Building Robust Medical Imaging AI for Low Resource Settings скачать в хорошем качестве

Building Robust Medical Imaging AI for Low Resource Settings

This session explores how we can design and deploy medical imaging AI systems that perform reliably in low-resource healthcare environments. Drawing from real-world projects on pediatric X-ray denoising and brain tumor segmentation, the session highlights the technical and ethical challenges of domain shift, data scarcity, and bias in medical datasets. Participants will learn practical approaches, including data-centric design, structure-aware modeling, and domain adaptation, and gain insight into how equitable, context-aware medical AI can bridge the healthcare gap across Africa and beyond.