What is the clinical evidence from this study?

Study design: Observational. Population: Perceived stress (n=8). Intervention: HRV sensors and accelerometers. Primary outcome: Prediction accuracy of perceived stress levels.

August 1, 2015

Modeling perceived stress via HRV and accelerometer sensor streams

Key Result

Integrating HRV and accelerometer sensor data with a Bagging classifier achieved a prediction accuracy of 85.7% for monitoring perceived stress levels in daily life.

Study Design

Type

Observational (n=8)

Structured PICO

Does integrating HRV and accelerometer data accurately predict perceived stress levels in daily life?

Population

8 participants monitored continuously for about 2 weeks to assess perceived stress levels using HRV and accelerometer sensors.

Exposure

Integration of HRV sensors and accelerometers with machine learning (Bagging classifier with feature selection)

Outcome

Prediction accuracy of perceived stress levels

Integrating HRV and accelerometer data using machine learning can predict perceived stress levels in daily life with 85.7% accuracy.

Abstract

Discovering and modeling of stress patterns of human beings is a key step towards achieving automatic stress monitoring, stress management and healthy lifestyle. As various wearable sensors become popular, it becomes possible for individuals to acquire their own relevant sensory data and to automatically assess their stress level on the go. Previous studies for stress analysis were conducted in the controlled laboratory and clinic settings. These studies are not suitable for stress monitoring in one's daily life as various physical activities may affect the physiological signals. In this paper, we address such issue by integrating two modalities of sensors, i.e., HRV sensors and accelerometers, to monitor the perceived stress levels in daily life. We gathered both the heart and the motion data from 8 participants continuously for about 2 weeks. We then extracted features from both sensory data and compared the existing machine learning methods for learning personalized models to interpret the perceived stress levels. Experimental results showed that Bagging classifier with feature selection is able to achieve a prediction accuracy 85.7%, indicating our stress monitoring on daily basis is fairly practical.

Bookmark