Web-Based Skin Cancer and Skin Disease Detection Using Image Analysis and Machine Learning
International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)
DOI: 10.15662/ijeetr.2026.0802164 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von S. Malathi und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Skin diseases are among the most common health conditions worldwide, and early diagnosis plays a crucial role in effective treatment and prevention of severe complications. Recent advances in artificial intelligence, particularly deep learning, have shown promising results in automated skin disease classification using medical images. However, many existing systems focus primarily on prediction accuracy while neglecting reliability assessment and explainability, which are critical for real-world clinical adoption. To address these limitations, this paper presents an Intelligent Skin Disease Diagnosis System with Reliability-Aware Explainable AI. The proposed system utilizes a hybrid architecture that combines deep learning and machine learning techniques. A pre-trained MobileNetV2 convolutional neural network is employed as a feature extractor to capture discriminative visual characteristics from dermoscopic or skin lesion images. The extracted deep features are then fed into a machine learning classifier to predict the skin disease category along with a confidence score. This hybrid approach reduces computational complexity while maintaining high classification performance, making the system suitable for real-time web-based deployment. To enhance transparency and trustworthiness, the system integrates Gradient-weighted Class Activation Mapping (Grad-CAM) to generate visual explanations highlighting the regions of the image that most influence the model’s predictions. Beyond simple visualization, a novel Grad-CAM Reliability Score (GRS) is introduced to quantitatively evaluate the reliability of each prediction based on activation intensity and spatial distribution. This reliability score, combined with prediction confidence and probability margin analysis, enables robust decision validation. When the system detects low confidence or unreliable predictions, it automatically classifies the outcome as “Uncertain” and recommends medical consultation instead of providing a potentially misleading diagnosis. This reliability-aware decision mechanism significantly reduces the risk of false predictions and improves clinical safety. Additionally, the system provides severity estimation, precautionary guidance, and a user-friendly web interface to enhance accessibility. Experimental evaluation using benchmark skin disease datasets demonstrates that the proposed system achieves competitive classification accuracy while offering superior explainability and reliability assessment compared to traditional deep learning approaches. The integration of explainable AI with quantitative reliability validation represents a key contribution of this work. The proposed framework can serve as an effective decision-support tool for preliminary skin disease screening and future clinical AI systems.
Zitieren
APA
Malathi, S., Sivakumar, S., & Nishal, V. (2026). Web-Based Skin Cancer and Skin Disease Detection Using Image Analysis and Machine Learning. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802164
Vancouver
Malathi S, Sivakumar S, Nishal V. Web-Based Skin Cancer and Skin Disease Detection Using Image Analysis and Machine Learning. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802164
BibTeX
@article{Malathi2026,
author = {Malathi, S. and Sivakumar, S. and Nishal, V.},
title = {Web-Based Skin Cancer and Skin Disease Detection Using Image Analysis and Machine Learning},
journal = {International Journal of Engineering & Extended Technologies Research},
year = {2026},
volume = {08},
number = {02},
doi = {10.15662/ijeetr.2026.0802164}
}
EUVIMED