A statistical model for person with multiple chronic conditions: the ProACT project
Alessandra Pascala and Stéphane Deparis, IBM Research
12-1pm 8th Mar 2018
Abstract
The seminar will present the H2020 project ProACT and will demonstrate the contribution of the IBM Research team in developing a statistical model for the person with multimorbidities. ProACT targets Europe’s 50 million multimorbid patients to proactively self-manage and offset the EU’s annual €700billion cost of chronic disease management. ProACT aims at providing and evaluating an open application programming interface to integrate a variety of new and existing technologies to advance ‘home based’ integrated care (IC). Cloud based data analytics are developed to determine correlations between technology use and the influence of support actors to impact on the health and quality of life of patients. Development of a novel data aggregation and cloud platform system will enable data analysis for improvement of IC, effective measurement of results and comparison of efficiency and costs, so that the relationship between patients and their personalized care network is optimized. Proof of concept trials (120 patients in total, with associated care/support actors) are carried out within Health Services (Ireland and Belgium) with associated living lab facilities to ensure patient co-design technology approaches. Clinical status information, therapies and activity tools will be deployed for the conditions of: chronic heart failure (CHF), diabetes and chronic obstructive pulmonary disease (COPD).
The contribution of IBM Research is scattered over several dimensions:
- Develop an integrated, extensible model and representation of the individual, covering relevant psychological, biological, physical and social aspects in relation to patient status, therapy and activity.
- Develop a representation for analytical models that can be transferred across sites. These models will be expert-driven or data-driven and understandable both by humans and machines.
- Develop a set of analytical models centred on the person and the care ecosystem.
- Develop of the InterACT cloud
The seminar will end with a demo showing the model for the person with multimorbidities.
Short Bio
Alessandra Pascale received the B.Sc. degree (with honors) from Politecnico di Bari in 2006 and the M.Sc. degree (mark 110/110) from Politecnico di Milano in 2009, both in Telecommunication Engineering, and the Ph.D. degree in Information Technology from Politecnico di Milano in 2013. She was also enrolled in the ASP (Alta Scuola Politecnica) program, a school of excellence between Politecnico di Milano and Politecnico di Torino. She was visiting researcher at University of California, Berkeley, in the TOPL group, from February to July 2012. At the present time she works as Research Scientist at the IBM Dublin Research Lab, Ireland. Her research interests are in the field of statistics and machine learning, in particular statistical signal processing, statistical estimation and prediction theory and distributed/cooperative estimation approaches. She applied her expertises in several fields, traffic systems, wireless networks and healthcare to name a few. She has taken parts in numerous research projects, both EU funded and in direct contact with clients. She is now PI of the H2020 project ProACT focused at improving home-based self care for elderly people with multimorbidity.
Dr Stéphane Deparis is a research scientist at IBM Research lab in Dublin. His research interests are centered on Preference Modelling, Risk analysis and Decision Support. He has worked on various projects ranging from automatic graph building of medical knowledge to analysing how lay people understand visual representations of probability. He is working on a risk model to support the care of Persons with Multimorbidity for ProACT. He received a Masters degree in Optimization and Decision theory from Paris Dauphine University, and the Ingénieur Civil des Mines degree from Mines Nancy in 2008. He received a PhD from Ecole Centrale Paris in 2012 for his thesis “The effect of multicriteria conflict on the expression of preferences, an empirical approach”. He also taught preference modelling and decision aid at Ecole Centrale Paris while completing his PhD