Rahayu Noveandini
Universitas Mahakarya Asia PSDKU Jakarta

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

SEGMENTASI DAN KLASIFIKASI LESI KULIT MENGGUNAKAN YOLO V11 Rahayu Noveandini; Maria Sri Wulandari; Farhan Rasyad
Journal of Information System, Informatics and Computing Vol 10 No 1 (2026): JISICOM (June 2026)
Publisher : Sekolah Tinggi Manajemen Informatika dan Komputer Jayakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52362/jisicom.v10i1.2385

Abstract

Skin cancer poses a serious threat if not detected early. Limited access to dermatology services calls for a reliable digital solution. This study proposes DermaScan, a web application for the early diagnosis of skin cancer based on the YOLOv11 model, capable of simultaneously segmenting and classifying skin lesions. The model was trained using the pre-processed HAM10000 dataset, including data augmentation and conversion of annotations to the YOLO format. The training process utilized a GPU and mixed precision for efficiency. Evaluation results demonstrate high performance with mAP50 = 0.91 and mAP50-95 = 0.735 for detection, as well as mAP50 (Seg) = 0.905 and mAP50-95 (Seg) = 0.706 for segmentation, proving good accuracy in mapping and identifying lesions. The best performa model is implemented into a web application using Flask (backend) and React.js (frontend) with a single-page application interface. Users can upload images of lesions and receive real-time prediction results, including disease type, severity level, and medical recommendations. DermaScan serves as a fast and easily accessible non-invasive screening tool. This system has the potential to raise public awareness and support healthcare professionals in the early detection of skin cancer, although it is not a substitute for clinical diagnosis