Fig 4 - uploaded by Vincenzo Randazzo
Content may be subject to copyright.
Source publication
This paper presents a comparison between two recurrent neural networks (RNN) for arterial blood pressure (ABP) estimation. ABP is a parameter closely related to the cardiac activity, for this reason its monitoring implies decreasing the risk of heart disease. In order to predict the ABP values (both systolic and diastolic), electrocardiographic (EC...
Context in source publication
Context 1
... purpose of Neural Network Output-Error (NNOE) is the identification of nonlinear dynamic systems in stochastic environment [15]. Fig. 4 describes the procedure that must be executed when attempting to identify a dynamical system. The experimental phase is represented by the description of the dataset í µí± í µí± , which describes the entire system in its operating region with a proper choice of sampling ...
Citations
... In particular, the application of AI in ECG analysis has recently gained tremendous momentum due to the fact that ECG constitutes an ideal substrate for AI application, being a low-cost and widely adopted cardiological tool [18]. Different groups have reported favorable results obtained with AI-based ECG analysis in several clinical settings, such as prediction of underlying atrial fibrillation in patients presenting with sinus rhythm [19], arterial blood pressure estimation [20][21][22][23][24], estimation of age and sex [25], prediction of underlying cardiac contractile dysfunction [26] and of hyperkalemia [27], arrhythmia classification [28-30], detection of hypertrophic cardiomyopathy [31], early detection of cardiovascular autonomic neuropathy [32], drug development [33], and, more generally, heartbeat classification [34][35][36]. The high-level discrimination capabilities of such AI models, which have shown very good predictive performances [37][38][39], together with the quickness, availability, and cost-effectiveness of the ECG, highlight the high potential of AI-based ECG analysis. ...
Short QT syndrome (SQTS) is an inherited cardiac ion-channel disease related to an increased risk of sudden cardiac death (SCD) in young and otherwise healthy individuals. SCD is often the first clinical presentation in patients with SQTS. However, arrhythmia risk stratification is presently unsatisfactory in asymptomatic patients. In this context, artificial intelligence-based electrocardiogram (ECG) analysis has never been applied to refine risk stratification in patients with SQTS. The purpose of this study was to analyze ECGs from SQTS patients with the aid of different AI algorithms to evaluate their ability to discriminate between subjects with and without documented life-threatening arrhythmic events. The study group included 104 SQTS patients, 37 of whom had a documented major arrhythmic event at presentation and/or during follow-up. Thirteen ECG features were measured independently by three expert cardiologists; then, the dataset was randomly divided into three subsets (training, validation, and testing). Five shallow neural networks were trained, validated, and tested to predict subject-specific class (non-event/event) using different subsets of ECG features. Additionally, several deep learning and machine learning algorithms, such as Vision Transformer, Swin Transformer, MobileNetV3, EfficientNetV2, ConvNextTiny, Capsule Networks, and logistic regression were trained, validated, and tested directly on the scanned ECG images, without any manual feature extraction. Furthermore, a shallow neural network, a 1-D transformer classifier, and a 1-D CNN were trained, validated, and tested on ECG signals extracted from the aforementioned scanned images. Classification metrics were evaluated by means of sensitivity, specificity, positive and negative predictive values, accuracy, and area under the curve. Results prove that artificial intelligence can help clinicians in better stratifying risk of arrhythmia in patients with SQTS. In particular, shallow neural networks’ processing features showed the best performance in identifying patients that will not suffer from a potentially lethal event. This could pave the way for refined ECG-based risk stratification in this group of patients, potentially helping in saving the lives of young and otherwise healthy individuals.
... From a machine learning perspective, current approaches that use PPG for BP prediction can broadly be categorized into approaches based on extracting hand-crafted features [6,10,35] and approaches that employ the entire signal and sometimes also its derivatives [25,29]. The latter are usually based on a certain deep neural network (DNN) architecture. ...
... The latter are usually based on a certain deep neural network (DNN) architecture. In DNN, these signals or their spectral representations are then usually either used directly in an endto-end learning scheme to predict BP from the shape information [25] or are transformed into a spectrogram beforehand [38]. Even hybrid approaches have already been investigated [3,29]. ...
... From a machine learning perspective, current approaches that use PPG for BP prediction can broadly be categorized into approaches based on extracting hand-crafted features [6,10,35] and approaches that employ the entire signal and sometimes also its derivatives [25,29]. The latter are usually based on a certain deep neural network (DNN) architecture. ...
