This project will implement three key functionalities to transform MIDATA into a full-featured and universal platform for patient-controlled data aggregation, consent management, and analytics. 1) Efficient interfaces between MIDATA and existing e-health platforms and clinical information systems to obtain a comprehensive data set across traditional and novel data sources and to facilitate the development of new tools for risk assessment and treatment decisions. 2) Machine Learning techniques to make personalized predictions concerning each patient’s disease progression. For this purpose, algorithms capable of finding complex patterns in aggregated data (from mobile apps, medical reports, and environmental data) will be developed. 3) Systemic ethical oversight for data-driven personalized health research. We will create a toolbox of processes, mechanisms and governance solutions for the MIDATA platform. Systemic oversight principles will define the tools tailored to patient-driven data aggregation. To ensure real-world utility and compatibility with daily clinical and research practice, these developments and governance infrastructure will be implemented and tested in the framework of a clinical research study (MS) carried out at the University Hospital in Zurich.
Multiple sclerosis is an autoimmune disease associated with recurring inflammation of brain and spinal cord. Worldwide there are over a million cases of MS with around 10’000 affected individuals in Switzerland. MS is especially suited to tackle the challenges arising with the acquisition and aggregation of health-related personal data and to probe the development of new tools for prediction of disease evolution since an interplay of genetic, environmental and lifestyle factors strongly influence disease course. MS is highly heterogeneous in clinical presentation and shows strong variability in disease course and response to different treatments. MS is ideal to establish personalized approaches that aim to integrate all available patient-related information to predict context-dependent disease progression and guide treatment decisions for individual patients.
