Hybrid Methodologies for Segmentation and Classification of Skin Diseases: A Study
Hybrid Methodologies for Segmentation and Classification of Skin Diseases: A Study作者机构：Department of Public Health and Informatics Jahangirnagar University Savar Dhaka Bangladesh Department of Computer Science and Engineering Jahangirnagar University Savar Dhaka Bangladesh
出 版 物：《Journal of Computer and Communications》 (电脑和通信（英文）)
年 卷 期：2021年第9卷第4期
页 面：67-84页
学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）]
主 题：Segmentation Feature Extraction Classification Machine Learning
摘 要：Skin disorders are a serious global health problem for humans. These disorders become dangerous when they grow into the malignant stage. Hence, it is necessary to detect these diseases at their early stage. A mobile-based automated skin disease detection system is vital for detecting skin diseases. This system also offers cure or treatment plans to the affected person through the short message service (SMS) or electronic mail (e-mail). An effective skin disease detection system consists of three processes: segmentation, feature extraction, and classification. Several hybrid methodologies are already developed for the above-mentioned processes for detecting skin diseases at the initial stage. This research gives a standard hybrid framework for detecting skin diseases and highlights some design requirements for achieving high accuracy. Existing state-of-the-art hybrid methods of three processes for detecting skin diseases along with their limitations are also summarized. It also identifies the challenges for developing an effective skin disease detection system and gives future research directions.
维普期刊数据库