A new study has utilized mobile technology to monitor asthma patients and predict asthma attacks using machine learning. The AAMOS-00 observational study collected a rich multi-dimensional dataset from 22 participants located in all four nations of the UK. Patients were given smart monitoring devices (smartwatch, smart peak flow meter, and smart inhaler) to collect data daily for six months, in addition to continuing daily questionnaires. Furthermore, daily location was used to link with weather, air quality, and pollen reports in the local area, providing a comprehensive overview of each participant’s environment.
The study was conducted over several periods of national lockdowns in the UK due to the COVID-19 pandemic. Despite the challenges, the average retention in the study was 123 days, which included daily tasks, much longer than the initial estimate. However, the study’s narrow inclusion criteria meant that patients selected already had an interest in monitoring and had experienced a severe asthma attack in the past 12 months.
The AAMOS-00 study’s system utilized the Mobistudy app, an open-source platform for facilitating mobile-based studies. Patients were located across all four nations of the UK. Social media recruitment, invitation letters, and email invitations were utilized to recruit participants. Four daily tasks and two weekly tasks were assigned to each participant, with patients also carrying out a passive monitoring task for smart relief inhaler usage.
After six months of monitoring, each participant was asked to complete a questionnaire about the acceptability and usability of the system, to assess the current implementation and guide future development. The study’s results provide a valuable dataset for the development of better asthma attack prediction algorithms and demonstrate the potential of mobile technology in asthma monitoring for self-management.
