Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimizing Photoplethysmography-Based Sleep Staging Models by Leveraging Temporal Context for Wearable Devices Applications

Joseph A. P. Quino, Diego A. C. Cardenas, Marcelo A. F. Toledo, Felipe M. Dias, Estela Ribeiro, Jose E. Krieger, Marco A. Gutierrez

TL;DR

This work proposes an adapted sleep staging model based on top-performing state-of-the-art methods and evaluates its performance with different PPG segment sizes and concatenate 30-second PPG segments over 15-minute intervals to leverage longer segment contexts.

Abstract

Accurate sleep stage classification is crucial for diagnosing sleep disorders and evaluating sleep quality. While polysomnography (PSG) remains the gold standard, photoplethysmography (PPG) is more practical due to its affordability and widespread use in wearable devices. However, state-of-the-art sleep staging methods often require prolonged continuous signal acquisition, making them impractical for wearable devices due to high energy consumption. Shorter signal acquisitions are more feasible but less accurate. Our work proposes an adapted sleep staging model based on top-performing state-of-the-art methods and evaluates its performance with different PPG segment sizes. We concatenate 30-second PPG segments over 15-minute intervals to leverage longer segment contexts. This approach achieved an accuracy of 0.75, a Cohen's Kappa of 0.60, an F1-Weighted score of 0.74, and an F1-Macro score of 0.60. Although reducing segment size decreased sensitivity for deep and REM stages, our strategy outperformed single 30-second window methods, particularly for these stages.

Optimizing Photoplethysmography-Based Sleep Staging Models by Leveraging Temporal Context for Wearable Devices Applications

TL;DR

This work proposes an adapted sleep staging model based on top-performing state-of-the-art methods and evaluates its performance with different PPG segment sizes and concatenate 30-second PPG segments over 15-minute intervals to leverage longer segment contexts.

Abstract

Accurate sleep stage classification is crucial for diagnosing sleep disorders and evaluating sleep quality. While polysomnography (PSG) remains the gold standard, photoplethysmography (PPG) is more practical due to its affordability and widespread use in wearable devices. However, state-of-the-art sleep staging methods often require prolonged continuous signal acquisition, making them impractical for wearable devices due to high energy consumption. Shorter signal acquisitions are more feasible but less accurate. Our work proposes an adapted sleep staging model based on top-performing state-of-the-art methods and evaluates its performance with different PPG segment sizes. We concatenate 30-second PPG segments over 15-minute intervals to leverage longer segment contexts. This approach achieved an accuracy of 0.75, a Cohen's Kappa of 0.60, an F1-Weighted score of 0.74, and an F1-Macro score of 0.60. Although reducing segment size decreased sensitivity for deep and REM stages, our strategy outperformed single 30-second window methods, particularly for these stages.
Paper Structure (10 sections, 5 figures, 1 table)

This paper contains 10 sections, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Dataset
  4. Preprocessing
  5. Super Windows
  6. SleepPPG-Net adaptation
  7. Training and Evaluation
  8. Results
  9. Discussion
  10. Conclusion

Figures (5)

  • Figure 1: PPG signal split into 30-second windows.
  • Figure 2: Block representations super windows. Each window W represents 30 seconds of PPG signal.
  • Figure 3: Feature extraction in the original and adapted model. The red arrow indicates skip connection and the green arrow indicates convolution output.
  • Figure 4: Based on kotzen_sleepppg-net_2023. Diagram of the adapted SleepPPG-Net model. $Q$ is the number of windows in the input.
  • Figure 5: Confusion matrix of the adapted SleepPPG-Net with different configurations.