KindSleep: Knowledge-Informed Diagnosis of Obstructive Sleep Apnea from Oximetry

• URL: http://arxiv.org/abs/2603.04755v1
• Date: Thu, 05 Mar 2026 03:00:34 GMT
• Title: KindSleep: Knowledge-Informed Diagnosis of Obstructive Sleep Apnea from Oximetry
• Authors: Micky C Nnamdi, Wenqi Shi, Cheng Wan, J. Ben Tamo, Benjamin M Smith, Chad A Purnell, May D Wang,
• Abstract summary: We introduce KindSleep, a deep learning framework that integrates clinical knowledge with single-channel patient-specific oximetry signals and clinical data for precise OSA diagnosis.<n>KindSleep first learns to identify clinically interpretable concepts, such as desaturation indices and respiratory disturbance events, directly from raw oximetry signals.<n>It then fuses these AI-derived concepts with multimodal clinical data to estimate the Apnea-Hypopnea Index (AHI)
• Score: 5.901247752047518
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Obstructive sleep apnea (OSA) is a sleep disorder that affects nearly one billion people globally and significantly elevates cardiovascular risk. Traditional diagnosis through polysomnography is resource-intensive and limits widespread access, creating a critical need for accurate and efficient alternatives. In this paper, we introduce KindSleep, a deep learning framework that integrates clinical knowledge with single-channel patient-specific oximetry signals and clinical data for precise OSA diagnosis. KindSleep first learns to identify clinically interpretable concepts, such as desaturation indices and respiratory disturbance events, directly from raw oximetry signals. It then fuses these AI-derived concepts with multimodal clinical data to estimate the Apnea-Hypopnea Index (AHI). We evaluate KindSleep on three large, independent datasets from the National Sleep Research Resource (SHHS, CFS, MrOS; total n = 9,815). KindSleep demonstrates excellent performance in estimating AHI scores (R2 = 0.917, ICC = 0.957) and consistently outperforms existing approaches in classifying OSA severity, achieving weighted F1-scores from 0.827 to 0.941 across diverse populations. By grounding its predictions in a layer of clinically meaningful concepts, KindSleep provides a more transparent and trustworthy diagnostic tool for sleep medicine practices.

