Leader: Giuseppe Calcagno; Other collaborator(s):
Our proposal focuses on organizing home-based physical activity programs aimed at fostering expertise, autonomy, and improved self-efficacy among participants. The project's general goal, aligned with the objectives of the funding call, is to support physical efficiency, cognitive improvement, and overall well-being through an online cognitive and physical training program. This initiative promotes best practices to encourage the Italian population to adopt a daily active lifestyle.
The program will involve multi-component interventions, including physical exercise, cognitive and psychological training, and social interaction promotion. Active participation of individuals in planning and decision-making will ensure a person-centered approach, fostering sustainability, inclusion, equity, and social cohesion. The activities will also prioritize enjoyment to ensure long-term adherence.
Key Objectives:
Monitoring Tools:
The project will utilize an integrated platform to track participants’ vital and behavioral parameters via various market-available devices. Data will be analyzed by specialists and monitored with AI algorithms to detect health anomalies. An accompanying app will guide exercises, provide biofeedback, and ensure safety.
Implementation Protocol:
Evaluation:
Timeline:
Expected Outcomes:
The project aims to: (a) Demonstrate improvements in cognitive function, physical efficiency, and well-being. (b) Identify target groups that benefit most from this approach. (c) Develop optimal intervention strategies. (d) Explore mechanisms driving the effectiveness of home-based physical activity.
During the second phase of the project related to Spoke4-PNRR, the first "beta" version of the PRINN application’s user interface was implemented. This version introduced the possibility for users to access a personal profile, through which they can manage different features that enhance the training experience and provide more personalized support. The main objective of this version was to enhance the available options for users and incorporate essential data to optimize the training plan, adapted to the individual's specific physical conditions.
The profile section allows users to modify and update their data, including information such as:
A key functionality introduced in this phase is the connection to external devices such as heart rate monitors. This enables real-time monitoring of the user's physiological parameters, such as heart rate during physical activity. These data are displayed in the user interface, allowing the user to maintain exercise within a “safe” and personalised range. The application’s integrated algorithm uses these parameters to provide immediate feedback and corrective suggestions when necessary.
The application provides access to a training section where users can follow their personalized exercise plans and view their daily schedules. The interface displays the exercises to be performed, along with detailed instructions and demonstration videos. Furthermore, users have the option to review their previously completed training plans, which include all relevant parameters for each session (duration and intensity of the exercises, heart rate recorded during sessions, and progress towards preset goals). This feature allows users to closely monitor their performance and assess any improvements, making the training experience more interactive and personalised.
An important item introduced in the beta version is the ability to schedule video calls with the assigned coach. This feature allows users to receive real-time personalized support, discuss potential issues, or receive specific guidance on correct exercise execution. Coaches can evaluate user performance through feedback gathered by the application and provide recommendations to improve training.
Furthermore, the interface is structured to provide real-time visual and postural feedback. By utilizing the device's camera, the application monitors user movements during exercise execution, systematically comparing these movements to predefined models to identify any postural deviations or incorrect techniques. At the end of each session, users receive a comprehensive report that includes the precision score and suggested corrections.
Additionally, the system enables users to review historical performance data, facilitating comparisons between previous results and current outcomes, thereby promoting a framework for continuous and measurable improvement.
The third phase of the project related to Spoke4-PNRR focused on implementing the coach interface to facilitate comprehensive monitoring of training sessions and to ensure a personalised approach for the user.
The dashboard allows coaches to monitor users' real-time heart rate data, monitored through external sensors, to assess cardiovascular responses during training sessions. It also provides access to detailed workout histories, including metrics such as duration, intensity, and adherence to the training plan. Additionally, coaches can monitor progress toward personalized fitness goals, enabling adjustments to optimize the effectiveness of the user's training program.
The key element of this third phase was the interaction between the coach and artificial intelligence (AI).
