A Machine Vision Approach to Preliminary Skin Lesion Assessments

TL;DR

This study compares rule-based and machine learning methods for early skin lesion assessment, showing that custom CNNs trained from scratch outperform transfer learning models on small medical datasets.

Contribution

It introduces a comprehensive system combining the ABCD rule with machine learning, highlighting the effectiveness of custom CNNs over transfer learning in skin lesion classification.

Findings

Custom CNN achieved 78.5% accuracy and 86.5% recall.

Transfer learning with EfficientNet-B0 underperformed due to domain shift.

Handcrafted features limited by complexity reduction.

Abstract

Early detection of malignant skin lesions is critical for improving patient outcomes in aggressive, metastatic skin cancers. This study evaluates a comprehensive system for preliminary skin lesion assessment that combines the clinically established ABCD rule of dermoscopy (analyzing Asymmetry, Borders, Color, and Dermoscopic Structures) with machine learning classification. Using a 1,000-image subset of the HAM10000 dataset, the system implements an automated, rule-based pipeline to compute a Total Dermoscopy Score (TDS) for each lesion. This handcrafted approach is compared against various machine learning solutions, including traditional classifiers (Logistic Regression, Random Forest, and SVM) and deep learning models. While the rule-based system provides high clinical interpretability, results indicate a performance bottleneck when reducing complex morphology to five numerical features. Experimental findings show that transfer learning with EfficientNet-B0 failed significantly due to domain shift between natural and medical images. In contrast, a custom three-layer Convolutional Neural Network (CNN) trained from scratch achieved 78.5% accuracy and 86.5% recall on median-filtered images, representing a 19-point accuracy improvement over traditional methods. The results demonstrate that direct pixel-level learning captures diagnostic patterns beyond handcrafted features and that purpose-built lightweight architectures can outperform large pretrained models for small, domain-specific medical datasets.