Related papers

• Stanford Sleep Bench: Evaluating Polysomnography Pre-training Methods for Sleep Foundation Models [18.499236999143474]
  We release Stanford Sleep Bench, a large-scale PSG dataset comprising 17,467 recordings totaling over 163,000 hours from a major sleep clinic.<n>Our results show that multiple pretraining methods achieve comparable performance for sleep staging, apnea diagnosis, and age estimation.<n>To facilitate and advance sleep research, we will release Stanford Sleep Bench along with pretrained model weights, training pipelines, and evaluation code.
  arXiv  Detail & Related papers  (2025-12-10T12:37:29Z)
• Exploring the Efficacy of Convolutional Neural Networks in Sleep Apnea Detection from Single Channel EEG [0.0]
  This paper presents a novel approach to the detection of sleep apnea using a Convolutional Neural Network (CNN) trained on single channel EEG data.<n>The proposed CNN achieved an accuracy of 85.1% and a Matthews Correlation Coefficient (MCC) of 0.22, demonstrating a significant potential for home based applications.
  arXiv  Detail & Related papers  (2025-08-16T04:58:00Z)
• MobileNetV2: A lightweight classification model for home-based sleep apnea screening [3.463585190363689]
  This study proposes a novel lightweight neural network model leveraging features extracted from electrocardiogram (ECG) and respiratory signals for early OSA screening.<n> ECG signals are used to generate feature spectrograms to predict sleep stages, while respiratory signals are employed to detect sleep-related breathing abnormalities.<n>By integrating these predictions, the method calculates the apnea-hypopnea index (AHI) with enhanced accuracy, facilitating precise OSA diagnosis.
  arXiv  Detail & Related papers  (2024-12-28T01:37:25Z)
• Thermal Imaging and Radar for Remote Sleep Monitoring of Breathing and Apnea [42.00356210257671]
  We show the first comparison of radar and thermal imaging for sleep monitoring. Our thermal imaging method detects apneas with an accuracy of 0.99, a precision of 0.68, a recall of 0.74, an F1 score of 0.71, and an intra-class correlation of 0.73. We present a novel proposal for classifying obstructive and central sleep apnea by leveraging a multimodal setup.
  arXiv  Detail & Related papers  (2024-07-16T17:26:50Z)
• What Radio Waves Tell Us about Sleep [34.690382091650314]
  We develop an advanced machine learning algorithm for passively monitoring sleep and nocturnal breathing from radio waves reflected off people while asleep. We show that the model captures the sleep hypnogram (with an accuracy of 81% for 30-second epochs categorized into Wake, Light Sleep, Deep Sleep, or REM) and detects sleep apnea (AUROC = 0.88) The model uncovers informative interactions between sleep stages and a range of diseases including neurological, psychiatric, cardiovascular, and immunological disorders.
  arXiv  Detail & Related papers  (2024-05-20T02:41:21Z)
• Sleep Activity Recognition and Characterization from Multi-Source Passively Sensed Data [67.60224656603823]
  Sleep Activity Recognition methods can provide indicators to assess, monitor, and characterize subjects' sleep-wake cycles and detect behavioral changes. We propose a general method that continuously operates on passively sensed data from smartphones to characterize sleep and identify significant sleep episodes. Thanks to their ubiquity, these devices constitute an excellent alternative data source to profile subjects' biorhythms in a continuous, objective, and non-invasive manner.
  arXiv  Detail & Related papers  (2023-01-17T15:18:45Z)
• Convolutional Neural Networks for Sleep Stage Scoring on a Two-Channel EEG Signal [63.18666008322476]
  Sleep problems are one of the major diseases all over the world. Basic tool used by specialists is the Polysomnogram, which is a collection of different signals recorded during sleep. Specialists have to score the different signals according to one of the standard guidelines.
  arXiv  Detail & Related papers  (2021-03-30T09:59:56Z)
• MSED: a multi-modal sleep event detection model for clinical sleep analysis [62.997667081978825]
  We designed a single deep neural network architecture to jointly detect sleep events in a polysomnogram. The performance of the model was quantified by F1, precision, and recall scores, and by correlating index values to clinical values.
  arXiv  Detail & Related papers  (2021-01-07T13:08:44Z)
• UNITE: Uncertainty-based Health Risk Prediction Leveraging Multi-sourced Data [81.00385374948125]
  We present UNcertaInTy-based hEalth risk prediction (UNITE) model. UNITE provides accurate disease risk prediction and uncertainty estimation leveraging multi-sourced health data. We evaluate UNITE on real-world disease risk prediction tasks: nonalcoholic fatty liver disease (NASH) and Alzheimer's disease (AD) UNITE achieves up to 0.841 in F1 score for AD detection, up to 0.609 in PR-AUC for NASH detection, and outperforms various state-of-the-art baselines by up to $19%$ over the best baseline.
  arXiv  Detail & Related papers  (2020-10-22T02:28:11Z)
• Automatic detection of microsleep episodes with deep learning [55.41644538483948]
  Brief fragments of sleep shorter than 15 s are defined as microsleep episodes (MSEs) maintenance of wakefulness test (MWT) is often used in a clinical setting to assess vigilance. MSEs are mostly not considered in the absence of established scoring criteria defining MSEs. We aimed for automatic detection of MSEs with machine learning based on raw EEG and EOG data as input.
  arXiv  Detail & Related papers  (2020-09-07T11:38:40Z)
• Automatic scoring of apnea and hypopnea events using blood oxygen saturation signals [0.0]
  DAS-KSVD was applied to detect and classify apnea and hypopnea events from signals obtained from the Sleep Heart Health Study database. For moderate to severe OSAH screening, a receiver operating characteristic curve analysis of the results shows an area under the curve of 0.957 and diagnostic sensitivity and specificity of 87.56% and 88.32%, respectively.
  arXiv  Detail & Related papers  (2020-03-22T15:17:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.