SleepLiteCNN: Energy-Efficient Sleep Apnea Subtype Classification with 1-Second Resolution Using Single-Lead ECG

2025-07-31

AI Summary

To address the need for real-time, low-power monitoring of sleep apnea (SA) subtypes—obstructive, central, and mixed—on wearable devices, this paper proposes SleepLiteCNN, a lightweight convolutional neural network specifically designed for single-lead ECG signals to enable continuous, second-level subtype classification. Methodologically, it processes 1-second ECG segments as input and benchmarks performance against both classical machine learning and deep learning baselines. To optimize for edge deployment, SleepLiteCNN integrates 8-bit quantization with FPGA hardware co-design, significantly reducing energy consumption and resource utilization without compromising accuracy. Experimental evaluation on public datasets achieves 95.2% classification accuracy and 92.1% macro-F1 score, with only 1.8 μJ inference energy per sample and over 40% reduction in FPGA resource usage. This work delivers a high-accuracy, ultra-low-power solution for real-time SA subtype screening at the edge.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Apnea is a common sleep disorder characterized by breathing interruptions lasting at least ten seconds and occurring more than five times per hour. Accurate, high-temporal-resolution detection of sleep apnea subtypes - Obstructive, Central, and Mixed - is crucial for effective treatment and management. This paper presents an energy-efficient method for classifying these subtypes using a single-lead electrocardiogram (ECG) with high temporal resolution to address the real-time needs of wearable devices. We evaluate a wide range of classical machine learning algorithms and deep learning architectures on 1-second ECG windows, comparing their accuracy, complexity, and energy consumption. Based on this analysis, we introduce SleepLiteCNN, a compact and energy-efficient convolutional neural network specifically designed for wearable platforms. SleepLiteCNN achieves over 95% accuracy and a 92% macro-F1 score, while requiring just 1.8 microjoules per inference after 8-bit quantization. Field Programmable Gate Array (FPGA) synthesis further demonstrates significant reductions in hardware resource usage, confirming its suitability for continuous, real-time monitoring in energy-constrained environments. These results establish SleepLiteCNN as a practical and effective solution for wearable device sleep apnea subtype detection.

Problem

• Classify sleep apnea subtypes with 1-second ECG resolution
• Optimize energy efficiency for wearable device deployment
• Balance accuracy and computational complexity in real-time detection

Innovation

• Single-lead ECG for apnea subtype classification
• Compact CNN optimized for wearable energy efficiency
• FPGA synthesis enables real-time low-power monitoring