SERF: Interpretable Sleep Staging using Embeddings, Rules, and Features

TL;DR

SERF introduces an interpretable sleep staging method that combines deep learning embeddings with clinical features, achieving high accuracy and interpretability for sleep analysis from polysomnogram data.

Contribution

The paper presents SERF, a novel approach that integrates deep neural network embeddings with rule-based features for interpretable sleep staging, outperforming existing methods.

Findings

SERF surpasses state-of-the-art interpretable sleep staging by 2%.

SERF achieves 0.766 κ and 0.870 AUC-ROC, close to black-box models.

The method provides meaningful clinical feature interpretations.

Abstract

The accuracy of recent deep learning based clinical decision support systems is promising. However, lack of model interpretability remains an obstacle to widespread adoption of artificial intelligence in healthcare. Using sleep as a case study, we propose a generalizable method to combine clinical interpretability with high accuracy derived from black-box deep learning. Clinician-determined sleep stages from polysomnogram (PSG) remain the gold standard for evaluating sleep quality. However, PSG manual annotation by experts is expensive and time-prohibitive. We propose SERF, interpretable Sleep staging using Embeddings, Rules, and Features to read PSG. SERF provides interpretation of classified sleep stages through meaningful features derived from the AASM Manual for the Scoring of Sleep and Associated Events. In SERF, the embeddings obtained from a hybrid of convolutional and recurrent neural networks are transposed to the interpretable feature space. These representative interpretable features are used to train simple models like a shallow decision tree for classification. Model results are validated on two publicly available datasets. SERF surpasses the current state-of-the-art for interpretable sleep staging by 2%. Using Gradient Boosted Trees as the classifier, SERF obtains 0.766 $\kappa$ and 0.870 AUC-ROC, within 2% of the current state-of-the-art black-box models.