The primary goal of this collaboration is to gradually increase user independence in performing exercises, while maintaining a high level of safety and movement quality. The integration of AI provides immediate technical support to users, enabling the coach to focus on more specific and personalized aspects of the training journey. The interaction between the coach and AI has been developed to make the user as independent as possible during exercise execution, minimizing the need for constant intervention from the coach.
The AI uses preloaded instructional videos in the application as a reference to correct the user's movements in real-time. These videos represent ideal models of exercise execution, which the AI uses to compare the user's movements and identify postural errors or incorrect techniques. During the exercise, the coach provides continuous feedback to the AI, adapting automatic corrections to the user's physical condition, specific needs, and progress.
At the end of each exercise, both the AI and the coach assign a score to the user based on the accuracy and correctness of the execution. The initial score from the AI is based on a comparison between the user's movements and the ideal models in the instructional videos. The coach enhances this evaluation by providing personalized feedback tailored to the user's specific characteristics. This combined evaluation system ensures immediate and constructive feedback, supporting the user's continuous improvement.
Over time, the AI gradually calibrates itself, adjusting its assessments and corrections based on feedback from the coach. As the AI receives more input from the coach regarding scores and necessary corrections, its evaluation system becomes increasingly precise and tailored to the user. This calibration process enables the AI to enhance its ability to autonomously correct the user, providing increasingly effective, targeted, and individualized guidance.
Brief description of the activities and of the intermediate results:
During the fourth phase of the Spoke4-PNRR project, the development of the application prototype was completed, consolidating existing features and integrating enhancements based on the feedback collected. These interventions were crucial to ensure the application effectively addresses the needs of users, providing them with an accessible and intuitive tool.
The team conducted an internal testing phase to systematically identify potential bugs, assess usability metrics, and recommend targeted modifications to streamline functionality and enhance the overall user experience. Both the posture detection system using the camera and the monitoring system through wearable devices have been optimized. Additionally, the aesthetics of the user interfaces have been improved to ensure greater ease of use for the application.
Moreover, qualified coaches with degrees in sports science were recruited and provided with specialized training on the use of the application. In particular, this training focused on advanced features, such as real-time monitoring, feedback generated by artificial intelligence, and manual feedback provided by the coach, to optimize the user’s acquisition of correct movement execution. Additionally, the training addressed the creation and customization of training plans tailored to user-provided data and individual needs.
Main policy, industrial and scientific implications:
The application could support health policies aimed at improving public health by promoting healthy and active lifestyles, which are particularly crucial for preventing chronic diseases such as obesity, diabetes, and cardiovascular conditions. Through advanced technologies, the project can be integrated into national and regional wellness programs that promote active aging, offering a preventive approach to the healthcare and social costs associated with an aging population. Additionally, the application could support public health initiatives by reducing disparities in the availability of fitness and wellness resources, ensuring greater equity and inclusivity.
The application represents an innovation in the wellness and fitness technology sector, providing advanced solutions that combine artificial intelligence, physiological monitoring devices, movement tracking for correct exercise execution, and user-friendly interfaces.
The ability to personalize training programs through data analysis, dynamic feedback, and precise movement detection ensured proper exercise execution and safety, potentially providing the application with a competitive advantage over less advanced solutions. Furthermore, the integration with external devices, such as heart rate monitors and smartwatches, could further enhance real-time monitoring and personalized adaptation capabilities.
The integration of AI-driven systems allows for the exploration of new approaches to encourage lasting behavioural changes by providing personalized and motivational feedback to users. Data collection and analysis from the app's usage could support longitudinal studies on improving the quality of life in adult and elderly populations, enhancing the understanding of the mechanisms linking physical activity, mental health, and overall well-being. Moreover, the project could open the way for the development of assistive technologies and home-based rehabilitation programs, with potential applications in different fields.
Primary Outcomes:
Secondary Outcomes:
Key Results and Impact:
The integration of AI, personalized feedback, and real-time monitoring in the PRINN application demonstrates significant progress in enhancing both the training experience and the efficiency of human-coach interactions, setting the stage for further advancements in digital training technologies.