Respiratory Rate Estimation Using U-Net-Based Cascaded Framework From Electrocardiogram and Seismocardiogram Signals

An efficient model for extracting respiratory and blood oxygen saturation data from photoplethysmogram signals by removing motion artifacts using heuristic-aided ensemble learning model

Patients with surgical, pulmonary, and cardiac problems, continual monitoring of Oxygen Saturation of a Person (SpO2) and Respiratory Rate (RR) is essential. Similarly, the persons with cardiopulmonary health issues, RR estimation is crucial. The performance of the ventilator assistance and lung medicines are evaluated using SpO2 and RR. For the persons, those who are living alone with respiratory illnesses need a compulsory estimation of RR. In case of serious illness, the RR might face abrupt changes. The immobility of the disturbance and RR makes the RR evaluation from the PhotoPlethysmoGraphic (PPG) signals is a difficult challenge. So, an efficient RR and SpO2 estimation framework from the PPG signal using the deep learning method is developed in this paper. At first, the PPG signal is collected from standard data sources. The collected PPG signals undergo signal pre-processing. The pre-processing procedures include Motion Artifacts (MA) removal and filtering techniques. The pre-processed signals are split into distinct windows. From the split windows of the signals, the spectral features, RR, and Respiratory Peak Variance (RPV) features are extracted. The retrieved features are selected optimally with the help of Advanced Golden Tortoise Beetle Optimizer (AGTBO). The weights are chosen optimally with the same AGTBO. The optimally selected features are fused with the optimal features to get the weighted optimal features. These weighted optimal features are fed into the Ensemble Learning-based RR and SpO2 Estimation Network (ELRR-SpO2EN). The ensemble learning model is developed by combining Multilayer Perceptron (MLP), AdaBoost, and Attention-based Long Short Term Memory (A-LSTM). The performance of the developed RR and SpO2 estimation model is compared with other existing techniques. The experimental analysis results revealed that the proposed AGTBO-ELRR-SpO2EN model attained 96 % accuracy for the second dataset, which is higher than the conventional models such as MLP (90 %), Adaboost (92 %), A-LSTM (92 %), and MLP-ADA-ALSTM (94 %). Thus, it has been confirmed that the designed RR and SpO2 estimation framework from PPG signals is more efficient than the other conventional models.

An open-access simultaneous electrocardiogram and phonocardiogram database

Objective.The EPHNOGRAM project aimed to develop a low-cost, low-power device for simultaneous electrocardiogram (ECG) and phonocardiogram (PCG) recording, with additional channels for environmental audio to enhance PCG through active noise cancellation. The objective was to study multimodal electro-mechanical activities of the heart, offering insights into the differences and synergies between these modalities during various cardiac activity levels.Approach.We developed and tested several hardware prototypes of a simultaneous ECG-PCG acquisition device. Using this technology, we collected simultaneous ECG and PCG data from 24 healthy adults during different physical activities, including resting, walking, running, and stationary biking, in an indoor fitness center. The data were annotated using a robust software that we developed for detecting ECG R-peaks and PCG S1 and S2 components, and overseen by a human expert. We also developed machine learning models using ECG-based, PCG-based, and joint ECG-PCG features, like R-R and S1-S2 intervals, to classify physical activities and analyze electro-mechanical dynamics.Main results.The results show a significant coupling between ECG and PCG components, especially during high-intensity exercise. Notable micro-variations in S2-based heart rate show differences in the heart's electrical and mechanical functions. The Lomb-Scargle periodogram and approximate entropy analyses confirm the higher volatility of S2-based heart rate compared to ECG-based heart rate. Correlation analysis shows stronger coupling between R-R and R-S1 intervals during high-intensity activities. Hybrid ECG-PCG features, like the R-S2 interval, were identified as more informative for physical activity classification through mRMR feature selection and SHAP value analysis.Significance.The EPHNOGRAM database, is available on PhysioNet. The database enhances our understanding of cardiac function, enabling future studies on the heart's mechanical and electrical interrelationships. The results of this study can contribute to improved cardiac condition diagnoses. Additionally, the designed hardware has the potential for integration into wearable devices and the development of multimodal stress test technologies.

A Deep Learning Approach to Using Wearable Seismocardiography (SCG) for Diagnosing Aortic Valve Stenosis and Predicting Aortic Hemodynamics Obtained by 4D Flow MRI

In this paper, we explored the use of deep learning for the prediction of aortic flow metrics obtained using 4-dimensional (4D) flow magnetic resonance imaging (MRI) using wearable seismocardiography (SCG) devices. 4D flow MRI provides a comprehensive assessment of cardiovascular hemodynamics, but it is costly and time-consuming. We hypothesized that deep learning could be used to identify pathological changes in blood flow, such as elevated peak systolic velocity ([Formula: see text]) in patients with heart valve diseases, from SCG signals. We also investigated the ability of this deep learning technique to differentiate between patients diagnosed with aortic valve stenosis (AS), non-AS patients with a bicuspid aortic valve (BAV), non-AS patients with a mechanical aortic valve (MAV), and healthy subjects with a normal tricuspid aortic valve (TAV). In a study of 77 subjects who underwent same-day 4D flow MRI and SCG, we found that the [Formula: see text] values obtained using deep learning and SCGs were in good agreement with those obtained by 4D flow MRI. Additionally, subjects with non-AS TAV, non-AS BAV, non-AS MAV, and AS could be classified with ROC-AUC (area under the receiver operating characteristic curves) values of 92%, 95%, 81%, and 83%, respectively. This suggests that SCG obtained using low-cost wearable electronics may be used as a supplement to 4D flow MRI exams or as a screening tool for aortic valve disease.

