Sensors

Research

18 pages, 17852 KiB

Open AccessArticle

2D-WinSpatt-Net: A Dual Spatial Self-Attention Vision Transformer Boosts Classification of Tetanus Severity for Patients Wearing ECG Sensors in Low- and Middle-Income Countries

by Ping Lu, Andrew P. Creagh, Huiqi Y. Lu, Ho Bich Hai, VITAL Consortium, Louise Thwaites and David A. Clifton

Sensors 2023, 23(18), 7705; https://doi.org/10.3390/s23187705 - 6 Sep 2023

Cited by 3 | Viewed by 1809

Abstract

Tetanus is a life-threatening bacterial infection that is often prevalent in low- and middle-income countries (LMIC), Vietnam included. Tetanus affects the nervous system, leading to muscle stiffness and spasms. Moreover, severe tetanus is associated with autonomic nervous system (ANS) dysfunction. To ensure early [...] Read more.

Tetanus is a life-threatening bacterial infection that is often prevalent in low- and middle-income countries (LMIC), Vietnam included. Tetanus affects the nervous system, leading to muscle stiffness and spasms. Moreover, severe tetanus is associated with autonomic nervous system (ANS) dysfunction. To ensure early detection and effective management of ANS dysfunction, patients require continuous monitoring of vital signs using bedside monitors. Wearable electrocardiogram (ECG) sensors offer a more cost-effective and user-friendly alternative to bedside monitors. Machine learning-based ECG analysis can be a valuable resource for classifying tetanus severity; however, using existing ECG signal analysis is excessively time-consuming. Due to the fixed-sized kernel filters used in traditional convolutional neural networks (CNNs), they are limited in their ability to capture global context information. In this work, we propose a 2D-WinSpatt-Net, which is a novel Vision Transformer that contains both local spatial window self-attention and global spatial self-attention mechanisms. The 2D-WinSpatt-Net boosts the classification of tetanus severity in intensive-care settings for LMIC using wearable ECG sensors. The time series imaging—continuous wavelet transforms—is transformed from a one-dimensional ECG signal and input to the proposed 2D-WinSpatt-Net. In the classification of tetanus severity levels, 2D-WinSpatt-Net surpasses state-of-the-art methods in terms of performance and accuracy. It achieves remarkable results with an F1 score of 0.88 ± 0.00, precision of 0.92 ± 0.02, recall of 0.85 ± 0.01, specificity of 0.96 ± 0.01, accuracy of 0.93 ± 0.02 and AUC of 0.90 ± 0.00. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

13 pages, 2210 KiB

Open AccessArticle

Enhancing Safety in Hyperbaric Environments through Analysis of Autonomic Nervous System Responses: A Comparison of Dry and Humid Conditions

by Carlos Sánchez, Alberto Hernando, Juan Bolea, David Izquierdo, Germán Rodríguez, Agustín Olea, María Teresa Lozano and María Dolores Peláez-Coca

Sensors 2023, 23(11), 5289; https://doi.org/10.3390/s23115289 - 2 Jun 2023

Cited by 1 | Viewed by 1272

Abstract

Diving can have significant cardiovascular effects on the human body and increase the risk of developing cardiac health issues. This study aimed to investigate the autonomic nervous system (ANS) responses of healthy individuals during simulated dives in hyperbaric chambers and explore the effects [...] Read more.

Diving can have significant cardiovascular effects on the human body and increase the risk of developing cardiac health issues. This study aimed to investigate the autonomic nervous system (ANS) responses of healthy individuals during simulated dives in hyperbaric chambers and explore the effects of the humid environment on these responses. Electrocardiographic- and heart-rate-variability (

H R V

)-derived indices were analyzed, and their statistical ranges were compared at different depths during simulated immersions under dry and humid conditions. The results showed that humidity significantly affected the ANS responses of the subjects, leading to reduced parasympathetic activity and increased sympathetic dominance. The power of the high-frequency band of the

