AggRegated System Of sensOrs and Multimedia Monitors: Technology for innovAtion and personalizaTion of rEhabilitation Care. (ROOMMATE)

Study Procedures

The study involves three successive phases:

Phase 1 During the first phase, after obtaining Ethics Committee approval, Living Labs will be planned. In these sessions, end users will test the proposed solutions and provide feedback on acceptability and usability. This phase is critical for defining user and system requirements. Additionally, health professionals will be trained as system coaches to support patients and their families in the subsequent phases.

The coaches will be selected from health professionals (physical therapists, occupational therapists, speech therapists, psychologists, nurses) with at least three professionals per center. They will complete a one-week course in digital literacy (e-literacy), covering:

• Technical and operational skills for managing and using digital technologies.
• Information and navigation skills.
• Specific training on the use of the Khymeia Home Kit.
• Training on using the inertial sensors developed by UNIFI. The training will conclude with a written and practical exam.

Subsequently, a multicenter multiple cohort RCT (cmRCT) will be conducted, consisting of two trials:

• A first RCT (Phase 2 of the study).
• A second pilot RCT (Phase 3 of the study). The study will be carried out in an inpatient setting for individuals with subacute stroke outcomes at the two clinical centers: IRCCS Fondazione Don Carlo Gnocchi in Florence, Italy, and Elias University Emergency Hospital in Bucharest, Romania.

The cmRCT design was chosen to address some limitations of traditional RCTs, such as issues with recruitment, ethics, patient preferences, and treatment comparison. Moreover, the cmRCT allows the simultaneous evaluation of different interventions.

In ROOMMATE, eligible patients who consent to baseline (T0) and final (T1, three weeks later) assessments may also be offered an additional intervention, forming the ROOMMATE Study Cohort. A subset of these patients will be randomized to receive an offer to participate in the ROOMMATE Interventions (B or B1), while those not randomized will form the control group (A, receiving usual care). This design will allow two successive RCTs to be conducted.

Phase 2

The first RCT will compare usual care (A) with a preliminary version of the ROOMMATE solution (B), which includes:

• Usual care.
• The Khymeia Home Kit monitor.
• Educational and rehabilitative content developed by clinical partners.
• Coaching for technology use. The sample size for this RCT is 140 patients, randomly assigned to Group A or B, with competitive enrollment across the two clinical centers.

The Khymeia Home Kit, supplemented with educational content, can be used by patients alone or with a caregiver. The educational materials will include:

• General information about stroke pathogenesis and prognosis.
• Strategies and advice for managing the condition.
• Health-related content, such as details on daily hospitalization routines, the roles of various care professionals, and discharge management (including bureaucratic aspects).
• Information on local support networks. Additional cognitive and motor rehabilitation content will be developed by a multidisciplinary team and customized by department therapists to match each patient's needs and recovery potential.

The coach will be available in person five days a week to support all end users in using the ROOMMATE technology.

Phase 3

Following the completion of the first RCT, a second pilot RCT with 60 patients will test usual care (A) against a mature version of the ROOMMATE solution (B1). This includes:

• Usual care.
• Motor and cognitive treatment supported by inertial sensor assessment for personalized interventions.
• Integration of clinical assessments with IMU-derived reports.
• Educational and rehabilitative content developed by clinical partners.
• Coaching for technology use. In Group B1, the Khymeia Home Kit (with educational and rehabilitation content) will be enhanced with inertial sensors developed by UNIFI. These sensors will measure hand motility and provide tailored rehabilitation interventions.

The ROOMMATE intervention will last three weeks per patient. During this period:

• Group B patients will receive training in using the technology from the coach.
• Patients and their families will have self-directed or supervised access to HKK features, including educational, rehabilitative, and infotainment content.
• Therapists will recommend the most appropriate content based on the patient's needs.
• In Group B1, the physician will use inertial sensor data to quantify motor deficits and tailor the treatment accordingly.

Assessments (T0 and T1)

At baseline (T0) and post-treatment (T1), the following will be assessed:

• Demographics: Sex, age, education.
• Clinical data: Type of stroke, time since stroke, lesion laterality.
• Primary and secondary outcomes: Disability in ADLs (modified Barthel Index, mBI). Manual dexterity. Upper limb motor skills. Neuropsychological and global cognitive functioning. Anxiety and depression. Quality of life.

At post-treatment (T1), also the following will be assessed:

• Primary and secondary outcomes: Usability, feasibility, and social impact measures. User experience and education evaluation. Wearability of sensors. Social impact and patient satisfaction.

Inclusion criteria

• outcome of ischemic or hemorrhagic stroke in subacute phase (< 3 months after the event)
• willingness to participate in the project, with informed consent signed by the subject or, when necessary, by the support administrator.

Exclusion criteria

• severe visual and/or auditory impairments that cannot be corrected
• severe cognitive and/or speech impairments that interfere with the ability to use the HKK system and the SensoMode accessory independently and/or in the
• absence of a caregiver who can assist the patient in using the system;
• skin lesions that prevent wearing the inertial sensors;
• presence of signs of clinical instability, defined by a score greater than zero on the Clinical Instability Scale (SIC).