Clinical Assessment
Add-on Module
About
The Clinical Assessment add-on module includes the following functionality:
Health Advisory
The health advisory add-on feature allows for personalised monitoring of health data trends on a daily, weekly, and monthly basis. This supports early identification of modifications in individual data patterns which may help users recognise emerging changes in health and wellbeing.
The system compares current measures in physiological and behavioural data against up-to-date baseline averages, detecting any mild or moderate deviations outside of these personalised ranges. A user is able to assess both short- and long-term changes in the data, taking advantage of our advisory support to uncover meaningful interpretations in patterns and trends.
Recovery Score
This add-on feature enables a user to assess how efficiently an individual recovers from periods of elevated stress on a daily basis using the Garmin-specific stress metric. This facilitates the monitoring of short-term recovery profiles, as well as longer-term changes in recovery patterns that may indicate chronic stress development.
The daily score works by incorporating the proportion of time spent in states of recovery, the frequency of stressful episodes experienced, the efficiency of recovery following these episodes, and the speed at which recovery is reached. Importantly, the Garmin stress metric is not recorded if physical activity is detected. Therefore, a corresponding daily confidence score is provided to indicate the level of data completeness, allowing the user to assess any potential impact of missing data on the output.
Drowsiness Detection
This add-on feature facilitates personalised drowsiness risk estimation using physiological and behavioural data collected from a smartwatch. This tool can help users make informed decisions about individual alertness, supporting the management of safety and performance. Importantly, this feature takes advantage of Garmin-specific metrics such as stress and body battery and is not yet applicable to other wearable brands.
The algorithm considers: 1) the individual’s recent history of drowsiness; 2) personal mean averages and deviations in a range of signals, including heart rate and stress, updated on a rolling basis; and 3) the time of day. These factors are combined using a Bayesian model to estimate drowsiness risk within a 45-minute window.