H R V

after removing the influence of respiration,

\bar{P_{H F ⊥}}

, and the number of pairs of successive normal-to-normal intervals that differ by more than 50 ms divided by the total number of normal-to-normal intervals,

\bar{p N N 50}

, indices were found to be the most informative in distinguishing the ANS responses of subjects between the two datasets. Additionally, the statistical ranges of the

H R V

indices were calculated, and the classification of subjects as “normal” or “abnormal” was determined based on these ranges. The results showed that the ranges were effective at identifying abnormal ANS responses, indicating the potential use of these ranges as a reference for monitoring the activity of divers and avoiding future immersions if many indices are out of the normal ranges. The bagging method was also used to include some variability in the datasets’ ranges, and the classification results showed that the ranges computed without proper bagging represent reality and its associated variability. Overall, this study provides valuable insights into the ANS responses of healthy individuals during simulated dives in hyperbaric chambers and the effects of humidity on these responses. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Graphical abstract

15 pages, 6469 KiB

Open AccessArticle

Accuracy of a Smartwatch to Assess Heart Rate Monitoring and Atrial Fibrillation in Stroke Patients

by Claudia Meza, Jesus Juega, Jaume Francisco, Alba Santos, Laura Duran, Maite Rodriguez, Jose Alvarez-Sabin, Laia Sero, Xavier Ustrell, Saima Bashir, Joaquín Serena, Yolanda Silva, Carlos Molina and Jorge Pagola

Sensors 2023, 23(10), 4632; https://doi.org/10.3390/s23104632 - 10 May 2023

Cited by 4 | Viewed by 7973

Abstract

(1) Background: Consumer smartwatches may be a helpful tool to screen for atrial fibrillation (AF). However, validation studies on older stroke patients remain scarce. The aim of this pilot study from RCT NCT05565781 was to validate the resting heart rate (HR) measurement and [...] Read more.

(1) Background: Consumer smartwatches may be a helpful tool to screen for atrial fibrillation (AF). However, validation studies on older stroke patients remain scarce. The aim of this pilot study from RCT NCT05565781 was to validate the resting heart rate (HR) measurement and the irregular rhythm notification (IRN) feature in stroke patients in sinus rhythm (SR) and AF. (2) Methods: Resting clinical HR measurements (every 5 min) were assessed using continuous bedside ECG monitoring (CEM) and the Fitbit Charge 5 (FC5). IRNs were gathered after at least 4 h of CEM. Lin’s concordance correlation coefficient (CCC), Bland-Altman analysis, and mean absolute percentage error (MAPE) were used for agreement and accuracy assessment. (3) Results: In all, 526 individual pairs of measurements were obtained from 70 stroke patients—age 79.4 years (SD ± 10.2), 63% females, BMI 26.3 (IQ 22.2–30.5), and NIHSS score 8 (IQR 1.5–20). The agreement between the FC5 and CEM was good (CCC 0.791) when evaluating paired HR measurements in SR. Meanwhile, the FC5 provided weak agreement (CCC 0.211) and low accuracy (MAPE 16.48%) when compared to CEM recordings in AF. Regarding the accuracy of the IRN feature, analysis found a low sensitivity (34%) and high specificity (100%) for detecting AF. (4) Conclusion: The FC5 was accurate at assessing the HR during SR, but the accuracy during AF was poor. In contrast, the IRN feature was acceptable for guiding decisions regarding AF screening in stroke patients. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

12 pages, 2063 KiB

Open AccessArticle

An Experimental Method for Bio-Signal Denoising Using Unconventional Sensors

by Rodrigo Aviles-Espinosa, Henry Dore and Elizabeth Rendon-Morales

Sensors 2023, 23(7), 3527; https://doi.org/10.3390/s23073527 - 28 Mar 2023

Cited by 2 | Viewed by 2074

Abstract

In bio-signal denoising, current methods reported in the literature consider purely simulated environments, requiring high computational powers and signal processing algorithms that may introduce signal distortion. To achieve an efficient noise reduction, such methods require previous knowledge of the noise signals or to [...] Read more.

