New AI-powered app aims to improve tropical skin disease detection

NTDs are a group of diseases endemic in 149 countries and estimated to affect over 1 billion people worldwide, many of them children. They disproportionately affect marginalised communities, where a lack of trained staff and diagnostic tools makes early detection a major challenge.

SkincAIr's mission is to change this reality by putting accessible AI solutions directly in the hands of those closest to NTD patients.

The project consortium, led by Universidad Politécnica de Madrid, includes 12 partners from Europe and Sub-Saharan Africa, combining clinical, academic, technical, legal, and marketing expertise. Using deep learning to analyse skin images captured via smartphone, SkincAIr provides frontline healthcare workers with fast and accurate diagnostic assistance, helping to improve early detection and treatment in regions with limited medical resources.

The model is co-funded by the Global Health EDCTP3 Joint Undertaking and the European Commission under the Horizon Europe Research and Innovation programme. The technology will be trained using diverse datasets from multiple African regions to ensure accurate diagnosis for different skin tones, age groups and geographic backgrounds - an approach supporting fair and inclusive access to healthcare.

Built on robust data pipelines and advanced AI architectures such as convolutional neural networks and vision transformers, SkincAIr integrates real-world patient data and bias detection methods to provide a reliable and scalable diagnostic solution.

At King's College London, we're proud to lead the Explainability and Fairness workstream, ensuring the AI models behind SkincAIr are not only accurate-but also fair and trustworthy for real-world deployment.

The model is set to be deployed in five countries - Kenya, Senegal, Ethiopia, Nigeria and the Democratic Republic of the Congo - where skin NTDs are prevalent and access to early diagnosis is limited. Local clinics and health research institutions will assist with testing and deployment, as well as the creation of the first public dataset of skin NTD images in Sub-Saharan Africa, so its AI models will be validated under real clinical conditions.

Beyond its technological innovation, SkincAIr is designed to deliver tangible social impact. By equipping frontline health workers with advanced AI tools, the project will not only strengthen healthcare capacity in remote and underserved communities, but also raise awareness around prevention, diagnosis, and treatment pathways.

SkincAIr aims to highlight the gaps and systemic challenges in managing NTDs with the goal of engaging policymakers, governments and other key stakeholders, paving the way for more inclusive and equitable healthcare strategies in Sub-Saharan Africa.