... The latter are usually based on a certain deep neural network (DNN) architecture. In DNN, these signals or their spectral representations are then usually either used directly in an endto-end learning scheme to predict BP from the shape information [25] or are transformed into a spectrogram beforehand [38]. Even hybrid approaches have already been investigated [3,29]. ...
Photoplethysmographic (PPG) signals offer diagnostic potential beyond heart rate analysis or blood oxygen level monitoring. In the recent past, research focused extensively on non-invasive PPG-based approaches to blood pressure (BP) estimation. These approaches can be subdivided into regression and classification methods. The latter assign PPG signals to predefined BP intervals that represent clinically relevant ranges. The former predict systolic (SBP) and diastolic (DBP) BP as continuous variables and are of particular interest to the research community. However, the reported accuracies of BP regression methods vary widely among publications with some authors even questioning the feasibility of PPG-based BP regression altogether. In our work, we compare BP regression and classification approaches. We argue that BP classification might provide diagnostic value that is equivalent to regression in many clinically relevant scenarios while being similar or even superior in terms of performance. We compare several established neural architectures using publicly available PPG data for SBP regression and classification with and without personalization using subject-specific data. We found that classification and regression models perform similar before personalization. However, after personalization, the accuracy of classification based methods outperformed regression approaches. We conclude that BP classification might be preferable over BP regression in certain scenarios where a coarser segmentation of the BP range is sufficient.
... Continuous cuff-less measurement devices on the other hand are mostly invasive and pose a high risk for infections or thrombosis. To overcome these limitations, recent studies have shown that blood pressure can be determined by surrogate measurements, such as the pulse transit time (PTT) [1] or the pulse arrival time (PAT) [2,3]. The PTT is commonly calculated as the time delay of the pulse wave between a proximal and a distal photoplethysmography (PPG) signal, whereas the PAT is calculated between the Rpeak of the ECG and a distal PPG signal. ...
... Several studies have been investigating the blood pressure estimation from surrogate parameters. Many works have been published on ECG and PPG signal processing with deep or recurrent neural networks [2,3]. ...
Commonly used blood pressure measurement devices have noticeable limitations in accuracy, measuring time, comfort or safety. To overcome these limitations, we developed and tested a surrogate-based, non-invasive blood pressure measurement method using an RGB-camera. Our proposed method employs the relation between the pulse transit time (PTT) and blood pressure. Two remote photoplethysmography (rPPG) signals at different distances from the heart are extracted to calculate the temporal delay of the pulse wave. In order to establish the correlation between the PTT values and the blood pressure, a regression model is trained and evaluated. Tests were performed with five subjects, where each subject was recorded fifteen times for 30 seconds. Since the physiological parameters of the cardiac system are different for each person, an individual calibration is required to obtain the systolic and diastolic blood pressure from the PTT values. The calibration results are limited by the small number of samples and the accuracy of the reference system. However, our results show a strong correlation between the PTT values and the blood pressure and we obtained a mean error of 0.18 +/- 5.50 mmHg for the diastolic blood pressure and 0.01 +/- 7.71 mmHg for the systolic pressure, respectively.
... In particular, PPG emerged as a potentially useful signal ( Fig. 2) [14]; indeed, many studies point out a clear relationship between PPG and ABP. Since PPG and ECG can easily be integrated into wearable devices [15−17], they can provide the inputs of deep learning approaches for ABP estimation, as already investigated in previous works [18,19]. Initially, indirect approaches using features derived from PPG and ECG were the most used: He et al. [20] and Shriram et al. [21] showed a strong negative correlation between ABP and pulse transit time (PTT), but pulse wave velocity (PWV) [22] and pulse arrival time (PAT) [23] were also studied. ...
