Bluish Veil Detection and Lesion Classification using Custom Deep Learnable Layers with Explainable Artificial Intelligence (XAI)

2025-07-10

AI Summary

This study addresses the challenges of low inter-observer agreement and high annotation cost in identifying the blue-white veil (BWV) feature for early melanoma diagnosis. To this end, we propose a weakly supervised, interpretable deep learning framework. First, a color-thresholding–based automatic annotation algorithm generates high-fidelity BWV pseudo-labels. Second, a customized deep convolutional neural network is designed, wherein standard ReLU activations are replaced with learnable activation layers to better capture BWV-specific textural and chromatic patterns. Third, explainable AI (XAI) techniques—including gradient-weighted class activation mapping (Grad-CAM)—are integrated to enhance decision transparency. Evaluated on PH2, ISIC, Derm7pt, and a combined dataset, our method achieves BWV detection accuracies of 85.71%, 95.00%, 90.00%, and 95.05%, respectively—significantly outperforming baseline models. This work establishes a novel paradigm for automated, interpretable BWV recognition in dermoscopic imaging.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Transparent, Interpretable, Explainable ML
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Melanoma, one of the deadliest types of skin cancer, accounts for thousands of fatalities globally. The bluish, blue-whitish, or blue-white veil (BWV) is a critical feature for diagnosing melanoma, yet research into detecting BWV in dermatological images is limited. This study utilizes a non-annotated skin lesion dataset, which is converted into an annotated dataset using a proposed imaging algorithm based on color threshold techniques on lesion patches and color palettes. A Deep Convolutional Neural Network (DCNN) is designed and trained separately on three individual and combined dermoscopic datasets, using custom layers instead of standard activation function layers. The model is developed to categorize skin lesions based on the presence of BWV. The proposed DCNN demonstrates superior performance compared to conventional BWV detection models across different datasets. The model achieves a testing accuracy of 85.71% on the augmented PH2 dataset, 95.00% on the augmented ISIC archive dataset, 95.05% on the combined augmented (PH2+ISIC archive) dataset, and 90.00% on the Derm7pt dataset. An explainable artificial intelligence (XAI) algorithm is subsequently applied to interpret the DCNN's decision-making process regarding BWV detection. The proposed approach, coupled with XAI, significantly improves the detection of BWV in skin lesions, outperforming existing models and providing a robust tool for early melanoma diagnosis.

Problem

• Detect bluish veil in skin lesions using deep learning
• Classify melanoma with custom CNN layers and XAI
• Improve BWV detection accuracy across diverse datasets

Innovation

• Custom deep learnable layers replace standard activation
• Color threshold algorithm creates annotated dataset
• Explainable AI interprets DCNN's BWV detection decisions