Pediatric sleep staging from airflow signals via persistence curve approximations

Sleep staging is a challenging task, typically performed manually by sleep technologists based on electroencephalogram (EEG) and other biosignals of patients taken during overnight sleep studies. Recent work aims to leverage automated algorithms to perform sleep staging not based on EEG signals, but rather based on the airflow signals of subjects. We aim to show improved sleep staging performance based on airflow signals alone, and furthermore that this method is viable on pediatric subjects.

Prior work uses ideas from topological data analysis (TDA), specifically Hermite function expansions of persistence curves (HEPC) to featurize airflow signals. However, finite order HEPC captures only partial information. In this work, we propose Fourier approximations of persistence curves (FAPC), and use this technique to perform sleep staging based on airflow signals.

We analyze performance using an XGBoost model on 1155 pediatric sleep studies taken from the Nationwide Children's Hospital Sleep DataBank (NCHSDB), and find that FAPC methods provide complimentary information to HEPC methods alone, leading to a 4.9% increase in performance over baseline methods.

The proposed TDA featurization technique allows an improvement over existing methods to classify sleep stage from respiratory signals. The experimental evaluation of this technique shows the promise of this method on pediatric subjects.

The results included in this work suggest the potential of TDA based featurization methods to improve other time-series signal processing problems.