RE-SAMPLE - Privacy-Preserving Machine Learning
(dies ist ein Projekt zusammen mit dem Forschungsteam an der Universität Twente in den Niederlanden)
In RE-SAMPLE (= REal-time data monitoring for Shared, Adaptive, Multi-domain and Personalised prediction and decision making for Long-term Pulmonary care Ecosystems), a European Horizon 2020 project, we aim to take a giant leap in the field of complex chronic condition (CCC) management building upon and going beyond existing initiatives, towards evidence-based, inclusive, preventive care and targeted treatment. This enables to “treat a person, not the disease(s)”. Multi-morbidity is highly prevalent in patients with Chronic Obstructive Pulmonary Disease (COPD). Timely and preventive care is essential, as exacerbations of COPD and complications are detrimental to patients.
The RE-SAMPLE objectives are to increase the understanding of COPD and co-existing morbidities, to identify individual multi-morbid exacerbations, to establish evidence of effective interventions for chronic disease management, and to develop tailored referral to a multidisciplinary, adaptive eHealth programme for COPD patients with comorbidities. Predictive modelling through privacy-preserving Artificial Intelligence (AI), will increase the understanding, and evidence of effective interventions for disease management. The inclusive design and citizen science approach in RE-SAMPLE will provide credible and accepted tools and a patient-centred eHealth approach. RE-SAMPLE will act upon the need for diversified, personalized care to alleviate the overall societal and economic burden of CCCs.
Within the RE-SAMPLE project, we are developing cryptographic protocols in order to protect data confidentiality in machine learning (ML) tasks specific to managing multi-morbid complex chronic conditions.