International Research Journal of Engineering and Technology (IRJET) Volume: 12 Issue: 01 | Jan 2025
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
SkinAI-Skin Disease Detection and Classification Using Machine Learning Chandrakant Barde1, Priyanka Bhamare2, Krushna Shinde3, Prachi Pawar4, Sakshi Chaudhari5 1 Assistant Professor, Dept. of Computer Engineering, GES’s R. H. Sapat College of Engineering, Nashik,
Maharashtra, India
2,3,4,5 Student, Dept. of Computer Engineering, GES’s R. H. Sapat College of Engineering, Nashik, Maharashtra, India
---------------------------------------------------------------------***--------------------------------------------------------------------Artificial Intelligence (AI), and more specifically, deep Abstract - The SkinAI research explores a cutting-edge learning, has revolutionized many fields, including approach to detecting and classifying skin diseases using healthcare. Convolutional Neural Networks (CNNs), a a Convolutional Neural Network (CNN) integrated into a type of deep learning model particularly adept at image webbased platform. Skin diseases are widespread, but recognition tasks, have shown great promise in medical early detection, especially of critical conditions like diagnostics, especially in dermatology. Skin disease melanoma, is essential for effective treatment. diagnosis relies heavily on visual assessments, making Traditional diagnostic methods are often inaccessible to it an ideal candidate for automation through AI. By individuals in remote areas, creating a demand for leveraging CNNs, it becomes possible to classify skin automated solutions. Our system allows patients and diseases based on images, allowing for a faster and healthcare professionals to upload or capture images of more accurate diagnosis process. This research focuses skin lesions via the web and receive real-time diagnostic on developing a web-based system for skin disease results and suggested treatments. The backend utilizes a detection and classification, integrating CNNs to CNN model trained on dermatological datasets to automate the diagnostic process. Users, including classify common skin diseases like eczema, psoriasis, and patients and medical professionals, can upload or melanoma. This paper provides a detailed breakdown of capture images of skin lesions, which are then analyzed the algorithm, system architecture, and model, and in real-time by a CNN model trained on dermatological compares our approach with existing machine learning datasets. The system not only identifies the skin methods. Results demonstrate high accuracy in disease condition but also suggests potential treatments, detection and a user-friendly experience, making this a offering a preliminary diagnosis that can guide further promising tool for telemedicine applications. medical consultations. The primary objective of this research is to create an accessible, user-friendly Key Words: Skin Disease Detection, Symptom-Based platform that can be utilized both as a diagnostic tool Recommendation, Machine Learning, Home Remedies Recommendation and an educational resource. By automating the initial assessment of skin diseases, the system aims to bridge 1. INTRODUCTION the gap between patients and dermatologists, particularly in areas where healthcare services are In recent years, skin diseases have become one of the scarce. This paper outlines the development process of most prevalent health concerns worldwide, affecting the system, the CNN model architecture, and the millions of individuals across various age groups and technological frameworks used, while also presenting a demographics. With conditions ranging from benign detailed comparison with existing methods for skin issues like acne to more severe cases such as disease detection. melanoma, the demand for accurate, efficient, and 1.1 Features of the Skin disease detection and timely diagnosis has never been greater. classification web application: Dermatological examinations are critical for diagnosing Our system is designed to automate the detection these conditions, but access to specialized care can be and classification of skin diseases, with a focus on limited, particularly in remote or underserved areas. accessibility and ease of use. Key features include: This limitation emphasizes the need for automated solutions that can assist both patients and healthcare User Roles: professionals in the early detection and classification of 1. Patients: Can upload or capture images but have skin diseases. restricted access to data modification.
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 658