Skin Disease Classification using Deep Learning with Grad-CAM Visualization
International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)
DOI: 10.15662/ijeetr.2026.0802245 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von C. Leelavathi und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Skin cancer is one of the most widespread and terminal skin diseases all over the world. Prompt diagnosis is quite vital in improving the survival rates of patients through proper diagnosis. The new trends in DL have enabled the automatic analysis of skin lesions through the use of classification and object recognition. To have a complete examination of the skin diseases, this writing presents a two-branch DL architecture by integrating image-level classification and localization of abnormality regions. A set of a transfer-learning architecture, including VGG16, ResNet50, InceptionV3, DenseNet121, Xception, ConvNeXt-Tiny, and an ensemble are trained on labeled dermoscopic images to classify the images into groups. YOLO based models such as YOLOv5s6, YOLOv8, YOLOv9 and YOLOv11 are trained with notes of the YOLO formatted bounding boxes. It has been extensively experimented and ConvNeXt-Tiny is the most successful at classification (94.88% accuracy), and YOLOv8 is the most successful at recognition (73.9% mAP, 72.2% precision, and 79.6% recall). The explainable AI techniques based on gradient-based camera (Grad-CAM) are applied to visualize the significant regions influencing predictions and simplify the model. Moreover, an application based on Flask web application is being developed to enable real-time inference, which is a merger of detection, classification, and visual description to assist health care specialists in making decisions
Zitieren
APA
Leelavathi, C., Sowjanya, A., Harika, T., & Reddy, A. V. K. (2026). Skin Disease Classification using Deep Learning with Grad-CAM Visualization. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802245
Vancouver
Leelavathi C, Sowjanya A, Harika T, Reddy AVK. Skin Disease Classification using Deep Learning with Grad-CAM Visualization. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802245
BibTeX
@article{Leelavathi2026,
author = {Leelavathi, C. and Sowjanya, A. and Harika, T. and Reddy, A. Vamshi Krishna},
title = {Skin Disease Classification using Deep Learning with Grad-CAM Visualization},
journal = {International Journal of Engineering & Extended Technologies Research},
year = {2026},
volume = {08},
number = {02},
doi = {10.15662/ijeetr.2026.0802245}
}
EUVIMED