Development of a Wearable Device for Stress Index Using Photoplethysmography Signal Analysis

EasyChair Preprint 13701

Abstract

This study focuses on developing a device that can measure PPG signals while wearing a wearable band and measuring HRV-based stress index using PPG signals measured at the wrist. We propose an evaluation method for HRV parameters using deep learning of PPG signals. An integrated model of 1D Convolutional Neural Networks (1DCNN) and Residual Networks (ResNet) are used to learn a model for the parameters. The model was uploaded to a microprocessor through quantization for edge computing and calculated into a stress index of the PPG signals. The device uses an ESP32 microprocessor and a MAX30105 PPG sensor module to measure PPG signals. This study aims to implement a system that optimizes edge device computing algorithms through PPG measurement in wearable devices. This study aims to implement a system that optimizes the stress index algorithm through edge device computing of PPG signals measurement on wearable devices.

Keyphrases: Edge device computing, Photoplethysmography, heart rate variability, stress index

@booklet{EasyChair:13701,
  author    = {Shinhye Kim and Donghwan Hwang},
  title     = {Development of a Wearable Device for Stress Index Using Photoplethysmography Signal Analysis},
  howpublished = {EasyChair Preprint 13701},
  year      = {EasyChair, 2024}}