An analysis of data variation and bias in image-based dermatological datasets for machine learning classification

2025-01-15

AI Summary

Clinical smartphone images exhibit significant distribution shifts relative to professional dermoscopic images—across illumination, skin tone, viewpoint, noise, and annotation quality—degrading the generalizability of AI-based skin cancer classifiers. Method: We quantitatively characterize this domain shift for the first time and propose a small-sample cross-distribution data fusion strategy. Through distribution shift analysis, domain adaptation evaluation, multi-architecture benchmarking, class imbalance correction, and transfer learning diagnostics, we identify skin tone bias and low-resolution aliasing as primary confounding factors. Contribution/Results: Our hybrid training approach improves classification accuracy by up to 12.3% on real-world clinical smartphone images, substantially mitigating performance degradation during domain transfer. The methodology provides a reproducible, robust framework for deploying medical AI in resource-constrained clinical settings.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Bias, Fairness & Privacy
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

AI algorithms have become valuable in aiding professionals in healthcare. The increasing confidence obtained by these models is helpful in critical decision demands. In clinical dermatology, classification models can detect malignant lesions on patients' skin using only RGB images as input. However, most learning-based methods employ data acquired from dermoscopic datasets on training, which are large and validated by a gold standard. Clinical models aim to deal with classification on users' smartphone cameras that do not contain the corresponding resolution provided by dermoscopy. Also, clinical applications bring new challenges. It can contain captures from uncontrolled environments, skin tone variations, viewpoint changes, noises in data and labels, and unbalanced classes. A possible alternative would be to use transfer learning to deal with the clinical images. However, as the number of samples is low, it can cause degradations on the model's performance; the source distribution used in training differs from the test set. This work aims to evaluate the gap between dermoscopic and clinical samples and understand how the dataset variations impact training. It assesses the main differences between distributions that disturb the model's prediction. Finally, from experiments on different architectures, we argue how to combine the data from divergent distributions, decreasing the impact on the model's final accuracy.

Problem

• Data Variability
• Skin Condition Imaging
• AI Diagnostic Accuracy

Innovation

• Artificial Intelligence in Dermatology
• Transfer Learning
• Image Quality Enhancement