Continuous vital signal monitoring is becoming more relevant in preventing diseases that afflict a large part of the world’s population; for this reason, healthcare equipment should be easy to wear and simple to use. Non-intrusive and non-invasive detection methods are a basic requirement for wearable medical devices, especially when these are used in sports applications or by the elderly for self-monitoring. Arterial blood pressure (ABP) is an essential physiological parameter for health monitoring. Most blood pressure measurement devices determine the systolic and diastolic arterial blood pressure through the inflation and the deflation of a cuff. This technique is uncomfortable for the user and may result in anxiety, and consequently affect the blood pressure and its measurement. The purpose of this paper is the continuous measurement of the ABP through a cuffless, non-intrusive approach. The approach of this paper is based on deep learning techniques where several neural networks are used to infer ABP, starting from photoplethysmogram (PPG) and electrocardiogram (ECG) signals. The ABP was predicted first by utilizing only PPG and then by using both PPG and ECG. Convolutional neural networks (ResNet and WaveNet) and recurrent neural networks (LSTM) were compared and analyzed for the regression task. Results show that the use of the ECG has resulted in improved performance for every proposed configuration. The best performing configuration was obtained with a ResNet followed by three LSTM layers: this led to a mean absolute error (MAE) of 4.118 mmHg on and 2.228 mmHg on systolic and diastolic blood pressures, respectively. The results comply with the American National Standards of the Association for the Advancement of Medical Instrumentation. ECG, PPG, and ABP measurements were extracted from the MIMIC database, which contains clinical signal data reflecting real measurements. The results were validated on a custom dataset created at Neuronica Lab, Politecnico di Torino.
... convolution neural networks. It will be also further analyzed the use of the recurrent neural networks, already used in this experiment [22]. Furthermore, a deeper study of specificity and sensitivity w.r.t. the single patient will be conducted; an unsupervised patient clustering will be exploited for tuning supervised neural networks specific for each cluster, such an approach could be useful to improve the generalization capability of the proposed approach. ...
Cardiovascular Diseases represent the leading cause of deaths in the world. Arterial Blood Pressure (ABP) is an important physiological parameter that should be properly monitored for the purposes of prevention. This work applies the neural network output-error (NNOE) model to ABP forecasting. Three input configurations are proposed based on ECG and PPG for estimating both systolic and diastolic blood pressures. The double channel configuration is the best performing one by means of the mean absolute error w.r.t the corresponding invasive blood pressure signal (IBP); indeed, it is also proven to be compliant with the ANSI/AAMI/ISO 81060-2:2013 regulation for non invasive ABP techniques. Both ECG and PPG correlations to IBP signal are further analyzed using Spearman’s correlation coefficient. Despite it suggests PPG is more closely related to ABP, its regression performance is worse than ECG input configuration one. However, this behavior can be explained looking to human biology and ABP computation, which is based on peaks (systoles) and valleys (diastoles) extraction.
Blood pressure (BP) estimation is one of the most popular and long-standing topics in health-care monitoring area. The utilization of machine learning (ML) and deep learning (DL) for BP prediction has made remarkable progress recently along with the development of ML and especially DL technologies, and the release of large-scale available datasets. In this survey, we present a comprehensive, systematic review about the recent advance of ML and DL for BP prediction. To start with, we systematically sort out the current progress from four perspectives. Then, we summarized commonly-used datasets, evaluation metrics as well as evaluation procedures (especially the usually ignored splitting strategy operation), which is followed by a critical analysis about the reported results. Next, we discussed several practical issues as well as newly-emerging techniques appeared in the research community of BP prediction. Also, we introduced the potential application of several advanced ML technologies in BP estimation. Last, we discussed the question of what a good BP estimator should look like?, and then a general proposal for an objective evaluation of model performance is given from the perspective of an ML researcher. Through this survey, we wish to provide a comprehensive, systematic, up-to-date (to Feb, 2022) review of related research on BP prediction using ML & DL methods, which may be helpful to researchers in this area. We also appeal an objective view of the progress reported in the relevant literatures in a more systematic manner. The experimental data & code and other useful resources are available at https://github.com/v3551G/BP-prediction-survey.
Applications such as surveillance, banking and healthcare deal with sensitive data whose confidentiality and integrity depends on accurate human recognition. In this sense, the crucial mechanism for performing an effective access control is authentication, which unequivocally yields user identity. In 2018, just in North America, around 445K identity thefts have been denounced. The most adopted strategy for automatic identity recognition uses a secret for encrypting and decrypting the authentication information. This approach works very well until the secret is kept safe. Electrocardiograms (ECGs) can be exploited for biometric purposes because both the physiological and geometrical differences in each human heart correspond to uniqueness in the ECG morphology. Compared with classical biometric techniques, e.g. fingerprints, ECG-based methods can definitely be considered a more reliable and safer way for user authentication due to ECG inherent robustness to circumvention, obfuscation and replay attacks. In this paper, the ECG WATCH, a non-expensive wristwatch for recording ECGs anytime, anywhere, in just 10 s, is proposed for user authentication. The ECG WATCH acquisitions have been used to train a shallow neural network, which has reached a 99% classification accuracy and 100% intruder recognition rate.
