Oral Lesion Image Classification Using Convolutional Neural Network (CovNets) Method Based on MobileNetV2 Architecture
Abstract
Oral lesions, which can appear in various areas of the oral cavity, are often an early indication of oral cancer, one of the most common cancers worldwide and a leading cause of cancer death in South Asia, Southeast Asia, and the Western Pacific. Oral cancer can affect various parts of the mouth and throat, with contributing factors including genetics, smoking, and viral infections. Early detection is critical for effective management of oral cancer, allowing for early treatment that increases the chance of cure and reduces the risk of complications. This study used a Convolutional Neural Network (CNN) to detect images of oral lesions, including benign and malignant lesions, by utilizing the TensorFlow Object Detection API and data from the Oral Images Dataset. Testing with 40 images (20 benign and 20 malignant lesions) showed an accuracy of 92.5%, a precision of 95%, and a recall of 90%, demonstrating the potential of CNN in efficiently detecting oral lesions.
Downloads
References
Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov and Liang-Chieh Chen. 2019. MobileNetV2: Inverted Residuals and Linear Bottlenecks.
Sayuti Rahman, Arnes Sembiring, Dodi Siregar, Husnul Khair, I Gusti Prahmana, Ratih Puspadini and Muhammad Zen. 2023. PYTHON : DASAR DAN PEMROGRAMAN BERORIENTASI OBJEK.
Humaira Nazir and Wakambum Monalisa, 2020. Study of Precancerous Lesions and Conditions by Clinical Examination, Chemiluminescence, and Toluidine Blue as Early Detection Tool- Retrospective Study. https://aimdrjournal.com/wp-content/uploads/2021/06/DE1_OA_Humaira-edit.pdf
Aarushi Shah, Manan Shah, Aum Pandya, Rajat Sushra, Ratnam Sushra, Manya Mehta, Keyur Patel, and Kaushal Patel. 2023. A comprehensive study on skin cancer detection using artificial neural network (ANN) and convolutional neural network (CNN).
Rohim, A., Sari, Y. A., & Tibyani. (2019). Convolution neural network (cnn) untuk pengklasifikasian citra makanan tradisional. Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komputer, 3(7), 7038–7042.
Swasono, D. I., Wijaya, M. A. R., & Hidayat, M. A. (2023). Klasifikasi Penyakit pada Citra Buah Jeruk Menggunakan Convolutional Neural Networks (CNN) dengan Arsitektur Alexnet. INFORMAL: Informatics Journal, 8(1), 68.
Tsany, A., & Dzaky, R. (2021). Deteksi Penyakit Tanaman Cabai Menggunakan Metode Convolutional Neural Network. E-Proceeding of Engineering, 8(2), 3039–3055.
Ulya, H., Darmanti, S., & Ferniah, R. S. (2020). Pertumbuhan Daun Tanaman Cabai (Capsicum annuum L.) yang Diinfeksi Fusarium oxysporum pada Umur Tanaman yang Berbeda. Jurnal Akademika Biologi, 9(1), 1–6.
Wulandari, I., Yasin, H., & Widiharih, T. (2020). Klasifikasi Citra Digital Bumbu Dan Rempah Dengan Algoritma Convolutional Neural Network (Cnn). Jurnal Gaussian, 9(3), 273–282. https://doi.org/10.14710/j.gauss.v9i3.27416
Yuliany, S., Aradea, & Andi Nur Rachman. (2022). Implementasi Deep Learning pada Sistem Klasifikasi Hama Tanaman Padi Menggunakan Metode Convolutional Neural Network (CNN). Jurnal Buana Informatika, 13(1), 54–65.
Ezar Al Rivan, M., & Giovri Riyadi, A. (2021). Perbandingan Arsitektur LeNet dan AlexNet Pada Metode Convolutional Neural Network Untuk Pengenalan American Sign Language. Jurnal Komputer Terapan, 7(Vol. 7 No. 1 (2021)), 53–61.
Hayati, E., Mahmud, T., & Fazil, R. (2012). PENGARUH JENIS PUPUK ORGANIK DAN VARIETAS TERHADAP PERTUMBUHAN DAN HASIL TANAMANCABAI (Capsicum annum L.). Urnal Floratek, 7(2), 173–181.
Hidayat, D. (2022). Klasifikasi Jenis Mangga Berdasarkan Bentuk Dan Tekstur Daun Menggunakan Metode Convolutio Nalneural Network(Cnn) Classification of Types of Mango Based on Leave Shape and Texture Using Convolutio Nalneural Network(Cnn) Method. Journal of Information Technology and Computer Science (INTECOMS), 5(1), 98–103.
Ilahiyah, S., & Nilogiri, A. (2018). Implementasi Deep Learning Pada Identifikasi Jenis Tumbuhan Berdasarkan Citra Daun Menggunakan Convolutional Neural Network. JUSTINDO (Jurnal Sistem Dan Teknologi Informasi Indonesia), 3(2), 49–56.
