Treffer: Classification of anterior cruciate ligament (ACL) injury based on convolutional neural network.

Image classification plays an important role in the domain of deep learning. This paper focuses on building a decision support system to perform knee joint Anterior Cruciate Ligament (ACL) Magnetic Resonance Imaging (MRI) analysis. A Convolutional Neural Network (CNN) based classification system is developed to assist medical experts in making decisions about the types of ACL injuries. A total of 515 samples were collected from Radiology Department, Hospital Kuala Lumpur (HKL) forming a database of MRI images of the knee with further focus on coronal-view only. All experimental work was carried out using Python. Classification ACL injury is based on 3 types i.e. complete tear, partial tear, and normal classes. Results show that the accuracy of the ACL injury classification using confusion matrix analysis is 83.90% and consequently Area Under ROC curve (AUC) is 94.8%. These results outperform our previous works on similar issue, and it has been shown that proposed deep learning is able to improve the diagnosis of ACL injuries from knee MRI images. [ABSTRACT FROM AUTHOR]

Copyright of AIP Conference Proceedings is the property of American Institute of Physics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)