Objective:A study was run to test the sensitivity of multiple workload indices to the differing cognitive demands of four military monitoring task scenarios and to investigate relationships between indices.Background:Various psychophysiological indices of mental workload exhibit sensitivity to task factors. However, the psychometric properties of multiple indices, including the extent to which they intercorrelate, have not been adequately investigated.Method:One hundred fifty participants performed in four task scenarios based on a simulation of unmanned ground vehicle operation. Scenarios required threat detection and/or change detection. Both single- and dual-task scenarios were used. Workload metrics for each scenario were derived from the electroencephalogram (EEG), electrocardiogram, transcranial Doppler sonography, functional near infrared, and eye tracking. Subjective workload was also assessed.Results:Several metrics showed sensitivity to the differing demands of the four scenarios. Eye fixation d...

The psychometrics of mental workload: multiple measures are sensitive but divergent.

Advancements in wireless sensor network technologies have enabled the proliferation of miniaturized body-worn sensors, capable of long-term pervasive biomedical signal monitoring. Remote cardiovascular monitoring has been one of the beneficiaries of this development, resulting in non-invasive, photoplethysmography (PPG) sensors being used in ambulatory settings. Wrist-worn PPG, although a popular alternative to electrocardiogram, suffers from motion artifacts inherent in daily life. Hence, in this paper, we present a novel deep learning framework (CorNET) to efficiently estimate heart rate (HR) information and perform biometric identification (BId) using only a wrist-worn, single-channel PPG signal collected in ambulant environment. We have formulated a completely personalized data-driven approach, using a four-layer deep neural network. Two convolution neural network layers are used in conjunction with two long short-term memory layers, followed by a dense output layer for modeling the temporal sequence inherent within the pulsatile signal representative of cardiac activity. The final dense layer is customized with respect to the application, functioning as: regression layer—having a single neuron to predict HR; classification layer—two neurons that identify a subject among a group. The proposed network was evaluated on the TROIKA dataset having 22 PPG records collected during various physical activities. We achieve a mean absolute error of 1.47 ± 3.37 beats per minute for HR estimation and an average accuracy of 96% for BId on 20 subjects. CorNET was further evaluated successfully in an ambulant use-case scenario with custom sensors for two subjects.

CorNET: Deep Learning Framework for PPG-Based Heart Rate Estimation and Biometric Identification in Ambulant Environment

The purpose of this research is to develop a real time system for individual verification with the electrocardiogram (ECG) signal. The ECG signal varies from person to person and it can be used as a new biometric for individual verification. The discrete wavelet transform extract signal features by wavelet coefficients of one-lead ECG signal. Using the proposed approach on 35 normal subjects and 10 arrhythmia patients, 100% verification rate was obtained for normal subjects and 81% verification rate for arrhythmia patients. Furthermore, the performance of the ECG verification system was evaluated by the false acceptance rate (FAR) and false rejection rate (FRR). The FAR was 0.83% and FRR was 0.86%for a database containing only 35 normal subjects. When 10 arrhythmia patients were added into the database, the FAR became 12.50% and the FRR 5.11%. The experimental results demonstrate that the proposed system works very well for normal subjects.

A Novel Personal Identity Verification Approach Using a Discrete Wavelet Transform of the ECG Signal

https://cyberleninka.org/article/n/140859.pdf

DSP-based arrhythmia classification using wavelet transform and probabilistic neural network

Smart clothes increase the efficiency of long-term non-invasive monitoring systems by facilitating the placement of sensors and increasing the number of measurement locations. Since the sensors are either garment-integrated or embedded in an unobtrusive way in the garment, the impact on the subject's comfort is minimized. However, the main challenge of smart clothing lies in the enhancement of signal quality and the management of the huge data volume resulting from the variable contact with the skin, movement artifacts, non-accurate location of sensors and the large number of acquired signals. This paper exposes the strategies and solutions adopted in the European 1ST project MyHeart to address these problems, from the definition of the body sensor network to the description of two embedded signal processing techniques performing on-body ECG enhancement and motion activity classification.

Combination of body sensor networks and on-body signal processing algorithms: the practical case of MyHeart project

The primary goal of this research is to develop an affordable real-time ambulatory electrocardiogram (ECG) monitor prototype. The system consists of an analogue signal conditioner, a micro-controller, external RAM and ROM, and a PCMCIA flash memory card and interfacing chip. Researches are carried out in the ECG data compression and real-time diagnosis. For the purpose of ECG analysis, a reliable QRS detection algorithm with as little computation as possible has to be developed. This paper reports our works in the development of QRS detection algorithms. They are derived and improved from the first derivative method as described in the paper by Friesen et al. Our results show that the maximum slope detection, with the QRS onset selected when two successive values of slope exceed the threshold, is the best method.

K.L. Chan

Papers

Development of QRS detection method for real-time ambulatory cardiac monitor