JMIR Publications

ABSTRACT

Background:

Sleep apnea is a respiratory disorder characterized by interruption to breathing during sleep. Depending on the presence of breathing effort, sleep apnea is divided into Obstructive Sleep Apnea (OSA) and Central Sleep Apnea (CSA) caused by different pathologies. Choosing the appropriate treatment relies on distinguishing between CSA and OSA.

Objective:

To develop a non-contact method to distinguish between obstructive versus central sleep apnea

Methods:

In this paper, five different computer vision-based algorithms were applied to infrared videos data to track and to analyze body movements to differentiate different types of sleep apnea (CSA vs. OSA). In the first two methods, supervised classifiers were trained to process optical flow information. In the remaining three methods, a convolutional neural network was designed to extract feature from optical flow and to distinguish OSA from CSA.

Results:

Overnight sleeping data of 42 participant (Age: 53±15 years, BMI: 30±7 kg/m2, 27 men and 15 women, #OSA: 16±30, #CSA: 3±7, AHI = 27±31 events/hour, sleep duration = 5±1 hour) were collected for this paper. The test and train data were recorded in two separate laboratory rooms. The best performing model (3D-CNN) obtained 95% accuracy and 89% F1-score in differentiating OSA versus CSA.

Conclusions:

In this paper, the first vision-based method was developed that differentiates apnea types (OSA vs. CSA). Developed algorithm tracks and analyses chest and abdominal movements captured via an infrared video camera. Unlike previously developed approaches, this method does not require any attachment to a user that could potentially alter the sleeping condition.