AI-powered remote personalized patient rehabilitation models for balance disorders

The HOLOBALANCE project, finalized in 2021, developed wearables and IoT devices for balance rehabilitation and remote patient monitoring. The EU-funded TeleRehaB DSS project will develop an AI-based decision support system (DSS) building on the existing Augmented Reality (AR) rehabilitation training platform and the tools from the HOLOBALANCE project.

TeleRehaB DSS targets the promotion of AI adoption in everyday clinical practice for balance rehabilitation training. An AI-based decision support system (DSS) will be developed expanding upon the existing Augmented Reality (AR) rehabilitation training platform, with its balance exercises, exergames, cognitive training and remote patient monitoring with wearables and IoT devices from HOLOBALANCE project (TRL6), to provide suggestive feedback for experts through the entire clinical rehabilitation pathway. The first component of AI models of TeleRehaB DSS will assess prognostic factors for risk of falls, treatment effectiveness, outcomes and side effects at baseline level, using a high volume of retrospective data for initial training. The other AI pillar of TeleRehaB DSS will introduce automated balance intervention planning and management functionality. The DSS will provide for each patient an optimal set of personalized rehabilitation activities, considering the best clinically effective treatment in conjunction with socio-economic effectiveness, and eHealth literacy. The latter will be evaluated with a quick and easy to use tool with simple tasks to assess patient’s level of technological awareness (i.e., use of smart devices, AR and IoT equipment), in order to predict how this is going to affect compliance and adherence with interventions that rely on the use of such novel technologies. Finally, the most beneficial use of AI in TeleRehaB DSS will consist of automated remote patient monitoring with wearables and IoT sensing devices, allowing rehabilitation training programs to be performed at home. The DSS will evaluate in real-time patient performance, symptoms occurrence with virtual AR physio’s providing corrective and motivational feedback as activities are performed. These performance evaluation measures will be fed back to the DSS to support experts with their most time- and effort-consuming activities of day-to-day patient management.

Key Aims & Advantages

  • Holistic Intervention: Addresses complex balance disorders by considering multifactorial data, including medical history, frailty, medication, and physical activity levels.
  • AI-Driven Personalization: Uses Machine Learning to tailor intervention types, intensity, and progression specifically to each patient’s risk profile and needs.
  • Advanced Remote Care: Brings clinical-grade rehabilitation into the home using AR avatars and wearable sensors, ensuring high-quality care for those in remote or underserved areas.
  • Economic Impact: Designed to significantly reduce healthcare costs by preventing falls and minimizing hospital re-admissions.
  • Proven Technology: Expands on the validated technologies of the HOLOBALANCE project to ensure reliability and user acceptance.

TeleRehaB DSS: eHealth Literacy Tool

Within the framework of the project, our laboratory, in collaboration with the National and Kapodistrian University of Athens (NKUA), has developed a pivotal tool for assessing users’ eHealth technological readiness. While this tool is being evaluated and optimized within the scope of the current project, it holds the potential to be extended to a wide range of eHealth applications.

  • Digital Proficiency Assessment: Evaluates patients’ digital skills regarding IoT, AR, and smart devices to personalize the rehabilitation program.
  • Interactive Testing: Incorporates interactive tasks (e.g., swipe, tap, drag & drop, typing, following arrows) to measure execution time, accuracy, and touch-screen interaction skills.
  • Performance Metrics: Captures critical data such as reaction time, fine motor control (finger dexterity), typing proficiency, and adherence to instructions.
  • Standardized Questionnaires: Integrates validated scales (eHEALS, STAM, MDPQ) to assess technology acceptance and mobile device proficiency.
  • Objective: To enhance adherence and minimize barriers to digital interventions, particularly for older adults.