DermAI: Clinical dermatology acquisition through quality-driven image collection for AI classification in mobile

TL;DR

DermAI is a smartphone app that improves dermatology AI by enabling real-time, quality-checked, and diverse skin lesion data collection during routine visits, enhancing model generalization.

Contribution

Introduces DermAI, a mobile app for real-time skin lesion data collection with quality control and local adaptation, addressing dataset bias and variability issues.

Findings

Models trained on public data failed to generalize to new samples.

Fine-tuning with local data improved model performance.

Diverse, standardized data collection is crucial for effective AI in dermatology.

Abstract

AI-based dermatology adoption remains limited by biased datasets, variable image quality, and limited validation. We introduce DermAI, a lightweight, smartphone-based application that enables real-time capture, annotation, and classification of skin lesions during routine consultations. Unlike prior dermoscopy-focused tools, DermAI performs on-device quality checks, and local model adaptation. The DermAI clinical dataset, encompasses a wide range of skin tones, ethinicity and source devices. In preliminary experiments, models trained on public datasets failed to generalize to our samples, while fine-tuning with local data improved performance. These results highlight the importance of standardized, diverse data collection aligned with healthcare needs and oriented to machine learning development.