DermAI: Clinical dermatology acquisition through quality-driven image collection for AI classification in mobile
This addresses data quality and diversity issues for dermatologists and AI developers, but it is incremental as it builds on existing mobile and AI tools.
The paper tackled the problem of limited AI adoption in dermatology due to biased datasets and variable image quality by introducing DermAI, a smartphone app for real-time capture and classification of skin lesions, where fine-tuning with local data improved performance over models trained on public datasets.
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.