SmartCHANGE - AI-based long-term health risk evaluation for driving behaviour change strategies in children and youth
Persone
(Responsabile)
(Collaboratore)
Abstract
Over the last few decades, public health authorities in developed countries have witnessed a substantial shift from communicable to non-communicable chronic diseases (NCDs). Indeed, mortality estimates from the World Health Organization (WHO) indicate that more than 70% of deaths worldwide, including up to 90% of deaths in the European region, are ascribed to NCDs. The total cost of NCDs was estimated to reach 5.5 trillion € in 2010, and this amount is projected to rise to more than 12 trillion € by 2030. Most NCDs share predisposing risk factors such as obesity and low levels of physical fitness resulting from unhealthy lifestyle including insufficient physical activity, prolonged time spent in sedentary pursuits, poor nutrition, inappropriate sleep duration, cigarette smoking and abusive alcohol consumption. However, all existing tools for modelling NCDs are used in adults, mostly after 35-40 years of age, when lifestyle risk factors have already been formed. To that end, creating risk calculators suitable for children and youth to predict future NCDs, and identify high risk individuals, could importantly improve the primary prevention of NCDs.
The overall goal of the SmartCHANGE project is to develop trustworthy, AI-based decision-support tools that will help health professionals and citizens reduce long-term risk of NCDs, by accurately assessing the risk of children and youth, including those with difficult-to-detect risks, and promoting delivery of optimised risk-lowering strategies. The following objectives are designed to achieve this goal:
• Build accurate machine learning models for predicting the risk of lifetime NCD (focusing on cardiovascular and metabolic disease) for children and youth.
• Make the risk-prediction models and AI tools trustworthy
• Develop tools for both health professionals and citizens that help improve citizens’ health by using the predictive models.
• Engage users – health professionals, educators, children, and families – in requirements elicitation and participatory design from the start.
• Investigate the feasibility and usability of SmartCHANGE tools through a proof-of-concept study in four different real-world healthcare scenarios in four countries.
• Develop recommendations for the implementation of the proposed solution and similar AI-based solutions for health risk prediction in children and youth aimed at health professionals, and disseminate through relevant institutions.
• Develop an exploitation and sustainability plan for the SmartCHANGE solution and its key elements.