In bio-signal denoising, current methods reported in the literature consider purely simulated environments, requiring high computational powers and signal processing algorithms that may introduce signal distortion. To achieve an efficient noise reduction, such methods require previous knowledge of the noise signals or to have certain periodicity and stability, making the noise estimation difficult to predict. In this paper, we solve these challenges through the development of an experimental method applied to bio-signal denoising using a combined approach. This is based on the implementation of unconventional electric field sensors used for creating a noise replica required to obtain the ideal Wiener filter transfer function and achieve further noise reduction. This work aims to investigate the suitability of the proposed approach for real-time noise reduction affecting bio-signal recordings. The experimental evaluation presented here considers two scenarios: (a) human bio-signals trials including electrocardiogram, electromyogram and electrooculogram; and (b) bio-signal recordings from the MIT-MIH arrhythmia database. The performance of the proposed method is evaluated using qualitative criteria (i.e., power spectral density) and quantitative criteria (i.e., signal-to-noise ratio and mean square error) followed by a comparison between the proposed methodology and state of the art denoising methods. The results indicate that the combined approach proposed in this paper can be used for noise reduction in electrocardiogram, electromyogram and electrooculogram signals, achieving noise attenuation levels of 26.4 dB, 21.2 dB and 40.8 dB, respectively. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

10 pages, 1636 KiB

Open AccessArticle

The r’-Wave Algorithm: A New Diagnostic Tool to Predict the Diagnosis of Brugada Syndrome after a Sodium Channel Blocker Provocation Test

by Giampaolo Vetta, Antonio Parlavecchio, Lorenzo Pistelli, Paolo Desalvo, Armando Lo Savio, Michele Magnocavallo, Rodolfo Caminiti, Anna Tribuzio, Alessandro Vairo, Diego La Maestra, Francesco Vetta, Giuseppe Dattilo, Francesco Luzza, Gianluca Di Bella, Roberta Rossini, Domenico Giovanni Della Rocca and Pasquale Crea

Sensors 2023, 23(6), 3159; https://doi.org/10.3390/s23063159 - 16 Mar 2023

Viewed by 2016

Abstract

A diagnosis of Brugada syndrome (BrS) is based on the presence of a type 1 electrocardiogram (ECG) pattern, either spontaneously or after a Sodium Channel Blocker Provocation Test (SCBPT). Several ECG criteria have been evaluated as predictors of a positive SCBPT, such as [...] Read more.

A diagnosis of Brugada syndrome (BrS) is based on the presence of a type 1 electrocardiogram (ECG) pattern, either spontaneously or after a Sodium Channel Blocker Provocation Test (SCBPT). Several ECG criteria have been evaluated as predictors of a positive SCBPT, such as the β-angle, the α-angle, the duration of the base of the triangle at 5 mm from the r’-wave (DBT- 5 mm), the duration of the base of the triangle at the isoelectric line (DBT- iso), and the triangle base/height ratio. The aim of our study was to test all previously proposed ECG criteria in a large cohort study and to evaluate an r’-wave algorithm for predicting a BrS diagnosis after an SCBPT. We enrolled all patients who consecutively underwent SCBPT using flecainide from January 2010 to December 2015 in the test cohort and from January 2016 to December 2021 in the validation cohort. We included the ECG criteria with the best diagnostic accuracy in relation to the test cohort in the development of the r’-wave algorithm (β-angle, α-angle, DBT- 5 mm, and DBT- iso.) Of the total of 395 patients enrolled, 72.4% were male and the average age was 44.7 ± 13.5 years. Following the SCBPTs, 24.1% of patients (n = 95) were positive and 75.9% (n = 300) were negative. ROC analysis of the validation cohort showed that the AUC of the r’-wave algorithm (AUC: 0.92; CI 0.85–0.99) was significantly better than the AUC of the β-angle (AUC: 0.82; 95% CI 0.71–0.92), the α-angle (AUC: 0.77; 95% CI 0.66–0.90), the DBT- 5 mm (AUC: 0.75; 95% CI 0.64–0.87), the DBT- iso (AUC: 0.79; 95% CI 0.67–0.91), and the triangle base/height (AUC: 0.61; 95% CI 0.48–0.75) (p < 0.001), making it the best predictor of a BrS diagnosis after an SCBPT. The r’-wave algorithm with a cut-off value of ≥2 showed a sensitivity of 90% and a specificity of 83%. In our study, the r’-wave algorithm was proved to have the best diagnostic accuracy, compared with single electrocardiographic criteria, in predicting the diagnosis of BrS after provocative testing with flecainide. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

