Virtual reality (VR) systems are increasingly using physiology to improve human training. However, these systems do not account for the complex intra-individual variability in physiology and human performance across multiple timescales and psychophysiological demands. To fill this gap, we propose a theory of multilevel variability where tractable neurobiological mechanisms generate complex variability in performance over time and in response to heterogeneous sources. Based on this theory, we also present a study that examines changes in cardiovascular activity and performance during a stressful shooting task in VR. We examined physiology and performance at three important levels of analysis: task-to-task, block-to-block, session-to-session. Findings indicated joint patterns of physiology and performance that notably varied by the level of analysis. At the task level, higher task difficulty worsened performance but did not change cardiovascular activation. At the block level, there were nonlinear changes in performance and heart rate variability. At the session level, performance improved while blood pressure decreased and heart rate variability increased across days. Of all the physiological metrics, only heart rate variability was correlated with marksmanship performance. Findings are consistent with our multilevel theory and highlight the need for VR and other affective computing systems to assess physiology across multiple timescales.
Objective. Most arrhythmias due to cardiovascular diseases alter the electrical activity, resulting in morphological alterations in electrocardiogram (ECG) recordings. ECG acquisition is a low-cost, non-invasive process and is commonly used for continuous monitoring as a diagnostic tool for cardiac abnormality identification. Our objective is to diagnose twenty-nine cardiac abnormalities and sinus rhythm using varied lead ECG signals. Approach. This work proposes a deep residual inception network with channel attention mechanism (RINCA) for twenty-nine cardiac arrhythmia classification (CAC) along with normal ECG from multi-label ECG signal with different lead combinations. The RINCA architecture employing the Inception-based convolutional neural network backbone uses residual skip connections with the channel attention mechanism. The Inception model facilitates efficient computation and prevents overfitting while exploring deeper networks through dimensionality reduction and stacked 1-dimensional convolutions. The residual skip connections alleviate the vanishing gradient problem. The attention modules selectively leverage the temporally significant segments in a sequence and predominant channels for multi-lead ECG signals, contributing to the decision-making. Main results. Exhaustive experimental evaluation on the large-scale 'PhysioNet/Computing in Cardiology Challenge (2021)' dataset demonstrates RINCA efficacy. On the hidden test data set, RINCA achieves the challenge metric score of 0.55, 0.51, 0.53, 0.51, and 0.53 (ranked 2nd, 5th, 4th, 5th and 4th) for the twelve-lead, six-lead, four-lead, three-lead, and two-lead combination cases, respectively. Significance. The proposed RINCA model is more robust against varied sampling frequency, recording time, and data with heterogeneous demographics than the existing art. The explainability analysis shows RINCA potential in clinical interpretations.
Automated cardiac abnormality detection from an everexpanding number of electrocardiogram (ECG) records has been widely used to assist physicians in the clinical diagnosis of a variety of cardiovascular diseases. Over the last few years, deep learning (DL) architectures have achieved state-of-the-art performances in various biomedical applications. In this work, we propose a bio-toolkit based on the DL framework comprising of stacked convolutional and long short term memory neural network blocks for multi-label ECG signal classification. Our team participated under the name "Cardio-Challengers" in the "PhysioNet/Computing in Cardiology Challenge 2020" and obtained a challenge metric score of 0.337.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.