Skip to main navigation Skip to search Skip to main content

SSA-HAN: Sleep-Stage-Aware Hierarchical Attention Network for Patient-Level Sleep Apnea Severity Assessment

  • Yanxue Yuan
  • , Kai Xing
  • , Libo Yin
  • , Hui Yang
  • , Xin Shi
  • , Liping Zhou
  • , Huanhuan Chang
  • , Ruyuan Wang
  • University of Science and Technology of China
  • Xi’an Central Hospital
  • XD Group Hospital
  • Shaanxi Second Provincial People’s Hospital
  • Xi’an Children’s Hospital
  • Xi'an Medical University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Sleep apnea severity assessment plays a central role in large-scale screening and longitudinal monitoring, but remains challenging due to pronounced intra-night variability and long-range temporal dependencies that are difficult to model with conventional detect-then-count or flat temporal approaches. In this work, we propose SSA-HAN, a sleep-stage-aware hierarchical attention network that formulates apnea-hypopnea index estimation as a structured whole-night reasoning problem. The proposed framework explicitly integrates sleep-stage context through cross-attention and aggregates information across multiple temporal scales, enabling consistent severity reasoning over long recordings. By jointly modeling fine-grained event evidence and global temporal structure, SSA-HAN produces severity-aware representations that align with clinical assessment principles. Evaluations on a private PPG-based dataset show that SSA-HAN achieves superior performance in both continuous AHI estimation and multi-class severity classification compared with representative baselines, with improved accuracy, agreement, and robustness across clinically relevant thresholds. These results suggest that incorporating sleep-stage awareness and hierarchical temporal aggregation provides an effective and clinically plausible solution for patient-level sleep apnea severity assessment.

Original languageEnglish
Title of host publicationProceedings of 2026 2nd International Conference on Artificial Intelligence and Computational Intelligence, AICI 2026
PublisherAssociation for Computing Machinery, Inc
Pages761-769
Number of pages9
ISBN (Electronic)9798400722790
DOIs
StatePublished - 19 Aug 2026
Event2026 2nd International Conference on Artificial Intelligence and Computational Intelligence, AICI 2026 - Zhengzhou, China
Duration: 6 Feb 2026 → 8 Feb 2026

Publication series

NameProceedings of 2026 2nd International Conference on Artificial Intelligence and Computational Intelligence, AICI 2026

Conference

Conference2026 2nd International Conference on Artificial Intelligence and Computational Intelligence, AICI 2026
Country/TerritoryChina
CityZhengzhou
Period6/02/26 → 8/02/26

Keywords

  • AHI estimation
  • Hierarchical attention
  • PPG-based monitoring
  • Sleep apnea assessment
  • Sleep-stage-aware learning
  • Structured temporal modeling

Fingerprint

Dive into the research topics of 'SSA-HAN: Sleep-Stage-Aware Hierarchical Attention Network for Patient-Level Sleep Apnea Severity Assessment'. Together they form a unique fingerprint.

Cite this