The guide for applicants to our Open Calls has recently been updated, incorporating important changes both in the deadlines and in the technical details of one of the challenges in this call. The most significant adjustment is the new deadline for submitting proposals, which is now 15 October at 15:00 (Brussels time), i.e. one day later than originally planned. With this extension, we aim to give participants additional time to refine their proposals and ensure the highest quality in their submissions.
In addition to the date change, the update includes modifications to Challenge 8: ‘Health Monitoring AI for Edge Devices with Privacy-First,’ which have been added for explanatory and clarifying purposes regarding the expected results. This challenge seeks to promote artificial intelligence solutions applied to health monitoring on Edge devices, with a focus on privacy and efficiency.
Among the points specified in this new version of the guide is the need to develop a functional prototype based on an application that runs on a Raspberry Pi or other similar device with health sensors, capable of making inferences directly on the device without relying on the cloud and integrated with DMWay. The implementation of a secure federated learning system is also expected, where training is performed locally and the aggregation of results is carried out on the server, applying privacy preservation techniques such as differential privacy or secure aggregation.
Applicants must also submit a technical report detailing the system architecture, the technologies used and the energy optimisation methods, together with experimental results evaluating energy consumption, model accuracy and customisation capabilities. Finally, a demonstrator or proof of concept is required to demonstrate the prototype's performance using synthetic or public data, simulating the early detection of stress or cardiac abnormalities and showing its performance in real time without compromising the privacy of personal data.
With these updates, we reaffirm our commitment to transparency, responsible innovation and the development of artificial intelligence technologies that respect privacy and promote a positive impact in the field of health. We invite all interested parties to consult the new guide and submit their proposals by 15 October at 15:00 (Brussels time).