An Adaptive Early Stopping Technique for DenseNet169-Based Knee Osteoarthritis Detection Model

Abstract

Knee osteoarthritis (OA) detection is an important area of research in health informatics that aims to improve the accuracy of diagnosing this debilitating condition. In this paper, we investigate the ability of DenseNet169, a deep convolutional neural network architecture, for knee osteoarthritis detection using X-ray images. We focus on the use of the DenseNet169 architecture and propose an adaptive early stopping technique that utilizes gradual cross-entropy loss estimation. The proposed approach allows for the efficient selection of the optimal number of training epochs, thus preventing overfitting. To achieve the goal of this study, the adaptive early stopping mechanism that observes the validation accuracy as a threshold was designed. Then, the gradual cross-entropy (GCE) loss estimation technique was developed and integrated to the epoch training mechanism. Both adaptive early stopping and GCE were incorporated into the DenseNet169 for the OA detection model. The performance of the model was measured using several metrics including accuracy, precision, and recall. The obtained results were compared with those obtained from the existing works. The comparison shows that the proposed model outperformed the existing solutions in terms of accuracy, precision, recall, and loss performance, which indicates that the adaptive early stopping coupled with GCE improved the ability of DenseNet169 to accurately detect knee OA.

Advancements in Artificial Intelligence for Medical Computer-Aided Diagnosis

“…Al-rimy et al [ 8 ] examined the use of DenseNet169 for knee osteoarthritis detection using X-ray images. An adaptive early-stopping technique with gradual cross-entropy loss estimation is proposed to improve the model’s performance.…”

Automated system for classifying uni-bicompartmental knee osteoarthritis by using redefined residual learning with convolutional neural network

“…Furthermore, a conventional machine learning classifier was employed to take advantage of the enriched feature space and improve the classification performance of Knee OA. The proposed models are evaluated and prove their contribution to the early classification of the disease with a 90.8 % accuracy rate [ 20 ]. developed a deep convolutional neural network (DenseNet169) architecture coupled with an adaptive early stopping technique that uses gradual cross-entropy loss estimation for knee OA detection using X-ray images.…”

CDK: A novel high-performance transfer feature technique for early detection of osteoarthritis

“…In the domain of KOA detection, leveraging DL algorithms enables the training process to identify distinct patterns and relevant features within radiographical images, aiding in the identification of indicators characteristic of the disease. 11 Individuals afflicted by KOA bear one of the highest burdens in terms of disability-adjusted life years. The pervasive impact of OA extends throughout the entire knee joint, impairing normal mobility.…”