RT info:eu-repo/semantics/article
T1 An explainable deep-learning architecture for pediatric sleep apnea identification from overnight airflow and oximetry signals
A1 Jiménez García, Jorge
A1 García Gadañón, María
A1 Gutiérrez Tobal, Gonzalo César
A1 Kheirandish Gozal, Leila
A1 Vaquerizo Villar, Fernando
A1 Álvarez González, Daniel
A1 Campo Matias, Félix del
A1 Gozal, David
A1 Hornero Sánchez, Roberto
K1 Pediatría
K1 Biomedical engineering
K1 Obstructive sleep apnea
K1 Children
K1 Airflow
K1 Apnea obstructiva del sueño
K1 Niños
K1 Flujo de aire
K1 32 Ciencias Médicas
AB Deep-learning algorithms have been proposed to analyze overnight airflow (AF) and oximetry (SpO2) signals to simplify the diagnosis of pediatric obstructive sleep apnea (OSA), but current algorithms are hardly interpretable. Explainable artificial intelligence (XAI) algorithms can clarify the models-derived predictions on these signals, enhancing their diagnostic trustworthiness. Here, we assess an explainable architecture that combines convolutional and recurrent neural networks (CNN + RNN) to detect pediatric OSA and its severity. AF and SpO2 were obtained from the Childhood Adenotonsillectomy Trial (CHAT) public database (n = 1,638) and a proprietary database (n = 974). These signals were arranged in 30-min segments and processed by the CNN + RNN architecture to derive the number of apneic events per segment. The apnea-hypopnea index (AHI) was computed from the CNN + RNN-derived estimates and grouped into four OSA severity levels. The Gradient-weighted Class Activation Mapping (Grad-CAM) XAI algorithm was used to identify and interpret novel OSA-related patterns of interest. The AHI regression reached very high agreement (intraclass correlation coefficient > 0.9), while OSA severity classification achieved 4-class accuracies 74.51% and 62.31%, and 4-class Cohen’s Kappa 0.6231 and 0.4495, in CHAT and the private datasets, respectively. All diagnostic accuracies on increasing AHI cutoffs (1, 5 and 10 events/h) surpassed 84%. The Grad-CAM heatmaps revealed that the model focuses on sudden AF cessations and SpO2 drops to detect apneas and hypopneas with desaturations, and often discards patterns of hypopneas linked to arousals. Therefore, an interpretable CNN + RNN model to analyze AF and SpO2 can be helpful as a diagnostic alternative in symptomatic children at risk of OSA.
PB Elsevier
SN 1746-8094
YR 2024
FD 2024
LK https://uvadoc.uva.es/handle/10324/62524
UL https://uvadoc.uva.es/handle/10324/62524
LA eng
NO Biomedical Signal Processing and Control, 2024, vol. 87, Part B, 105490
NO Producción Científica
DS UVaDOC
RD 22-dic-2024