Multiple Physiological Signals Fusion Techniques for Improving Heartbeat Detection: A Review

Abstract

This paper presents a review of the techniques found in the literature that aim to achieve a robust heartbeat detection from fusing multi-modal physiological signals (e.g., electrocardiogram (ECG), blood pressure (BP), artificial blood pressure (ABP), stroke volume (SV), photoplethysmogram (PPG), electroencephalogram (EEG), electromyogram (EMG), and electrooculogram (EOG), among others). Techniques typically employ ECG, BP, and ABP, of which usage has been shown to obtain the best performance under challenging conditions. SV, PPG, EMG, EEG, and EOG signals can help increase performance when included within the fusion. Filtering, signal normalization, and resampling are common preprocessing steps. Delay correction between the heartbeats obtained over some of the physiological signals must also be considered, and signal-quality assessment to retain the best signal/s must be considered as well. Fusion is usually accomplished by exploiting regularities in the RR intervals; by selecting the most promising signal for the detection at every moment; by a voting process; or by performing simultaneous detection and fusion using Bayesian techniques, hidden Markov models, or neural networks. Based on the results of the review, guidelines to facilitate future comparison of the performance of the different proposals are given and promising future lines of research are pointed out.

Digital Volume Pulse Measured at the Fingertip as an Indicator of Diabetic Peripheral Neuropathy in the Aged and Diabetic

“…In recent years, the percussion entropy index (PEI) using synchronized ECG and photoplethysmography (PPG) signals has been used effectively to assess BRS complexity in the aged and diabetic patients associated with type 2 diabetes-associated autonomic dysfunction [16,17]. The peak-to-peak intervals (PPIs) acquired by the PPG sensor were reported to assess HRV [18][19][20][21][22][23]. Moreover, PPG-derived digital volume pulse (DVP) signals were further used for clinical applications (i.e., ubiquitous blood pressure monitoring, congestive heart failure, and hypertension assessment) [24][25][26].…”

“…Although the above algorithms have been reported to increase the accuracy of PPI, most efforts have been made to reduce computation load [21][22][23]. The objectives of this study are to test two hypotheses.…”

ECG Monitoring Systems: Review, Architecture, Processes, and Key Challenges

“…Supporting processes were described in nearly half of the proposed work included in Table 1, such as data storage and data modeling. The most inclusive ECG monitoring system's lifecycles that included the primary processes definition were depicted in [55,58,72,75,78,83]. Few of the researches specified additional supporting processes, such as data storage or encryption.…”

“…However, most researchers working with ECG monitoring systems favor datasets from well-known databases, rather than creating a data acquisition system of their own, especially when they address diagnosis issues and feature extraction techniques, which constitute the remaining parts of the monitoring lifecycle. A thorough review of existing databases that provide single-lead and multi-lead ECG signals was depicted in [58].…”

“…However, exiting reviews related to ECG monitoring in the literature can be intuitive and incomprehensive [48]. They do not consider the latest technological trends [49][50][51], and they target very narrow research niches, such as wearable sensors [52][53][54][55], mobile sensors [56], disease diagnosis [57], heartbeat detection [58], emotion recognition [59], or ECG compression methods [60]. Hence, there is a need to provide a comprehensive, expert-verified taxonomy of ECG monitoring systems, a common architecture, and a complete set of processes to guide the classification, analysis, and design of these systems.…”