A novel method of artery stenosis diagnosis using transfer function and support vector machine based on transmission line model: A numerical simulation and validation study

Peripheral artery disease diagnosis based on deep learning-enabled analysis of non-invasive arterial pulse waveforms

This paper intends to investigate the feasibility of peripheral artery disease (PAD) diagnosis based on the analysis of non-invasive arterial pulse waveforms. We generated realistic synthetic arterial blood pressure (BP) and pulse volume recording (PVR) waveform signals pertaining to PAD present at the abdominal aorta with a wide range of severity levels using a mathematical model that simulates arterial blood circulation and arterial BP-PVR relationships. We developed a deep learning (DL)-enabled algorithm that can diagnose PAD by analyzing brachial and tibial PVR waveforms, and evaluated its efficacy in comparison with the same DL-enabled algorithm based on brachial and tibial arterial BP waveforms as well as the ankle-brachial index (ABI). The results suggested that it is possible to detect PAD based on DL-enabled PVR waveform analysis with adequate accuracy, and its detection efficacy is close to when arterial BP is used (positive and negative predictive values at 40 % abdominal aorta occlusion: 0.78 vs 0.89 and 0.85 vs 0.94; area under the ROC curve (AUC): 0.90 vs 0.97). On the other hand, its efficacy in estimating PAD severity level is not as good as when arterial BP is used (r value: 0.77 vs 0.93; Bland-Altman limits of agreement: -32%-+32 % vs -20%-+19 %). In addition, DL-enabled PVR waveform analysis significantly outperformed ABI in both detection and severity estimation. In sum, the findings from this paper suggest the potential of DL-enabled non-invasive arterial pulse waveform analysis as an affordable and non-invasive means for PAD diagnosis.

Clinically-Driven Virtual Patient Cohorts Generation: An Application to Aorta

The combination of machine learning methods together with computational modeling and simulation of the cardiovascular system brings the possibility of obtaining very valuable information about new therapies or clinical devices through in-silico experiments. However, the application of machine learning methods demands access to large cohorts of patients. As an alternative to medical data acquisition and processing, which often requires some degree of manual intervention, the generation of virtual cohorts made of synthetic patients can be automated. However, the generation of a synthetic sample can still be computationally demanding to guarantee that it is clinically meaningful and that it reflects enough inter-patient variability. This paper addresses the problem of generating virtual patient cohorts of thoracic aorta geometries that can be used for in-silico trials. In particular, we focus on the problem of generating a cohort of patients that meet a particular clinical criterion, regardless the access to a reference sample of that phenotype. We formalize the problem of clinically-driven sampling and assess several sampling strategies with two goals, sampling efficiency, i.e., that the generated individuals actually belong to the target population, and that the statistical properties of the cohort can be controlled. Our results show that generative adversarial networks can produce reliable, clinically-driven cohorts of thoracic aortas with good efficiency. Moreover, non-linear predictors can serve as an efficient alternative to the sometimes expensive evaluation of anatomical or functional parameters of the organ of interest.

Intelligent Oscillometric System for Automatic Detection of Peripheral Arterial Disease

Peripheral arterial disease (PAD) is a progressing arterial disorder that is associated with significant morbidity and mortality. The conventional PAD detection methods are invasive, cumbersome, or require expensive equipment and highly trained technicians. Here, we propose a new automated, noninvasive, and easy-to-use method for the detection of PAD based on characterizing the arterial system by applying an external varying pressure using a cuff. The superposition of the internal arterial pressure and the externally applied pressure were measured and mathematically modeled as a function of cuff pressure. A feature-based learning algorithm was then designed to identify PAD patterns by analyzing the parameters of the derived mathematical models. Genetic algorithm and principal component analysis were employed to select the best predictive features distinguishing PAD patterns from normal. A RUSBoost ensemble model using neural network as the base learner was designed to diagnose PAD from genetic algorithm selected features. The proposed method was validated on data collected from 14 PAD patients and 19 healthy individuals. It achieved a high accuracy, sensitivity, and specificity of 91.4%, 90.0%, and 92.1%, respectively, in detecting PAD. The effect of age, a confounding factor that may have impacted our analyzes, was not considered in this study. The proposed method shows promise toward noninvasive and accurate detection of PAD and can be integrated into routine oscillometric blood pressure measurements.

Deep Learning-Based Diagnosis of Peripheral Artery Disease via Continuous Property-Adversarial Regularization: Preliminary in Silico Study

