Explainable AI-Based Model for Detection and Classification of Cancerous and Non-Cancerous Skin Conditions

DOI

10.2991/978-94-6239-693-7_2How to use a DOI?

Keywords

Explainable AI; YOLOv8; Skin Lesion Detection; Medical Imaging; Deep Learning

Abstract

The paper presents an explainable AI (XAI) framework that enables detecting, localizing, and segmenting of skin lesions with the YOLOv8 object detector model, which is available as a Streamlit web application. This system consists of a trained YOLOv8 detector and other image-processing techniques (hair artifact removal, contrast-limited adaptive histogram equalization, and adaptive thresholding) and optimization of the lesion area (Watershed and GrabCut) to accurately outline the areas of lesions in dermoscopic photographs. Analysis of position analysis is carried out to produce heatmaps highlighting clinically important parts of the image. The system is in favor of real-time inference and visualization. It has been experimentally confirmed that this method is robust to variations in light levels, noise and background, and has the ability to obtain correct lesion delineation and good classification. Grad-CAM++ and LIME-inspired occlusion mapping is also introduced to make the model more interpretable, thus allowing clinicians to interpret and trust the model. In general, the framework is effective to combine high-performance deep learning and user-friendly explanatory visuals, which will develop reliable AI technologies in dermatology.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - A. Vikram
AU  - S. Poornima
PY  - 2026
DA  - 2026/06/16
TI  - Explainable AI-Based Model for Detection and Classification of Cancerous and Non-Cancerous Skin Conditions
BT  - Proceedings of the International Conference on Intelligent Systems for a Sustainable Future (ISSF 2026)
PB  - Atlantis Press
SP  - 5
EP  - 11
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-693-7_2
DO  - 10.2991/978-94-6239-693-7_2
ID  - Vikram2026
ER  -