12 pages, 1747 KiB

Open AccessArticle

Wireless Single-Lead versus Standard 12-Lead ECG, for ST-Segment Deviation during Adenosine Cardiac Stress Scintigraphy

by Luna Fabricius Ekenberg, Dan Eik Høfsten, Søren M. Rasmussen, Jesper Mølgaard, Philip Hasbak, Helge B. D. Sørensen, Christian S. Meyhoff and Eske K. Aasvang

Sensors 2023, 23(6), 2962; https://doi.org/10.3390/s23062962 - 9 Mar 2023

Cited by 2 | Viewed by 2870

Abstract

Wearable wireless electrocardiographic (ECG) monitoring is well-proven for arrythmia detection, but ischemia detection accuracy is not well-described. We aimed to assess the agreement of ST-segment deviation from single- versus 12-lead ECG and their accuracy for the detection of reversible ischemia. Bias and limits [...] Read more.

Wearable wireless electrocardiographic (ECG) monitoring is well-proven for arrythmia detection, but ischemia detection accuracy is not well-described. We aimed to assess the agreement of ST-segment deviation from single- versus 12-lead ECG and their accuracy for the detection of reversible ischemia. Bias and limits of agreement (LoA) were calculated between maximum deviations in ST segments from single- and 12-lead ECG during ⁸²Rb PET-myocardial cardiac stress scintigraphy. Sensitivity and specificity for reversible anterior-lateral myocardial ischemia detection were assessed for both ECG methods, using perfusion imaging results as a reference. Out of 110 patients included, 93 were analyzed. The maximum difference between single- and 12-lead ECG was seen in II (−0.019 mV). The widest LoA was seen in V5, with an upper LoA of 0.145 mV (0.118 to 0.172) and a lower LoA of −0.155 mV (−0.182 to −0.128). Ischemia was seen in 24 patients. Single-lead and 12-lead ECG both had poor accuracy for the detection of reversible anterolateral ischemia during the test: single-lead ECG had a sensitivity of 8.3% (1.0–27.0%) and specificity of 89.9% (80.2–95.8%), and 12-lead ECG a sensitivity of 12.5% (3.0–34.4%) and a specificity of 91.3% (82.0–96.7%). In conclusion, agreement was within predefined acceptable criteria for ST deviations, and both methods had high specificity but poor sensitivity for the detection of anterolateral reversible ischemia. Additional studies must confirm these results and their clinical relevance, especially in the light of the poor sensitivity for detecting reversible anterolateral cardiac ischemia. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

13 pages, 3597 KiB

Open AccessArticle

Using a Smartwatch to Record Precordial Electrocardiograms: A Validation Study

by Joske van der Zande, Marc Strik, Rémi Dubois, Sylvain Ploux, Saer Abu Alrub, Théo Caillol, Mathieu Nasarre, Dirk W. Donker, Eline Oppersma and Pierre Bordachar

Sensors 2023, 23(5), 2555; https://doi.org/10.3390/s23052555 - 25 Feb 2023

Cited by 9 | Viewed by 4182

Abstract

Smartwatches that support the recording of a single-lead electrocardiogram (ECG) are increasingly being used beyond the wrist, by placement on the ankle and on the chest. However, the reliability of frontal and precordial ECGs other than lead I is unknown. This clinical validation [...] Read more.

