SleepECG-Net: Explainable Deep Learning Approach With ECG for Pediatric Sleep Apnea Diagnosis
Year: 2024
Type: article
Abstract: Obstructive sleep apnea (OSA) in children is a prevalent and serious respiratory condition linked to cardiovascular morbidity. Polysomnography, the standard diagnostic approach, faces challenges in accessibility and complexity, leading to underdiagnosis. To simplify OSA diagnosis, deep learning (DL) algorithms have been developed using cardiac signals, but they often lack interpretability. Our study introduces a novel interpretable DL approach (SleepECG-Net) for directly estimating OSA severity in at-risk children. A combination of convolutional and recurrent neural networks (CNN-RNN) was trained on overnight electrocardiogram (ECG) signals. Gradient-weighted Class Activation Mapping (Grad-CAM), an eXplainable Artificial Intelligence (XAI) algorithm, was applied to explain model decisions and extract ECG patterns relevant to pediatric OSA. Accordingly, ECG signals from the semi-public Childhood Adenotonsillectomy Trial (CHAT, n = 1610) and Cleveland Family Study (CFS,n = 64), and the private University of Chicago (UofC, n = 981) databases were used. OSA diagnostic performance reached 4-class Cohen's Kappa of 0.410, 0.335, and 0.249 in CHAT, UofC, and CFS, respectively. The proposal demonstrated improved performance with increased severity along with heightened cardiovascular risk. XAI findings highlighted the detection of established ECG features linked to OSA, such as bradycardia-tachycardia events and delayed ECG patterns during apnea/hypopnea occurrences, focusing on clusters of events. Furthermore, Grad-CAM heatmaps identified potential ECG patterns indicating cardiovascular risk, such as P, T, and U waves, QT intervals, and QRS complex variations. Hence, SleepECG-Net approach may improve pediatric OSA diagnosis by also offering cardiac risk factor information, thereby increasing clinician confidence in automated systems, and promoting their effective adoption in clinical practice.
Language: English
Subfield: Physiology
Field: Medicine
Domain: Health Sciences
Open Access status: green
Funders: Ministerio de Ciencia, Innovación y Universidades, European Commission, Universidad de Valladolid, National Institutes of Health, Instituto de Salud Carlos III, European Regional Development Fund, Agencia Estatal de Investigación, National Heart, Lung, and Blood Institute, National Center for Research Resources, Interreg, European Social Fund, NextGenerationEU
Awards: G1050765037, G1105844351, G1525460525, G1622782223, G17239878, G1837804946, G1986609078, APP-mediated signaling, sleep perturbations, and Alzheimer's disease mouse models, G2243702923, G2260006100, Case Western Reserve University/Cleveland Clinic CTSA (UL1), G2371472412, G2519937635, G2855301824, G2862297731, A randomized controlled study of adenotonsillectomy for children, G3012971696, G318412399, G3214468647, Familial Aggregation and Natural History of Sleep Apnea, G3480869486, G3618234253, G3779662906, G3925316322, G3990393904, G410571670, G4249524944, G451917667, G491712853, G4949749077, G499116126, G5566929262, G5635532810, G5745121124, Cognitive Function in Snoring Children, G5821954751, G5967599077, G6218905885, G624877701, G6520718613, G670555432, G6731524417, G675437911, G678360218, G7084143925, G7511203839, G7538875822, G7703859750, Randomized Controlled Study of Adenotonsillectomy for Childhood Sleep Apnea, G7799207878, G7805626310, G7863371703, G8186789035, G8298419571, G8497050386, G8932742119, G9000009611, G9000037440, G9000068995, G9000076961, G9000132197, G9000199928, G9000281696, Coronary artery dysfunction in OSA: Role of mineralocorticoid receptors, G91634571, G929308097
OpenAlex ID: W4404238544