The growing population aging with multimorbidity (MM) is a problem to the sustainability of the healthcare sector. MM is typically associated with high healthcare usage and costs, which do not always translate into better outcomes for patients. Consequently, there is a need to develop new tools to manage this condition.
The IntelligentCare project, promoted by GLSMED LEARNING HEALTH S.A., aims at developing a patient centric solution to help manage MM condition using analytical methods to explore data from the electronic health records (EHR) and the measures reported remotely by the patients, related to outcomes (PROMs) and to life events/quality of life/physical activity, named as additional value variables (AVVs), using smart sensors and mobile solutions.
Keywords: Health Data Science, Value-based healthcare, Multimorbidity
The IntelligentCare project offers an innovative approach to manage the MM condition by introducing the concept of value to the patient (patient centric) in the process of characterization and prediction of patient interactions with the hospital.
Patient-centric medicine hopes to improve health outcomes of individual patients in everyday clinical practice, taking into account the patient’s objectives, preferences, values as well as the available economic resources.
The IntelligentCare project will add value by working on advanced analytical methodologies to improve patient outcomes with MM, evaluating the healthcare delivery and contribute to hospital resources optimization, moving towards value-based healthcare.
Promoter:
GLSMED – Nuno André da Silva
Industrial Co-Promoter:
Priberam – Carlos Amaral
Hospital da Luz Lisboa – Joâo Sequeira Carlos
Academic Co-promoters:
INESC ID – Mário Gaspar Silva
Instituto Superior Técnico I ISR Lisboa – José Santos-Victor
IST ID – Alexandre Bernardino
CMU Department:
Heinz College – Pedro Ferreira
Under the value-based healthcare framework, the project will develop methods to achieve three main goals:
(1) Early signaling and characterization of patients with MM using the EHR and PROMs/AVVs (Multimorbidity phenotypes)
(2) Identification of personalized clinical pathways for each cluster based on granular evidence (approach “patients like me” and not “one size fits all”)
(3) Development of tools for monitoring and predicting hospital interactions combining EHR data with PROMs/AVVs reported systematically.
To achieve these goals research will be conducted in advanced analytical algorithms, particularly deep learning methods, in order to fully explore EHR data (structured and unstructured), PROMs and AVVs recorded via smart sensors.