Smartwatches that support the recording of a single-lead electrocardiogram (ECG) are increasingly being used beyond the wrist, by placement on the ankle and on the chest. However, the reliability of frontal and precordial ECGs other than lead I is unknown. This clinical validation study assessed the reliability of an Apple Watch (AW) to obtain conventional frontal and precordial leads as compared to standard 12-lead ECGs in both subjects without known cardiac anomalies and patients with underlying heart disease. In 200 subjects (67% with ECG anomalies), a standard 12-lead ECG was performed, followed by AW recordings of the standard Einthoven leads (leads I, II, and III) and precordial leads V1, V3, and V6. Seven parameters (P, QRS, ST, and T-wave amplitudes, PR, QRS, and QT intervals) were compared through a Bland–Altman analysis, including the bias, absolute offset, and 95% limits of agreement. AW-ECGs recorded on the wrist but also beyond the wrist had similar durations and amplitudes compared to standard 12-lead ECGs. Significantly greater amplitudes were measured by the AW for R-waves in precordial leads V1, V3, and V6 (+0.094 mV, +0.149 mV, +0.129 mV, respectively, all p < 0.001), indicating a positive bias for the AW. AW can be used to record frontal, and precordial ECG leads, paving the way for broader clinical applications. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

29 pages, 7849 KiB

Open AccessArticle

A Deep Learning Architecture Using 3D Vectorcardiogram to Detect R-Peaks in ECG with Enhanced Precision

by Maroua Mehri, Guillaume Calmon, Freddy Odille and Julien Oster

Sensors 2023, 23(4), 2288; https://doi.org/10.3390/s23042288 - 18 Feb 2023

Cited by 4 | Viewed by 3598

Abstract

Providing reliable detection of QRS complexes is key in automated analyses of electrocardiograms (ECG). Accurate and timely R-peak detections provide a basis for ECG-based diagnoses and to synchronize radiologic, electrophysiologic, or other medical devices. Compared with classical algorithms, deep learning (DL) architectures have [...] Read more.

Providing reliable detection of QRS complexes is key in automated analyses of electrocardiograms (ECG). Accurate and timely R-peak detections provide a basis for ECG-based diagnoses and to synchronize radiologic, electrophysiologic, or other medical devices. Compared with classical algorithms, deep learning (DL) architectures have demonstrated superior accuracy and high generalization capacity. Furthermore, they can be embedded on edge devices for real-time inference. 3D vectorcardiograms (VCG) provide a unifying framework for detecting R-peaks regardless of the acquisition strategy or number of ECG leads. In this article, a DL architecture was demonstrated to provide enhanced precision when trained and applied on 3D VCG, with no pre-processing nor post-processing steps. Experiments were conducted on four different public databases. Using the proposed approach, high F1-scores of 99.80% and 99.64% were achieved in leave-one-out cross-validation and cross-database validation protocols, respectively. False detections, measured by a precision of 99.88% or more, were significantly reduced compared with recent state-of-the-art methods tested on the same databases, without penalty in the number of missed peaks, measured by a recall of 99.39% or more. This approach can provide new applications for devices where precision, or positive predictive value, is essential, for instance cardiac magnetic resonance imaging. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

13 pages, 1746 KiB

Open AccessArticle

Improved Estimation of Exercise Intensity Thresholds by Combining Dual Non-Invasive Biomarker Concepts: Correlation Properties of Heart Rate Variability and Respiratory Frequency

by Bruce Rogers, Marcelle Schaffarczyk and Thomas Gronwald

Sensors 2023, 23(4), 1973; https://doi.org/10.3390/s23041973 - 10 Feb 2023

Cited by 5 | Viewed by 3803

Abstract

Identifying exercise intensity boundaries has been shown to be important during endurance training for performance enhancement and rehabilitation. Unfortunately, even though surrogate markers show promise when assessed on a group level, substantial deviation from gold standards can be present in each individual. The [...] Read more.

