Pediatric SleepNet: a deep learning network for reliable pediatric sleep staging across developmental stages
Ayush Tripathi1,2, Arnav Gupta1,3, Wolfgang Ganglberger1,2
1Department of Neurology, Beth Israel Deaconess Medical Center, Boston, MA, United States.
Insights
A new deep learning model, pediatric SleepNet, accurately stages sleep in children across various ages and conditions. This advanced AI tool shows promise for improving pediatric sleep medicine research and clinical practice.
Area of Science:
- Pediatric Sleep Medicine
- Artificial Intelligence in Healthcare
- Computational Neuroscience
Background:
- Manual sleep staging in children is difficult due to developmental variations and inconsistent scoring, particularly in infants.
- Accurate sleep staging is crucial for diagnosing and managing pediatric sleep disorders.
Purpose of the Study:
- To develop and evaluate a multimodal deep learning model, pediatric SleepNet, for automated sleep staging in pediatric populations.
- To assess the model's performance across a wide age range and diverse clinical subgroups.
Main Methods:
- A U-Net-inspired encoder-decoder model (pediatric SleepNet) was trained using 9-channel physiological signals (EEG, EOG, EMG) from 9,150 pediatric polysomnograms (PSGs).
- Models were trained on three age groups (<6 months, 6-12 months, >1 year) and evaluated on 3,804 test recordings, with comparisons to U-Sleep and CAISR.
- Stratified analyses were conducted across ages, sexes, and disease categories, with external validation on two independent datasets.
Main Results:
- pediatric SleepNet demonstrated robust performance, with mean Cohen's Kappa increasing from 0.49 (0-6 months) to 0.72 (>12 years).
- The model significantly outperformed U-Sleep and CAISR in early developmental stages and showed comparable performance on external validation datasets (Kappa >0.69).
- Performance reductions were noted in children with epilepsy, Down syndrome, hydrocephalus, and other neurodevelopmental conditions.
Conclusions:
- pediatric SleepNet provides reliable sleep staging across pediatric development, ages, and diverse clinical conditions.
- The model's strong performance across internal and external datasets supports its utility in pediatric sleep medicine research and clinical applications.
Study Objectives:
Manual sleep staging in pediatric populations is challenging due to developmental variability and limited scoring consistency, especially in infants and toddlers. We developed a multimodal deep learning model for pediatric sleep staging and evaluated its performance across a broad age range and diverse clinical subgroups.
Methods:
We trained a U-Net-inspired encoder-decoder model (pediatric SleepNet) using 9-channel input signals: electroencephalography (EEG), electrooculography (EOG), and chin electromyography (EMG) using 35-epoch segments from clinical pediatric polysomnograms (PSGs). Models were trained separately on three age groups (<6 months, 6-12 months, >1 year) using 9150 PSGs, with 2455 PSGs reserved for validation. Evaluation was conducted on 3804 held-out test recordings. Performance was compared with U-Sleep and the Complete Artificial Intelligence Sleep Report (CAISR), and stratified analyses were performed across ages, sexes, and seven ICD-10-based disease categories. External validation was conducted on two independent datasets, CHAT and PATS.
Results:
pediatric SleepNet achieved robust performance across all age groups, with mean Cohen's Kappa increasing from 0.49 (0-6 months) to 0.72 (>12 years). It significantly outperformed U-Sleep and CAISR across early developmental stages. Three-class staging yielded mean Cohen's Kappa increasing from 0.66 (0-6 months) to 0.79 (>12 years). Sex-based differences were negligible. However, significant reductions in performance were observed in children with epilepsy, Down syndrome, hydrocephalus, and other neurodevelopmental conditions. External validation yielded Kappa values >0.69 comparable to the internal test set.
Conclusions:
pediatric SleepNet demonstrates reliable sleep staging across pediatric development. Its robust performance across age, disease, and external datasets supports its potential for clinical and research use in pediatric sleep medicine.
