A Clinically Informed Cross-Attention Deep Learning Framework for Automated Diagnosis of Obstructive Sleep Apnea
Abstract
Obstructive sleep apnea (OSA) is a highly prevalent breathing-related sleep disorder associated with a variety of cardiovascular comorbidities. Polysomnography (PSG) is the clinical gold standard for OSA diagnosis, with disease severity quantified by the apnea hypopnea index (AHI). However, deriving AHI requires manual annotation of overnight PSG recordings by certificated technicians, making the OSA diagnostic process costly and time-consuming. While it is highly desirable to automate this process with a reliable machine learning pipeline, conventional models often rely on handcrafted feature engineering and may therefore not sufficiently capture the complex dynamics of physiological signals. Deep learning models can learn nonlinear representations directly from raw PSG signals, but many existing works study PSG independently and give less consideration to clinical variables that are known to influence OSA risk, e.g., age, sex. More importantly, most prior research has focused on a single dataset, raising concerns about the robustness and generalizability of these models. To tackle these limitations, this project proposes a clinical knowledge-guided cross-attention deep learning framework for automated PSG annotation and AHI prediction from overnight PSG recordings, combined with clinical variables. The proposed framework employs a deep encoder to learn epoch-level features from PSG data and classify epochs based on whether an adverse respiratory event occurs, providing epoch-level PSG annotations that mimic manual evaluation in current practice. Then, the epoch-level embeddings will be integrated with clinical variables using a multi-head attention mechanism for patient-level AHI prediction. The framework is validated across four large real-world datasets.
Related
- KGRD: a knowledge-graph-augmented automated reasoning framework for diagnosis and counselling of paediatric rare genetic disorders
- A federated attention-based stacked LSTM framework for interpretable malaria diagnosis under simulated non-IID federated conditions
- A multi task deep learning framework for automated health condition prediction from retinal fundus images using hybrid attention and explainable AI
- Tiotropium and clinically important deterioration in mild-to-moderate chronic obstructive pulmonary disease: a post hoc win ratio analysis of the Tie-COPD study.
- Changes in parent-rated attention-deficit/hyperactivity disorder symptom scores and sleep parameters during individualized multidisciplinary clinical management in children with comorbid obstructive sleep apnea: a prospective cohort study in Vietnam
BibTeX
@article{li2026clinically,
title = {{A Clinically Informed Cross-Attention Deep Learning Framework for Automated Diagnosis of Obstructive Sleep Apnea}},
author = {Lingyi Li and Bing Si},
journal = {IEEE International Conference on Automation Science and Engineering (CASE) : [proceedings]. IEEE Conference on Automation Science and Engineering},
year = {2026},
month = oct,
doi = {10.1109/case69030.2026.11704404},
url = {https://doi.org/10.1109/case69030.2026.11704404},
note = {CC BY},
}
RIS
TY - JOUR
TI - A Clinically Informed Cross-Attention Deep Learning Framework for Automated Diagnosis of Obstructive Sleep Apnea
AU - Lingyi Li
AU - Bing Si
PY - 2026
DA - 2026/10/06
JO - IEEE International Conference on Automation Science and Engineering (CASE) : [proceedings]. IEEE Conference on Automation Science and Engineering
DO - 10.1109/case69030.2026.11704404
UR - https://doi.org/10.1109/case69030.2026.11704404
AB - Obstructive sleep apnea (OSA) is a highly prevalent breathing-related sleep disorder associated with a variety of cardiovascular comorbidities. Polysomnography (PSG) is the clinical gold standard for OSA diagnosis, with disease severity quantified by the apnea hypopnea index (AHI). However, deriving AHI requires manual annotation of overnight PSG recordings by certificated technicians, making the OSA diagnostic process costly and time-consuming. While it is highly desirable to automate this process with a reliable machine learning pipeline, conventional models often rely on handcrafted feature engineering and may therefore not sufficiently capture the complex dynamics of physiological signals. Deep learning models can learn nonlinear representations directly from raw PSG signals, but many existing works study PSG independently and give less consideration to clinical variables that are known to influence OSA risk, e.g., age, sex. More importantly, most prior research has focused on a single dataset, raising concerns about the robustness and generalizability of these models. To tackle these limitations, this project proposes a clinical knowledge-guided cross-attention deep learning framework for automated PSG annotation and AHI prediction from overnight PSG recordings, combined with clinical variables. The proposed framework employs a deep encoder to learn epoch-level features from PSG data and classify epochs based on whether an adverse respiratory event occurs, providing epoch-level PSG annotations that mimic manual evaluation in current practice. Then, the epoch-level embeddings will be integrated with clinical variables using a multi-head attention mechanism for patient-level AHI prediction. The framework is validated across four large real-world datasets.
SN - 2161-8070
SN - 2161-8089
ER -