Identifying exercise intensity boundaries has been shown to be important during endurance training for performance enhancement and rehabilitation. Unfortunately, even though surrogate markers show promise when assessed on a group level, substantial deviation from gold standards can be present in each individual. The aim of this study was to evaluate whether combining two surrogate intensity markers improved this agreement. Electrocardiogram (ECG) and gas exchange data were obtained from 21 participants who performed an incremental cycling ramp to exhaustion and evaluated for first (VT1) and second (VT2) ventilatory thresholds, heart rate (HR) variability (HRV), and ECG derived respiratory frequency (EDR). HRV thresholds (HRVT) were based on the non-linear index a1 of a Detrended Fluctuation Analysis (DFA a1) and EDR thresholds (EDRT) upon the second derivative of the sixth-order polynomial of EDR over time. The average of HRVT and EDRT HR was set as the combined threshold (Combo). Mean VT1 was reached at a HR of 141 ± 15, HRVT1 at 152 ± 14 (p < 0.001), EDRT1 at 133 ± 12 (p < 0.001), and Combo1 at 140 ± 13 (p = 0.36) bpm with Pearson’s r of 0.83, 0.78, and 0.84, respectively, for comparisons to VT1. A Bland–Altman analysis showed mean biases of 8.3 ± 7.9, −8.3 ± 9.5, and −1.7 ± 8.3 bpm, respectively. A mean VT2 was reached at a HR of 165 ± 13, HRVT2 at 167 ± 10 (p = 0.89), EDRT2 at 164 ± 14 (p = 0.36), and Combo2 at 164 ± 13 (p = 0.59) bpm with Pearson’s r of 0.58, 0.95, and 0.94, respectively, for comparisons to VT2. A Bland–Altman analysis showed mean biases of −0.3 ± 8.9, −1.0 ± 4.6, and −0.6 ± 4.6 bpm, respectively. Both the DFA a1 and EDR intensity thresholds based on HR taken individually had moderate agreement to targets derived through gas exchange measurements. By combining both non-invasive approaches, there was improved correlation, reduced bias, and limits of agreement to the respective corresponding HRs at VT1 and VT2. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

17 pages, 8852 KiB

Open AccessArticle

Non-Contact Monitoring of ECG in the Home Environment—Selecting Optimal Electrode Configuration

by Adam Bujnowski, Kamil Osiński, Piotr Przystup and Jerzy Wtorek

Sensors 2022, 22(23), 9475; https://doi.org/10.3390/s22239475 - 4 Dec 2022

Cited by 2 | Viewed by 2421

Abstract

Capacitive electrocardiography (cECG) is most often used in wearable or embedded measurement systems. The latter is considered in the paper. An optimal electrocardiographic lead, as an individual feature, was determined based on model studies. It was defined as the possibly highest value of [...] Read more.

Capacitive electrocardiography (cECG) is most often used in wearable or embedded measurement systems. The latter is considered in the paper. An optimal electrocardiographic lead, as an individual feature, was determined based on model studies. It was defined as the possibly highest value of the R-wave amplitude measured on the back of the examined person. The lead configuration was also analyzed in terms of minimizing its susceptibility to creating motion artifacts. It was found that the direction of the optimal lead coincides with the electrical axis of the heart. Moreover, the electrodes should be placed in the areas preserving the greatest voltage and at the same time characterized by the lowest gradient of the potential. Experimental studies were conducted using the developed measurement system on a group of 14 people. The ratio of the R-wave amplitude (as measured on the back and chest, using optimal leads) was less than 1 while the SNR reached at least 20 dB. These parameters allowed for high-quality QRS complex detection with a PPV of 97%. For the “worst” configurations of the leads, the signals measured were practically uninterpretable. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

13 pages, 1464 KiB

Open AccessArticle

Utility of the Full ECG Waveform for Stress Classification

by Katya Arquilla, Andrea K. Webb and Allison P. Anderson

Sensors 2022, 22(18), 7034; https://doi.org/10.3390/s22187034 - 17 Sep 2022

Cited by 6 | Viewed by 3853

Abstract

The detection of psychological stress using the electrocardiogram (ECG) signal is most commonly based on the detection of the R peak—the most prominent part of the ECG waveform—and the heart rate variability (HRV) measurements derived from it. For stress detection algorithms focused on [...] Read more.

