Architecture
The architecture of the GAME2AWE platform is illustrated bellow. The Device Layer is the first level of the platform architecture, where external devices communicate with the system. The information collected by the devices is forwarded to the Data Layer where it is modeled using machine learning algorithms. The information is then classified by the Analysis Layer for the purpose of accepting or rejecting an action/movement. In case of acceptance, information is perceived by the user through Game Mechanics. The game mechanisms inform the system about the progress of the user in the gameplay in order to make the corresponding adaptation of the game parameters. The Adaptation Layer provides the system intelligence to create adaptable gaming experiences by adjusting the parameters and elements of each game. Finally, the Interaction Layer provides the user interfaces.

Platform Assessment
The evaluation of the platform includes 3 axes: kinetic, cognitive and technological. A total of 60 users are required, whose participation is based on age (65 years) and health criteria (possibility of independent movement, absence of serious health problems). The methodology includes a control group and an intervention group applying a randomized controlled trial. For all users, measurements of motor and cognitive functions will be recorded before using the GAME2AWE platform, while after the intervention, the same measurements will be recorded only for the intervention team. The intervention team will use the GAME2AWE platform 1-2 times a week until each participant completes 24 user sessions.
Valid relative tests / scales will be used for the kinetic axis such as: Choice Stepping Reaction Time (CSRT), Berg Balance Scale (BBS), Functional Reach Test (FRT), 30 Second Sit to Stand Test (30SST), Motor Fitness Scale and Time Up and Go (TUG). The Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA) scales will be used for the cognitive axis.
At the level of technological acceptance of the platform by users, parameters such as Perceived Usefulness, Perceived Ease of Use and Output Quality (Technology Acceptance Model 3) are evaluated. The Physical Activity Enjoyment Scale (PACES) will be used to assess user satisfaction with the use of the platform. Qualitative research will be conducted through semi-structured interviews of users participating in the pilot test. The research will subjectively determine the measure of improving the quality of life.
Machine Learning
During the use of the platform, the system collects and stores performance data such as the score of each game, time of interaction with the game screens, frequency of game use, games completed successfully, number of errors, etc. The stored data is utilized by machine learning algorithms to determine more effective exercise programs to strengthen balance and strength to avoid falls but also to strengthen cognitive skills. In addition, the discovery of patterns in the stored data helps to classify the elderly at diagnostic levels with the aim of replacing traditional diagnostic tools.