DIGIPREDICT: physiological, behavioural and environmental predictors of asthma attacks-a prospective observational...

Area of Science:

• Digital health
• Artificial intelligence in medicine
• Asthma research

Background:

• Asthma attacks are a significant cause of morbidity and mortality.
• Early detection and prompt treatment are crucial for preventing severe asthma exacerbations.
• Subtle physiological and behavioral changes preceding attacks are often unrecognized, indicating a need for technological solutions.

Purpose of the Study:

• To identify early digital markers of asthma attacks using smart device sensors.
• To leverage health, environmental, and AI data for asthma attack prediction.
• To develop a personalized risk prediction model for early asthma attack warnings.

Main Methods:

• Prospective study of 300 participants (≥12 years) with recent moderate/severe asthma attacks.
• Utilized smartwatches, smart inhalers, peak flow meters, and cough monitoring apps over 6 months.
• Collected participant data (sociodemographics, health, technology acceptance) and environmental data (weather, air quality).

Main Results:

• Machine learning analysis of integrated datasets to develop a predictive model.
• Asthma attacks were confirmed via self-report and clinical records.
• Data analysis is ongoing to refine the risk prediction model.

Conclusions:

• The DIGIPREDICT study aims to establish a novel technological approach for asthma attack prediction.
• Early identification of digital biomarkers could lead to personalized interventions and improved asthma management.
• This research has the potential to significantly reduce asthma-related morbidity and mortality.