ICSEJOURNALICSEJOURNAL

Journal of Computer Science and Engineering (JCSE)Journal of Computer Science and Engineering (JCSE)

Vision impairment from cataract and glaucoma remains a leading global health challenge, with 2.2 billion people affected worldwide and Indonesia registering the highest prevalence in Southeast Asia and second highest globally (Ministry of Health, 2017–2030 Roadmap; WHO 2023). Early, accurate detection is essential to prevent irreversible blindness. In this study, we develop and evaluate a native-architecture Convolutional Neural Network (CNN) to classify cataract and glaucoma using three novel non-fundus (β€œreal-eye) image subsets. Trained for 100 epochs on a balanced dataset, our CNN achieves 98.67 % accuracy, precision of 98.8 %, and recall of 98.5 % on a held-out test set. The resulting TensorFlow-saved model is deployed in the browser via TensorFlow.js, enabling real-time, platform-agnostic inference without specialized hardware. This work demonstrates that non-fundus imaging, combined with lightweight CNN design and web deployment, can provide a practical, high-accuracy tool for early eye-disease screening in low-resource settings.

This study investigates eye disease classification using the Convolutional Neural Network (CNN) method with a native CNN architecture.The studys novel contribution lies in the utilization of three non-fundus image classes.The preprocessing stage involves resizing images to 100x100 pixels.The CNN model is implemented using 100 epochs for its hyperparameters.

Berdasarkan penelitian ini, terdapat beberapa arah penelitian lanjutan yang menjanjikan. Pertama, perlu dilakukan eksplorasi lebih lanjut terhadap jenis-jenis citra non-fundus lainnya, seperti citra yang diambil dengan berbagai kondisi pencahayaan atau sudut pandang, untuk meningkatkan robustitas model terhadap variasi kondisi pengambilan gambar. Kedua, pengembangan model dapat difokuskan pada integrasi dengan perangkat mobile, sehingga memungkinkan skrining penyakit mata secara langsung di lapangan dengan memanfaatkan kamera smartphone. Ketiga, penelitian selanjutnya dapat menginvestigasi penggunaan teknik transfer learning dengan memanfaatkan dataset citra mata yang lebih besar untuk meningkatkan akurasi dan efisiensi pelatihan model.

  1. Cataract: the relation between myopia and cataract morphology. | British Journal of Ophthalmology. cataract... bjo.bmj.com/content/71/6/405Cataract the relation between myopia and cataract morphology British Journal of Ophthalmology cataract bjo bmj content 71 6 405
Read online
File size752.34 KB
Pages7
DMCAReport

Related /

ads-block-test