Android Teledermatology Application for Skin Lesions Classification Based on Dermoscopic Images
DOI:
https://doi.org/10.33795/jartel.v14i4.6227Keywords:
Deep Learning, Hyperparameter Tuning, Pre-trained Model, Skin Lesions, Convolutional Neural NetworkAbstract
Skin cancer is a commonly diagnosed cancer worldwide, which is ranked third in Indonesia. While this type of cancer is curable in the early stage, there is an increase in the risk of death in the late stage. Thus, early detection is essential to minimize the spread of cancer. A prototype of an Android-based application was developed in this research to classify skin lesions. Dermoscopy images from the MNIST HAM10000 dataset comprised 10,015 images across seven skin cancer classes. Five pre-trained models from Keras, ResNet50V2, DenseNet121, InceptionV3, Xception, and MobileNet will be utilized for training and testing. During training, each model will be tested for classification performance based on changes in several hyperparameter variables, including learning rate, optimizer, batch size, and dropout.After testing, each model's performance improvement was obtained with adjustments to the training hyperparameters. The Xception model achieved the best accuracy at 92.5 percent. This model was then implemented in an Android-based application. The testing results on Android showed the same accuracy for original dermoscopic images from the test dataset. However, a decrease in accuracy was observed when testing with the camera, which noise or differences in resolution might cause
References
Urban, Katelyn, Sino Mehrmal, Prabhdeep Uppal, Rachel L. Giesey, and Gregory R. Delost. 2021. “The Global Burden of Skin Cancer: A Longitudinal Analysis From the Global Burden of Disease Study, 1990–2017.” JAAD International 2 (March): 98–108.
Wilvestra, Silvia, Sri Lestari, and Ennesta Asri. "Studi retrospektif kanker kulit di poliklinik ilmu kesehatan kulit dan kelamin RS Dr. M. Djamil Padang periode 2015-2017." Jurnal Kesehatan Andalas 7 (2018): 47-49.
Mikołajczyk, Agnieszka, and Michał Grochowski. "Data augmentation for improving deep learning in image classification problem." In 2018 international interdisciplinary PhD workshop (IIPhDW), pp. 117-122. IEEE, 2018.
Purnama, I. Ketut Eddy, Arta Kusuma Hernanda, Anak Agung Putri Ratna, Ingrid Nurtanio, Afif Nurul Hidayati, Mauridhi Hery Purnomo, Supeno Mardi Susiki, and Reza Fuad. "Disease classification based on dermoscopic skin images using convolutional neural network in teledermatology system." In 2019 international conference on computer engineering, network, and intelligent multimedia (CENIM), pp. 1-5. IEEE, 2019.
Rochmawanti, Ovy, Fitri Utaminingrum, and Fitra A. Bachtiar. "Analisis Performa Pre-Trained Model Convolutional Neural Network dalam Mendeteksi Penyakit Tuberkulosis." Jurnal Teknologi Informasi dan Ilmu Komputer 8, no. 4 (2021): 805-814.
Ali, Karar, Zaffar Ahmed Shaikh, Abdullah Ayub Khan, and Asif Ali Laghari. "Multiclass skin cancer classification using EfficientNets–a first step towards preventing skin cancer." Neuroscience Informatics 2, no. 4 (2022): 100034.
Agustina, Regita, Rita Magdalena, And Nor Kumalasari Caecar Pratiwi. "Klasifikasi Kanker Kulit menggunakan Metode Convolutional Neural Network dengan Arsitektur VGG-16." ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika 10, no. 2 (2022): 446.
Prastika, Indah Widhi, and Eri Zuliarso. "Deteksi penyakit kulit wajah menggunakan tensorflow dengan metode convolutional neural network." Jurnal Manajemen Informatika dan Sistem Informasi 4, no. 2 (2021): 84-91.
Tschandl, Philipp, Cliff Rosendahl, and Harald Kittler. "The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions." Scientific data 5, no. 1 (2018): 1-9.
Baig, Abdul Rauf, Qaisar Abbas, Riyad Almakki, Mostafa EA Ibrahim, Lulwah AlSuwaidan, and Alaa ES Ahmed. "Light-dermo: A lightweight pretrained convolution neural network for the diagnosis of multiclass skin lesions." Diagnostics 13, no. 3 (2023): 385.
Abiodun, Oludare Isaac, Aman Jantan, Abiodun Esther Omolara, Kemi Victoria Dada, Abubakar Malah Umar, Okafor Uchenwa Linus, Humaira Arshad, Abdullahi Aminu Kazaure, Usman Gana, and Muhammad Ubale Kiru. "Comprehensive review of artificial neural network applications to pattern recognition." IEEE access 7 (2019): 158820-158846.
Purwono, Purwono, Alfian Ma'arif, Wahyu Rahmaniar, Haris Imam Karim Fathurrahman, Aufaclav Zatu Kusuma Frisky, and Qazi Mazhar ul Haq. "Understanding of convolutional neural network (cnn): A review." International Journal of Robotics and Control Systems 2, no. 4 (2022): 739-748.
Putra, Wayan Suartika Eka. "Klasifikasi citra menggunakan convolutional neural network (CNN) pada caltech 101." Jurnal Teknik ITS 5, no. 1 (2016).
Muraina, Ismail. "Ideal dataset splitting ratios in machine learning algorithms: general concerns for data scientists and data analysts." In 7th International Mardin Scientific Research Conference, pp. 496-504. 2022.
Thanapol, Panissara, Kittichai Lavangnananda, Pascal Bouvry, Frédéric Pinel, and Franck Leprévost. "Reducing overfitting and improving generalization in training convolutional neural network (CNN) under limited sample sizes in image recognition." In 2020-5th International Conference on Information Technology (InCIT), pp. 300-305. IEEE, 2020
