Classification of REM-sleep-behavior-disorder based on convolutional neural network using cardiopulmonary coupling spectrogram

Reference:
Park J, Urtnasan E, Joo E, Lee K. Classification of REM sleep behavior disorder base on convolutional neural network using cardiopulmonary coupling spectrogram. Sleep Medicine 2017. DOI: 10.1016/j.sleep.2017.11.740

Objectives:
To propose a method of automatically classify patients of REM sleep behavior disorder (RBD) based on convolutional neural network (CNN) using cardiopulmonary coupling (CPC) spectrum induced from a single-channel electrocardiogram (ECG).

Conclusions:
The classification performance showed sensitivity, specificity and accuracy of 80.0%, 85.7% and 82.3% for RBD, respectively.

Practical Significance:

The ECG-derived CPC-based method has the potential for automatically classifying RBD patients in a home environment without any other physiological signals.

View Publication

Classification of REM-sleep-behavior-disorder based on convolutional neural network using cardiopulmonary coupling spectrogram