Webinars
Federated Learning (FL) is a machine learning approach that allows a model to be trained across multiple decentralised devices or servers holding local data samples, without exchanging them. Instead of sending data to a central server, updates to the model are computed locally on each device, and only model parameters are aggregated or combined. This approach minimises the risk of exposing sensitive user data. It strikes a balance between model performance and data privacy.
Strokes are a major cause of disability worldwide, with over a million cases and nearly half a million deaths annually in Europe, leading to a growing social and economic burden due to an ageing population. To improve personalized care and prevent relapse, the TRUSTroke project introduces an AI-based tool built on a FL clinical and patient-reported data to support clinicians, patients, and caregivers in reliably assessing disease progression and managing chronically disabled stroke patients.