A Comparison of Deep Learning Techniques for Arterial Blood Pressure Prediction

Abstract

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.

State-of-the-Art Deep Learning Methods on Electrocardiogram Data: Systematic Review

“…Only 2.6% (6/230) of studies that applied DL methods to ECG data to perform BP estimation were identified in the literature search. A combined architecture of ResNets and LSTM was proposed twice (33.3%), once by Miao et al [ 94 ], who achieved a mean error of −0.22 (SD 5.82) mm Hg for systolic BP (SBP) prediction and of −0.75 (SD 5.62) mm Hg for diastolic BP (DBP) prediction using data that originated from a private database, and once by Paviglianiti et al [ 96 ], who achieved a mean average error of 4.118 mm Hg for SBP and 2.228 mm Hg for DBP prediction using the Medical Information Mart for Intensive Care database. By contrast, Jeong and Lim [ 98 ] exercised a CNN-LSTM network on the Medical Information Mart for Intensive Care database and managed to predict SBP and DBP with a mean error of 0.0 (SD 1.6) mm Hg and 0.2 (SD 1.3) mm Hg, respectively.…”

“…A common pitfall in the use of calibrated techniques is that subsequent test data points do not differ significantly from the calibration value and thus yield small errors in prediction, whereas the data are presented as an aggregate pooled correlation plot or Bland-Altman plot with a correlation value that simply reflects the range of BPs across the population rather than patient-specific BP variation [326,327]. In our review of articles using DL for BP prediction, we did not encounter significant attempts to address the issue of BP variability in training data; in fact, many publications explicitly removed data points with hypertensive values or large pulse pressures from their data sets as "artifacts" [93][94][95][96]98].…”

Advances in Cuffless Continuous Blood Pressure Monitoring Technology Based on PPG Signals

“…After transfer learning-based calibration, the prediction performance was significantly improved. Based on PPG and ECG signals of 40 subjects selected from the MIMIC dataset, Paviglianiti et al [ 81 ] compared the performance of three deep learning models, ResNet, LSTM, and WaveNet. The results pointed out that ResNet combined with three LSTM layers achieved the best prediction performance, with an MAE of 4.118 and 2.228 mmHg for SBP and DBP, respectively.…”

Blood pressure estimation from photoplethysmography by considering intra- and inter-subject variabilities: guidelines for a fair assessment

“…Signalbased techniques, in turn, have been explored deep learning models in order to, jointly, extract features and perform estimation. Several methods, including combinations of them, have also been used: fully connected neural network [49], [50], convolutional neural network [32], [37], [48], [50], long short-term memory [32], [37], [42], [48]- [50], gated recurrent unit [50], AlexNet [42], [44], ResNet [42], [46], [49], WaveNet [49], U-Net [35], residual U-Net [39] and generative adversarial network [48].…”