–104
- DOI
- —
Abstract
Meniscus tears are serious knee abnormalities that can cause knee osteoarthritis disorder. Therefore, early detection and treatment of meniscus tears that may occur in the knee with computer-aided systems will prevent the progression of these disorders. In this study, an approach which can detect the meniscus tears automatically by using and comparing two different feature extraction methods have been presented. With these methods, features of the knee MR images were obtained and automatic meniscus tear classification was performed by such features. Four different classifiers have been used to model the features in the classification phase. The most successful classification results were obtained from the support vector machines (SVM) with a success rate of 90.13% and the extreme learning machines (ELM) with a success rate of 87.85% via the LBP feature extraction method. It is observed that better results are obtained than the ones in similar studies in the literature. It is aimed to improve the existing success with the use of deep feature extraction methods in the future.
Özet
Meniscus tears are serious knee abnormalities that can cause knee osteoarthritis disorder. Therefore, early detection and treatment of meniscus tears that may occur in the knee with computer-aided systems will prevent the progression of these disorders. In this study, an approach which can detect the meniscus tears automatically by using and comparing two different feature extraction methods have been presented. With these methods, features of the knee MR images were obtained and automatic meniscus tear classification was performed by such features. Four different classifiers have been used to model the features in the classification phase. The most successful classification results were obtained from the support vector machines (SVM) with a success rate of 90.13% and the extreme learning machines (ELM) with a success rate of 87.85% via the LBP feature extraction method. It is observed that better results are obtained than the ones in similar studies in the literature. It is aimed to improve the existing success with the use of deep feature extraction methods in the future.