Toward Reliable and Explainable Nail Disease Classification: Leveraging Adversarial Training and Grad-CAM Visualization

2026-02-04

AI Summary

This study addresses the challenge of early diagnosis of nail diseases, which is hindered by subtle visual differences, by proposing a deep learning–based automated classification method. Leveraging a publicly available dataset, the authors systematically evaluate several state-of-the-art convolutional neural network architectures—including InceptionV3, DenseNet201, EfficientNetV2, and ResNet50—and enhance model robustness through adversarial training. To improve clinical interpretability, decision visualization is achieved using Grad-CAM and SHAP techniques. Experimental results demonstrate that InceptionV3 and DenseNet201 achieve classification accuracies of 95.57% and 94.79%, respectively, offering high diagnostic performance while significantly improving reliability and clinical trustworthiness.

Technology Category

Computer Vision: Adversarial Attacks & RobustnessMachine Learning: Deep Neural Architectures and Foundation ModelsKnowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsWeb Mining and Content Analysis: Large pretrained models with web dataResponsible Web: Algorithmic accountability and transparency on the web

Abstract

Human nail diseases are gradually observed over all age groups, especially among older individuals, often going ignored until they become severe. Early detection and accurate diagnosis of such conditions are important because they sometimes reveal our body's health problems. But it is challenging due to the inferred visual differences between disease types. This paper presents a machine learning-based model for automated classification of nail diseases based on a publicly available dataset, which contains 3,835 images scaling six categories. In 224x224 pixels, all images were resized to ensure consistency. To evaluate performance, four well-known CNN models-InceptionV3, DenseNet201, EfficientNetV2, and ResNet50 were trained and analyzed. Among these, InceptionV3 outperformed the others with an accuracy of 95.57%, while DenseNet201 came next with 94.79%. To make the model stronger and less likely to make mistakes on tricky or noisy images, we used adversarial training. To help understand how the model makes decisions, we used SHAP to highlight important features in the predictions. This system could be a helpful support for doctors, making nail disease diagnosis more accurate and faster.

Problem

• nail disease classification
• early detection
• visual differences
• diagnostic accuracy
• automated diagnosis

Innovation

• adversarial training
• Grad-CAM
• explainable AI
• nail disease classification
• SHAP