Deep learning with fetal ECG recognition

Objective.Independent component analysis (ICA) is widely used in the extraction of fetal ECG (FECG). However, the amplitude, order, and positive or negative values of the ICA results are uncertain. The main objective is to present a novel approach to FECG recognition by using a deep learning strategy.Approach.A cross-domain consistent convolutional neural network (CDC-Net) is developed for the task of FECG recognition. The output of the ICA algorithm is used as input to the CDC-Net and the CDC-Net identifies which channel's signal is the target FECG.Main results.Signals from two databases are used to test the efficiency of the proposed method. The proposed deep learning method exhibits good performance on FECG recognition. Specifically, the Precision, Recall and F1-score of the proposed method on the ADFECGDB database are 91.69%, 91.37% and 91.52%, respectively. The Precision, Recall and F1-score of the proposed method on the Daisy database are 97.85%, 97.42% and 97.63%, respectively.Significance. This study is a proof of concept that the proposed method can automatically recognize the FECG signals in multi-channel ECG data. The development of FECG recognition technology contributes to automated FECG monitoring.

Towards Quantifying Stress in Patients with a History of Myocardial Infarction: Validating ECG-Derived Patch Features

Patients with prior myocardial infarction (MI) have an increased risk of experiencing a secondary event which is exacerbated by mental stress. Our team has developed a miniaturized patch with the capability to capture electrocardiogram (ECG), seismocardiogram (SCG) and photoplethysmogram (PPG) signals which may provide multimodal information to characterize stress responses within the post-MI population in ambulatory settings. As ECG-derived features have been shown to be informative in assessing the risk of MI, a critical first step is to ensure that the patch ECG features agree with gold-standard devices, such as the Biopac. However, this is yet to be done in this population. We, thus, performed a comparative analysis between ECG-derived features (heart rate (HR) and heart rate variability (HRV)) of the patch and Biopac in the context of stress. Our dataset contained post-MI and healthy control subjects who participated in a public speaking challenge. Regression analyses for patch and Biopac HR and HRV features (RMSSD, pNN50, SD1/SD2, and LF/HF) were all significant (p<0.001) and had strong positive correlations (r>0.9). Additionally, Bland-Altman analyses for most features showed tight limits of agreement: 0.999 bpm (HR), 11.341 ms (RMSSD), 0.07% (pNN50), 0.146 ratio difference (SD1/SD2), 0.750 ratio difference (LF/HF).Clinical relevance- This work demonstrates that ECG-derived features obtained from the patch and Biopac are in agreement, suggesting the clinical utility of the patch in deriving quantitative metrics of physiology during stress in post-MI patients. This has the potential to improve post-MI patients' outcomes, but needs to be further evaluated.

Dual-Sensor Signals Based Exact Gaussian Process-Assisted Hybrid Feature Extraction and Weighted Feature Fusion for Respiratory Rate and Uncertainty Estimations

Accurately estimating respiratory rate (RR) has become essential for patients and the elderly. Hence, we propose a novel method that uses exact Gaussian process regression (EGPR)-assisted hybrid feature extraction and feature fusion based on photoplethysmography and electrocardiogram signals to improve the reliability of accurate RR and uncertainty estimations. First, we obtain the power spectral features and use the multi-phase feature model to compensate for insufficient input data. Then, we combine four different feature sets and choose features with high weights using a robust neighbor component analysis. The proposed EGPR algorithm provides a confidence interval representing the uncertainty. Therefore, the proposed EGPR algorithm, including hybrid feature extraction and weighted feature fusion, is an excellent model with improved reliability for accurate RR estimation. Furthermore, the proposed EGPR methodology is likely the only one currently available that provides highly stable variation and confidence intervals. The proposed EGPR-MF, 0.993 breath per minute (bpm), and EGPR-feature fusion, 1.064 (bpm), show the lowest mean absolute error compared to the other models.

Identification of Characteristic Points in Multivariate Physiological Signals by Sensor Fusion and Multi-Task Deep Networks

Identification of characteristic points in physiological signals, such as the peak of the R wave in the electrocardiogram and the peak of the systolic wave of the photopletismogram, is a fundamental step for the quantification of clinical parameters, such as the pulse transit time. In this work, we presented a novel neural architecture, called eMTUnet, to automate point identification in multivariate signals acquired with a chest-worn device. The eMTUnet consists of a single deep network capable of performing three tasks simultaneously: (i) localization in time of characteristic points (labeling task), (ii) evaluation of the quality of signals (classification task); (iii) estimation of the reliability of classification (reliability task). Preliminary results in overnight monitoring showcased the ability to detect characteristic points in the four signals with a recall index of about 1.00, 0.90, 0.90, and 0.80, respectively. The accuracy of the signal quality classification was about 0.90, on average over four different classes. The average confidence of the correctly classified signals, against the misclassifications, was 0.93 vs. 0.52, proving the worthiness of the confidence index, which may better qualify the point identification. From the achieved outcomes, we point out that high-quality segmentation and classification are both ensured, which brings the use of a multi-modal framework, composed of wearable sensors and artificial intelligence, incrementally closer to clinical translation.