Classifying the Severity of Knee Osteoarthritis on X-ray Images Using NASNet Mobile
Journal of Applied Informatics and Computing · 16.04.2026 · 10 (2) · 1191-1199
DOI: 10.30871/jaic.v10i2.12246 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Applied Informatics and Computing, online erschienen am 16.04.2026. Verfasst von Adelia Damayanti und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Knee osteoarthritis is a joint condition characterized by cartilage breakdown in the knee, most frequently affecting older adults. Early diagnosis of knee osteoarthritis is critical because it can delay disease progression and improve patient quality of life. This study aims to classify the severity of knee osteoarthritis using X-ray images by employing a Convolutional Neural Network (CNN) with the NASNet Mobile model. The dataset consists of 1500 knee X-ray images, divided into three classes: normal, moderate, and severe. The model was trained with various batch sizes and learning rates. The highest accuracy of 92.53% was achieved on the k-fold cross-validation dataset using a batch size of 32 and a learning rate of 0.001, yielding a sensitivity of 92.72% and a specificity of 96.32%.
Zitieren
APA
Damayanti, A., Novitasari, D. C. R., Hakim, L., & Musfiroh, M. (2026). Classifying the Severity of Knee Osteoarthritis on X-ray Images Using NASNet Mobile. Journal of Applied Informatics and Computing, 10(2), 1191-1199. https://doi.org/10.30871/jaic.v10i2.12246
Vancouver
Damayanti A, Novitasari DCR, Hakim L, Musfiroh M. Classifying the Severity of Knee Osteoarthritis on X-ray Images Using NASNet Mobile. Journal of Applied Informatics and Computing. 2026;10(2):1191-1199. doi:10.30871/jaic.v10i2.12246
BibTeX
@article{Damayanti2026,
author = {Damayanti, Adelia and Novitasari, Dian Candra Rini and Hakim, Lutfi and Musfiroh, Musfiroh},
title = {Classifying the Severity of Knee Osteoarthritis on X-ray Images Using NASNet Mobile},
journal = {Journal of Applied Informatics and Computing},
year = {2026},
volume = {10},
number = {2},
pages = {1191--1199},
doi = {10.30871/jaic.v10i2.12246}
}
EUVIMED