31 Usability testing of an artificial intelligence enhanced mobile application for home-based cardiac rehabilitation

<h3>Background</h3> Home-based cardiac rehabilitation may be facilitated through mobile applications powered by artificial intelligence (AI). <h3>Aim</h3> RECAP trial Phase I assessed the usability of a home-based CR programme delivered using a mobile app powered by AI to prescribe exercise goals. <h3>Methods</h3> A mobile application and a web portal were developed to facilitate home based cardiac rehabilitation. RECAP Trial Phase I was a usability study aimed at assessing the functionality of the technology. Using consecutive sampling, data was collected from 15 patients. <h3>Results</h3> Phase I of the study was conducted with 15 patients, age range 49 to 77. Three female and 12 male. Two patients post coronary artery bypass grafting surgery and 13 post ACS. Ten participants had six-minute walk test (6 MWT ) data. The patients improved their mean 6MWT distance from 393.5m (initial visit) to 550.0m (final visit). The average System Usability Scale (a scale used to assess the usability of a system) score for the app was 82/100, receiving an overall rating of ‘Excellent’. For 15 participants, the mean step count in week one was 35876 (SD 26133) and in week eight 29169 (SD 15403). On average, patients were able to achieve physical activity goals close to those prescribed for them (figure 1). During the qualitative interviews following Phase I, patients stated that they felt motivated to do more each day because the accelerometers monitored their daily physical activity levels. Additionally, patients reported feeling safer as they felt they were being monitored at home. <h3>Conclusion</h3> We found the mobile application to be usable and well-received by the patients in Phase I of the study. Patients found the goals practical and felt motivated to do more.