The detection of psychological stress using the electrocardiogram (ECG) signal is most commonly based on the detection of the R peak—the most prominent part of the ECG waveform—and the heart rate variability (HRV) measurements derived from it. For stress detection algorithms focused on short-duration time windows, there is potential benefit in including HRV features derived from the detection of smaller peaks within the ECG waveform: the P, Q, S, and T waves. However, the potential drawback of using these small peaks is their smaller magnitude and subsequent susceptibility to noise, making them more difficult to reliably detect. In this work, we demonstrate the potential benefits of including smaller waves within binary stress classification using a pre-existing data set of ECG recordings from 57 participants (aged 18–40) with a self-reported fear of spiders during exposure to videos of spiders. We also present an analysis of the performance of an automated peak detection algorithm and the reliability of detection for each of the smaller parts of the ECG waveform. We compared two models, one with only R peak features and one with small peak features. They were similar in precision, recall, F1, area under ROC curve (AUC), and accuracy, with the greatest differences less than the standard deviations of each metric. There was a significant difference in the Akaike Information Criterion (AIC), which represented the information loss of the model. The inclusion of novel small peak features made the model

4.29 \times 10^{28}

times more probable to minimize the information loss, and the small peak features showed higher regression coefficients than the R peak features, indicating a stronger relationship with acute psychological stress. This difference and further analysis of the novel features suggest that small peak intervals could be indicative of independent processes within the heart, reflecting a psychophysiological response to stress that has not yet been leveraged in stress detection algorithms. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

22 pages, 8063 KiB

Open AccessArticle

Deep-Learning-Based Estimation of the Spatial QRS-T Angle from Reduced-Lead ECGs

by Ana Santos Rodrigues, Rytis Augustauskas, Mantas Lukoševičius, Pablo Laguna and Vaidotas Marozas

Sensors 2022, 22(14), 5414; https://doi.org/10.3390/s22145414 - 20 Jul 2022

Cited by 2 | Viewed by 3547

Abstract

The spatial QRS-T angle is a promising health indicator for risk stratification of sudden cardiac death (SCD). Thus far, the angle is estimated solely from 12-lead electrocardiogram (ECG) systems uncomfortable for ambulatory monitoring. Methods to estimate QRS-T angles from reduced-lead ECGs registered with [...] Read more.

The spatial QRS-T angle is a promising health indicator for risk stratification of sudden cardiac death (SCD). Thus far, the angle is estimated solely from 12-lead electrocardiogram (ECG) systems uncomfortable for ambulatory monitoring. Methods to estimate QRS-T angles from reduced-lead ECGs registered with consumer healthcare devices would, therefore, facilitate ambulatory monitoring. (1) Objective: Develop a method to estimate spatial QRS-T angles from reduced-lead ECGs. (2) Approach: We designed a deep learning model to locate the QRS and T wave vectors necessary for computing the QRS-T angle. We implemented an original loss function to guide the model in the 3D space to search for each vector’s coordinates. A gradual reduction of ECG leads from the largest publicly available dataset of clinical 12-lead ECG recordings (PTB-XL) is used for training and validation. (3) Results: The spatial QRS-T angle can be estimated from leads {I, II, aVF, V2} with sufficient accuracy (absolute mean and median errors of 11.4° and 7.3°) for detecting abnormal angles without sacrificing patient comfortability. (4) Significance: Our model could enable ambulatory monitoring of spatial QRS-T angles using patch- or textile-based ECG devices. Populations at risk of SCD, like chronic cardiac and kidney disease patients, might benefit from this technology. Full article

(This article belongs to the Special Issue ECG Signal Processing and Analysis, Computational Technology and Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

ECG Signal Processing and Analysis, Computational Technology and Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Benefits of Publishing in a Special Issue

Related Special Issue

Published Papers (12 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI