Artifacts classification and apnea events detection in neck photoplethysmography signals

Abstract

The novel pulse oximetry measurement site of the neck is a promising location for multi-modal physiological monitoring. Specifically, in the context of respiratory monitoring, in which it is important to have direct information about airflow. The neck makes this possible, in contrast to common photoplethysmography (PPG) sensing sites. However, this PPG signal is susceptible to artifacts that critically impair the signal quality. To fully exploit neck PPG for reliable physiological parameters extraction and apneas monitoring, this paper aims to develop two classification algorithms for artifacts and apnea detection. Features from the time, correlogram, and frequency domains were extracted. Two SVM classifiers with RBF kernels were trained for different window (W) lengths and thresholds (Thd) of corruption. For artifacts classification, the maximum performance was attained for the parameters combination of [W = 6s-Thd= 20%], with an average accuracy= 85.84%(ACC), sensitivity= 85.43%(SE) and specificity= 86.26%(SP). For apnea detection, the model [W = 10s-Thd= 50%] maximized all the performance metrics significantly (ACC= 88.25%, SE= 89.03%, SP= 87.42%). The findings of this proof of concept are significant for denoising novel neck PPG signals, and demonstrate, for the first time, that it is possible to promptly detect apnea events from neck PPG signals in an instantaneous manner. This could make a big impact in crucial real-time applications, like devices to prevent sudden-unexpected-death-in-epilepsy (SUDEP).

A comparative study in class imbalance mitigation when working with physiological signals

“…A total of 49 features were extracted from each preprocessed PPG segment, all of which were proposed and evaluated for processing PPG signals in previous studies ( 15 , 17 ), Tables 2 , 3 list those features and their brief description.…”

“…Table 3 shows the frequency-domain, correlogram and envelop features that were calculated over the window segments. The envelope features were obtained from the upper envelope of the PPG signal as described in ( 15 ). Features were standardized before further processing, and standardization parameters were strictly obtained from the training partition of the dataset.…”

“… Section-wise splits where each subject’s data was partitioned into three sections, then randomly choosing one of these sections and leaving it out for testing. This method was illustrated in ( 15 ). The partitioning was performed before windowing the data for feature extraction, and the sliding window was only allowed to traverse locally within the sections.…”

“…The raw PPG waveform is rarely used to detect apnoea, however, it was proven to be feasible especially when obtained from the neck since it becomes strongly modulated by respiration ( 14 ). Indeed it was shown in a later study that it was possible to detect apnoea events using PPG signals without the need for obtaining SpO2 measurements that pose a significant time delay ( 15 ).…”

Quantifying the Suitability of Biosignals Acquired During Surgery for Multimodal Analysis

“…Analysis of signal envelopes allows for detecting noisy peaks in a signal and has already been used for assessing the quality of pulse oximetry signals [45]. At first, signal amplitudes were scaled to [−1, 1] to normalize across the different types of signals.…”