Open Access

ARTICLE

CGB-Net: A Novel Convolutional Gated Bidirectional Network for Enhanced Sleep Posture Classification

Hoang-Dieu Vu^1,2, Duc-Nghia Tran³, Quang-Tu Pham¹, Ngoc-Linh Nguyen^4,*, Duc-Tan Tran^1,*

1 Faculty of Electrical and Electronic Engineering, Phenikaa School of Engineering, Phenikaa University, Yen Nghia, Hanoi, 12116, Vietnam
2 Graduate University of Sciences and Technology, Vietnam Academy of Science and Technology, Hanoi, 100000, Vietnam
3 Institute of Information Technology, Vietnam Academy of Science and Technology, Hanoi, 10072, Vietnam
4 International School, Vietnam National University, Hanoi, 10000, Vietnam

* Corresponding Authors: Ngoc-Linh Nguyen. Email: ; Duc-Tan Tran. Email:

Computers, Materials & Continua 2025, 85(2), 2819-2835. https://doi.org/10.32604/cmc.2025.068355

Received 26 May 2025; Accepted 29 July 2025; Issue published 23 September 2025

Abstract

This study presents CGB-Net, a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer, with direct applicability to gastroesophageal reflux disease (GERD) monitoring. Unlike conventional approaches limited to four basic postures, CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions, providing enhanced resolution for personalized health assessment. The architecture introduces a unique integration of three complementary components: 1D Convolutional Neural Networks (1D-CNN) for efficient local spatial feature extraction, Gated Recurrent Units (GRU) to capture short-term temporal dependencies with reduced computational complexity, and Bidirectional Long Short-Term Memory (Bi-LSTM) networks for modeling long-term temporal context in both forward and backward directions. This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data, surpassing the performance of simpler or previously published hybrid models. Experiments were conducted on a benchmark dataset consisting of 18 volunteers (age range: 19–24 years, mean 20.56 1.1 years; height 164.78 8.18 cm; weight 55.39 8.30 kg; BMI 20.24 2.04), monitored via a single abdominal accelerometer. A subject-independent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability. The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%, both reported with standard deviations over multiple runs, outperforming several baseline and state-of-the-art methods. By releasing the dataset publicly and detailing the model design, this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.

---

Keywords

---