Stress Level Detection and Evaluation from Phonation and PPG Signals Recorded in an Open-Air MRI Device

Abstract

This paper deals with two modalities for stress detection and evaluation—vowel phonation speech signal and photo-plethysmography (PPG) signal. The main measurement is carried out in four phases representing different stress conditions for the tested person. The first and last phases are realized in laboratory conditions. The PPG and phonation signals are recorded inside the magnetic resonance imaging scanner working with a weak magnetic field up to 0.2 T in a silent state and/or with a running scan sequence during the middle two phases. From the recorded phonation signal, different speech features are determined for statistical analysis and evaluation by the Gaussian mixture models (GMM) classifier. A database of affective sounds and two databases of emotional speech were used for GMM creation and training. The second part of the developed method gives comparison of results obtained from the statistical description of the sensed PPG wave together with the determined heart rate and Oliva–Roztocil index values. The fusion of results obtained from both modalities gives the final stress level. The performed experiments confirm our working assumption that a fusion of both types of analysis is usable for this task—the final stress level values give better results than the speech or PPG signals alone.

Analysis of Heart Pulse Transmission Parameters Determined from Multi-Channel PPG Signals Acquired by a Wearable Optical Sensor

“…In agreement with previous developments and realizations of wearable PPG sensors [15], [16], [18], our current prototype consists of similar functional parts:…”

“…The real-time algorithm for PTT determination from a multi-channel PPG signal must be fast, stable, sufficiently precise, and relatively easy to implement on the chosen HW platform. The technique used is similar to that used for one-channel PPG signal acquisition [15], [16]. The PPG signals are processed in the following way:…”

“…In the long term, we are interested in the analysis of stress in persons scanned in the MRI tomograph operating with a low stationary magnetic field B0 up to 0.2 T [15], [16]. For stress detection and quantization, we have previously used the PPG signal sampled from a wearable optical sensor on the fingers with parallel measurement of blood pressure values from an external wearable blood pressure monitor (BPM) device.…”

“…Scanning measuring phase: The high-resolution MR scan sequence is performed (further called "MRISC"). The head of the measured subject is situated near the receiving/transmitting RF coil [29] as in our previous experiments [15], [16] to maximize the negative vibration and noise effect on the tested person. The typical time duration of the currently used scan sequence is approximately 5 minutes (including the calibration pre-scan part).…”

Survey of Transfer Learning Approaches in the Machine Learning of Digital Health Sensing Data

“…Predicting glucose levels in patients with diabetes [176,177] and monitoring the emotional state or stress levels [178,179] Electro-oculogram (EOG) (Time series)…”