Periodic Reporting for period 3 - OCARIoT (smart childhood Obesity CARing solution using IoT potential)
Période du rapport: 2019-11-01 au 2021-04-30
Child obesity is the major pediatric public health concern, affecting around 224 million school-age children in the world. Its prevalence has tripled in many European countries since 1980, increasing in an alarming rate. Childhood obesity already affects more than one in three school-aged children in Brazil, Greece and Spain. Childhood is an important period for forming healthy behaviours in order to reduce obesity inequities. OCARIoT will promote the improvement of eating and physical disorders and also the prevention of the obesity onset for children (between 9 and 12 years old), the main objective of OCARIoT is to provide an IoT-based personalised coaching solution guiding children to adopt healthy eating and physical activity behaviour. The IoT network will allow us to observe child activity patterns of daily living, health evolution, physiological & behavioural parameters and environmental data.
The general objective is broken down into several objectives.
STO1: Integrate, secure and real-life test several IoT sensors, devices and wearables to capture data about eating behaviour, activity patterns, and physiological signals for enabling a smart assisted living environment that supports a Healthy and Active Lifestyle targeting childhood obesity.
STO2: Create obesity models by correlating behavioural patterns (for eating and physical activity mainly) detected on specific recordings of human generated signals to the risk of the development or aggravation of obesity.
STO3: Provide a Personalised IoT-based obesity-care (nutritional and sport) guidance System and IoT-based interventions to help and train children to improve their eating and activity behaviour by detecting subjects at risk for developing obesity or eating disorders and offering them enhanced monitoring and guidance in order to prevent further disease progression.
STO4: Develop the OCARIoT platform to enable a bi-directional and easy communication between users and the system.
STO5: Provide the integrated platform in order to demonstrate the efficiency of the proposed solution validating in two pilots in Europe and one pilot in Brazil with children between 9 and 12 years-old.
The main findings after piloting confirm the OCARIoT ecosystem is capable of assessing the behaviours of children, motivating and guiding them towards achieving personal behavioural goals. Although it is demonstrated that the environmental status can directly influence our health, at this first stage it is still not possible to relate the data collected to the quality of health of children in relation to their physical activities due to problems related to COVID-19, as the data were collected in their absence in schools. Finally, the feedback provided by families and children revealed that the OCARIoT solution created a high positive impression and satisfaction from UX and technology acceptance perspective. However, the OCARIoT solution needs further refinement to become a commercial tool, and that has been considered in the exploitation and business model.
The general objective is broken down into several objectives.
STO1: Integrate, secure and real-life test several IoT sensors, devices and wearables to capture data about eating behaviour, activity patterns, and physiological signals for enabling a smart assisted living environment that supports a Healthy and Active Lifestyle targeting childhood obesity.
STO2: Create obesity models by correlating behavioural patterns (for eating and physical activity mainly) detected on specific recordings of human generated signals to the risk of the development or aggravation of obesity.
STO3: Provide a Personalised IoT-based obesity-care (nutritional and sport) guidance System and IoT-based interventions to help and train children to improve their eating and activity behaviour by detecting subjects at risk for developing obesity or eating disorders and offering them enhanced monitoring and guidance in order to prevent further disease progression.
STO4: Develop the OCARIoT platform to enable a bi-directional and easy communication between users and the system.
STO5: Provide the integrated platform in order to demonstrate the efficiency of the proposed solution validating in two pilots in Europe and one pilot in Brazil with children between 9 and 12 years-old.
The main findings after piloting confirm the OCARIoT ecosystem is capable of assessing the behaviours of children, motivating and guiding them towards achieving personal behavioural goals. Although it is demonstrated that the environmental status can directly influence our health, at this first stage it is still not possible to relate the data collected to the quality of health of children in relation to their physical activities due to problems related to COVID-19, as the data were collected in their absence in schools. Finally, the feedback provided by families and children revealed that the OCARIoT solution created a high positive impression and satisfaction from UX and technology acceptance perspective. However, the OCARIoT solution needs further refinement to become a commercial tool, and that has been considered in the exploitation and business model.
The main achievements for this period can be summarized as follows:
- Demonstration that the OCARIoT ecosystem is capable of assessing the behaviours of children, motivating and guiding them towards achieving personal behavioural goals.
- Comparison between self-reported and accelerometer-measured physical activity in young versus older children showing that the use of wearable data for physical activity assessment should be preferred in younger children.
- Slope trend service for managing intelligent algorithms.
- Development and Integration in the OCARIoT dashboard as a tool to calculate and present the evolution of child BMI values according to weight, height, age and gender.
- Combined use of Bootstrap and Material style libraries.
- Inclusion of alternative measures for obesity categorization (waist circumference, body composition) liable to be measured by IoT.
- Collection of health variables from a triple perspective: the subject (child), family, school. With tools and interface personalised to each case.
- Overcoming ethical constraints for child user profiling on IoT devices, RFID recognition for IoT weight scale usage.
- Machine Learning models to predict obesity condition/non-healthy nutritional status and
- Next-day sedentary behaviour of children based on step counts can be predicted with 77% sensitivity and 70% accuracy, thereby providing the ground for realizing novel predictive capabilities in digital health systems.
