Accelerometry-Based and Gyroscope-Based Detection of Sleep Apnea Events: Systematic Review of Technical and Clinical Evidence

Sleep apnea is a common but frequently underdiagnosed disorder associated with increased cardiovascular and metabolic risk. Although polysomnography remains the diagnostic gold standard, its high cost, technical complexity, and limited availability restrict large-scale screening and continuous monitoring. Accelerometry-based and gyroscope-based approaches may represent a simplified alternative that can also be applied in home-based settings.

This systematic review aimed to evaluate the technical and clinical evidence for accelerometry-based and gyroscope-based detection of apnea and hypopnea events and to assess their potential clinical applicability.

A systematic literature search was conducted on April 12, 2025, in PubMed, Embase, Scopus, Web of Science, and IEEE Xplore following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines. Studies investigating accelerometers, gyroscopes, inertial measurement units, seismocardiography, or ballistocardiography for apnea or hypopnea detection were included. Extracted data comprised study characteristics, sensor configurations, signal processing approaches, validation strategies, diagnostic performance, and clinical applicability. Methodological quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool.

A total of 46 studies published between 2010 and 2025 were included. Most studies used accelerometers as the primary sensing modality, whereas gyroscopes were mainly applied in multimodal inertial measurement unit systems. Binary classification between apnea and nonapnea phases was the predominant analytical approach. Reported sensitivities ranged from 30% to 99%, depending on study settings, validation strategies, and evaluation levels. High diagnostic performance was mainly observed in simulation-based studies under controlled conditions. The included studies differed considerably in sensor placement, sampling frequencies, preprocessing pipelines, segmentation strategies, validation procedures, and reference standards. Only a minority of studies used polysomnography as the reference standard or included clinically representative populations. Furthermore, limited external validation, insufficient reproducibility, and a high risk of bias reduced comparability and clinical transferability.

Current evidence supports the feasibility of accelerometer-based respiratory monitoring for sleep apnea screening, particularly in wearable and home-monitoring settings. However, most available studies were conducted under controlled experimental conditions and included only limited clinical validation. Standardized validation procedures and clinically representative cohorts are necessary to improve comparability and clinical transferability.