Tiles Study

TILES (Tracking Individuals Performance with Sensors) was a project in collaboration with University of Southern California and University of Washington. The goal of the project is to try and model/predict job performance and other psychological states relevant to quality-of-life of nurses. The motivation comes from high reported burnout cases in the nurse population caused by long working hours and stressful situations faced by hospital staff. This added with the fact that any lapse in performance from healthcare providers can lead to critical conditions for patients makes remote performance monitoring an important and interesting research problem.

Studies done in controlled laboratory environments have shown the capability of monitoring emotions and performance using physiological signals such as inter-beat interval series, breathing signals, electrodermal activity etc. However, these studies have not translated well into real world conditions. This is due to various reasons ranging from noisy sensor data to lack of context on emotional/physical stimulus being provided to the subject.

The study collects physiological and other data such as location and audio from more than 200 hospital employees while they work on their daily tasks. My role in the project is using the inter-beat interval and breathing data provided by the smart shirt to derive heart rate and breathing features that are:

• Robust of instances of missing data and noise

• Provide more information about the psychological states of individuals

I have done this by deriving various complexity based features for heart rate analysis as well as complexity based features on the inter-breath interval and inhale-to-exhale ratio series derived from the breathing signals

Publications:

1. Abhishek Tiwari, Shrikanth Narayanan and Tiago H. Falk, Breathing Rate Complexity Features for “In-the-Wild” Stress and Anxiety Measurement, 27th European Signal Processing Conference, EUSIPCO 2019

2. Abhishek Tiwari, Shrikanth Narayanan and Tiago H. Falk, Stress and Anxiety Measurement “In-the-Wild” Using Quality-aware Multi-scale HRV Features, 41st International Engineering in Medicine and Biology Conference, EMBC 2019

3. Abhishek Tiwari, Raymundo Cassani, Shrikanth Narayanan and Tiago H. Falk, A Comparative Study of Stress and Anxiety Estimation in Ecological Settings Using a Smart-shirt and a Smart-bracelet, 41st International Engineering in Medicine and Biology Conference, EMBC 2019

4. Tiwari, A., Villatte, J. L., Narayanan, S., & Falk, T. H. . Prediction of Psychological Flexibility with multi-scale Heart Rate Variability and Breathing Features in an “in-the-wild” Setting. In 2019 8th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW) (pp. 297-303). IEEE.