- Applicability towards social robot-based food detection as well as coaching and health behaviour change in real-life contexts.
- Patent application: Security and Privacy Assessment Method and System for Acquisition of Hardware and Software Components for IoT Systems. Patent filed - INPI Process # BR 10 2020009179 4;
- Patent application: Multiple Layer Vulnerability Testing and Assessment Method and System for IoT Systems. Patent filed - (In Progress).
- Demonstration that the OCARIoT ecosystem is capable of assessing the behaviours of children, motivating and guiding them towards achieving personal behavioural goals.
- Comparison between self-reported and accelerometer-measured physical activity in young versus older children showing that the use of wearable data for physical activity assessment should be preferred in younger children.
- Slope trend service for managing intelligent algorithms.
- Development and Integration in the OCARIoT dashboard as a tool to calculate and present the evolution of child BMI values according to weight, height, age and gender.
- Combined use of Bootstrap and Material style libraries.
- Inclusion of alternative measures for obesity categorization (waist circumference, body composition) liable to be measured by IoT.
- Collection of health variables from a triple perspective: the subject (child), family, school. With tools and interface personalised to each case.
- Overcoming ethical constraints for child user profiling on IoT devices, RFID recognition for IoT weight scale usage.
- Machine Learning models to predict obesity condition/non-healthy nutritional status and
- Next-day sedentary behaviour of children based on step counts can be predicted with 77% sensitivity and 70% accuracy, thereby providing the ground for realizing novel predictive capabilities in digital health systems.
- Applicability towards social robot-based food detection as well as coaching and health behaviour change in real-life contexts.
- Patent application: Security and Privacy Assessment Method and System for Acquisition of Hardware and Software Components for IoT Systems. Patent filed - INPI Process # BR 10 2020009179 4;
- Patent application: Multiple Layer Vulnerability Testing and Assessment Method and System for IoT Systems. Patent filed - (In Progress).
The innovation potential of OCARIoT to improve the reduction of obesity rate in school-aged children from pilot sites by using the OCARIoT platform can be summarised as:
Health status monitoring with IoT sensors, devices and wearables
- Usage of universAAL IoT platform to enable the acquisition of data and consumption of services by sharing the compatible models (ontologies) that describe the measures and parameters we are interested in. At the moment of pilot implementation, the best IoT/wearable solution available in the market for a massive usage will be included into the universAAL IoT platform implementing the service model represented by its measurement.
- Integration of different heterogeneous IoT sources (including other IoT platforms such as FiWare) to enrich such data within semantic annotations and to infer new knowledge within reasoning algorithms. The awareness of the current systems’ situation and truthfulness of the data will be crucial for ensuring a precise decision making and algorithms performance.
- Enabling a close collaboration between the main actors of the OCARIoT platform, such as children, professional healthcare staff, educational staff and families.
Predictive models for preventing overweight and obesity problems
- Build upon the data, reviews and overall findings both on childhood obesity factors as well as prevention interventions derived from past clinical and technical studies.
- Integrate that historical data on childhood obesity with new evidence derived within the OCARIoT project in order to develop predictive models for preventing overweight and obesity problems
- Predictive models operating on the basis of multi-parametric user monitoring data automatically collected through the OCARIoT’s envisioned IoT infrastructure.
Gamification and user engagement
- Provision of an app specifically for children in order to attract them in its use and with a clear focus on treating obesity health problems.
- Combination of personalised health interventions with a gamification strategy addressing the specific needs of children with overweight and obesity problems.
- Enabling a motivational self-management process adapted to children’ intrinsic and extrinsic motivators.
Health status monitoring with IoT sensors, devices and wearables
- Usage of universAAL IoT platform to enable the acquisition of data and consumption of services by sharing the compatible models (ontologies) that describe the measures and parameters we are interested in. At the moment of pilot implementation, the best IoT/wearable solution available in the market for a massive usage will be included into the universAAL IoT platform implementing the service model represented by its measurement.
- Integration of different heterogeneous IoT sources (including other IoT platforms such as FiWare) to enrich such data within semantic annotations and to infer new knowledge within reasoning algorithms. The awareness of the current systems’ situation and truthfulness of the data will be crucial for ensuring a precise decision making and algorithms performance.
- Enabling a close collaboration between the main actors of the OCARIoT platform, such as children, professional healthcare staff, educational staff and families.
Predictive models for preventing overweight and obesity problems
- Build upon the data, reviews and overall findings both on childhood obesity factors as well as prevention interventions derived from past clinical and technical studies.
- Integrate that historical data on childhood obesity with new evidence derived within the OCARIoT project in order to develop predictive models for preventing overweight and obesity problems
- Predictive models operating on the basis of multi-parametric user monitoring data automatically collected through the OCARIoT’s envisioned IoT infrastructure.
Gamification and user engagement
- Provision of an app specifically for children in order to attract them in its use and with a clear focus on treating obesity health problems.
- Combination of personalised health interventions with a gamification strategy addressing the specific needs of children with overweight and obesity problems.
- Enabling a motivational self-management process adapted to children’ intrinsic and extrinsic motivators.