Machine Learning-Based Stress Detection Using Smartphones and Wearable Sensors

Author(s) Information

Robert Burns, California Polytechnic State University, San Luis ObispoFollow
Kassandra Martinez-Mejia, California Polytechnic State University, San Luis ObispoFollow

College - Author 1

College of Engineering

Department - Author 1

Computer Engineering Department

College - Author 2

College of Engineering

Department - Author 2

Electrical Engineering Department

Advisor

Kun Hua, College of Engineering, Electrical Engineering Department

Funding Source

Cal Poly Electrical Engineering Department

Date

10-2024

Abstract/Summary

Stress and mental health have been considered increasing global concerns in modern society, including some of the illnesses responsible for a large proportion of comorbidities and deaths worldwide, such as depression and cardiovascular disease. Recently, smartphones, smartwatches, and smart wristbands have become an integral part of our lives and have reached widespread usage [1]. This raised the question of whether we can detect and prevent stress with smartphones and wearable sensors wirelessly. The challenge of such an approach is that the weak ECG signal is prone to errors, and the battery life of the smart devices is limited. In this project, we propose to explore the generalization capabilities of Electrocardiogram (ECG)-based deep learning models and models based on handcrafted ECG features, i.e., Heart Rate Variability (HRV) features. We will work with the students to train efficient HRV models and machine learning models that use ECG signals as input. And then apply machine learning approaches to the processed ECG data and classify their stress levels accordingly [2]. At the end of the project, we will develop a smartphone application to display, process, and analyze the collected real-time ECG signals and provide appropriate advice to the users. The project aims to create a computational model to analyze Machine Learning based Stress detection using smartphones and wearable sensors. To this end, the student will: (1) Create a computational model that calculates stress features through the ECG detection algorithms. The mentor will help the student gain a solid understanding of the model that has been developed, and code it in Matlab [4]. (2) Next, the student will implement an intelligent approach (machine learning) to the acquired ECG signals, in which the collected data will be comprehensively classified into stressed and non-stressed states [5]. (3) In the final stage, the student will use the results developed to build up a smartphone application, in which the patients' stress data can be displayed, and analyzed, and future actions will also be accordingly suggested.

October 1, 2024.