This paper presents a novel deep learning-based arterial pulse wave analysis (PWA) approach to diagnosis of peripheral artery occlusive disease (PAD). Naïve application of deep learning to PAD diagnosis can be hampered by the fact that securing a large amount of longitudinal dataset encompassing diverse PAD severity as well as anatomical and physiological variability presents formidable challenge. Training of a deep neural network (DNN) to a small training dataset raises the risk of overfitting the PAD diagnosis algorithm only to the individuals in the training dataset while deteriorating its ability to generalize also to other individuals who may exhibit a large variability in anatomical and physiological characteristics beyond the training dataset. To overcome these obstacles, we propose a continuous property-adversarial regularization (CPAR) approach to robust generalization of a DNN against scarce datasets. Our approach fosters the exploitation of latent features that can facilitate the intended task independently of confounding property-induced disturbances. by regularizing the extraction of disturbance-dependent latent features in the network's feature extraction layer. By training and testing a deep convolutional neural network (CNN) for PAD diagnosis using scarce virtual datasets, we illustrated that the CNN trained by our approach was superior to a conventionally trained CNN in detecting and assessing the severity of PAD against disturbances originating from diversity in the patients' height and arterial stiffness: when trained with one-time pulse wave signal measurement at ankle and brachial arteries in a small number of patients, our approach achieved detection accuracy of >90% and severity assessment of 0.83 in r² value, which were >15% and >40% improvement over conventional approach without CPAR. In addition, we ascertained the advantage of our approach in efficient training and robust generalization of DNN by contrasting it to multi-task learning which promotes the exploitation (as opposed to regularization in CPAR) of disturbance-dependent latent features in fulfilling the intended tasks.

Detection and Severity Assessment of Peripheral Occlusive Artery Disease via Deep Learning Analysis of Arterial Pulse Waveforms: Proof-of-Concept and Potential Challenges

Toward the ultimate goal of affordable and non-invasive screening of peripheral occlusive artery disease (PAD), the objective of this work is to investigate the potential of deep learning-based arterial pulse waveform analysis in detecting and assessing the severity of PAD. Using an established transmission line model of arterial hemodynamics, a large number of virtual patients associated with PAD of a wide range of severity and the corresponding arterial pulse waveform data were created. A deep convolutional neural network capable of detecting and assessing the severity of PAD based on the analysis of brachial and ankle arterial pulse waveforms was constructed, evaluated for efficacy, and compared with the state-of-the-art ankle-brachial index (ABI) using the virtual patients. The results suggested that deep learning may diagnose PAD more accurately and robustly than ABI. In sum, this work demonstrates the initial proof-of-concept of deep learning-based arterial pulse waveform analysis for affordable and convenient PAD screening as well as presents challenges that must be addressed for real-world clinical applications.

Autodetect extracranial and intracranial artery stenosis by machine learning using ultrasound

BACKGROUND

and Purpose: This study proposed a machine learning method for identifying ≥50% stenosis of the extracranial and intracranial arteries.

PATIENTS AND METHODS

A total of 8211 patients with both carotid ultrasound and cerebral angiography were enrolled. Support vector machine (SVM) was employed as the machine learning classifier. Carotid Doppler parameters and transcranial Doppler parameters were used as the input features. Feature selection was performed using the Extra-Trees (extremely randomized trees) method.

RESULTS

For the machine learning method, the sensitivities and specificities of identifying stenosis of the extracranial arteries were 88.5%-100% and 96.0%-100%, respectively. The sensitivities and specificities of identifying stenosis of the intracranial arteries were 71.7%-100% and 88.9%-100%, respectively.

CONCLUSIONS

The SVM classifier with feature selection is an efficient method for identifying the stenosis of both intracranial and extracranial arteries. Comparing with traditional Doppler criteria, this machine learning method achieves up to 20% higher in accuracy and 45% in sensitivity, respectively.

N-Point Moving Average: A Special Generalized Transfer Function Method for Estimation of Central Aortic Blood Pressure

OBJECTIVE

N-point moving average (NPMA) is a simplified method of central aortic systolic pressure (CASP) estimation in comparison with the generalized transfer function (GTF). The fundamental difference or similarity between the methods is not established. This study investigates theoretical properties of NPMA relative to GTF and explores the integer and fractional denominator for the averaging process in the NPMA.

METHODS

Convolution of a specified square wave and the radial (or brachial) blood pressure waveform constituted the NPMA . A single uniform tube model-based TF (MTF) was employed to investigate potential physiological meaning of NPMA. In experimental analysis, invasive, simultaneously recorded aortic and radial pressure waveforms were obtained in 62 subjects under control conditions and following nitroglycerin administration. CASP was estimated by NPMA (), GTF ( ), and MTF (CASP _MTF) from radial waveforms by tenfold cross validation.

RESULTS

Theoretical analysis showed that NPMA was an inversed constant TF. Its spectrum matched that of MTF in low frequency (<4 Hz for radial and <5 Hz for brachial) by optimizing reflection coefficient and propagation time. Experiment results showed the NPMA optimized fractional denominator of K = 4.4 significantly decreased the mean difference between CASP_NPMA and measured CASP to 0.0 ± 4.7 mmHg from -1.8 ± 4.6 mmHg for integer denominator of K = 4. CASP_NPMA correlated with CASP_MTF and CASP _GTF (r² = 0.99 and 0.97, mean difference: -0.3 ± 1.8 and 0.5 ± 2.7 mmHg).

CONCLUSION

This study demonstrated that NPMA is similar in nature to the GTF.

PWPSim: A new simulation tool of pulse wave propagation in the human arterial tree

Hemodynamic simulation enhances investigations of pulse wave propagation phenomena and enables validation of new methods of pulse wave analysis. However, such simulation systems or tools are not readily available nor easily accessible. In this study, a new simulation tool of pulse wave propagation in the human arterial tree was developed based on a transmission line model (TLM). This paper describes the theory of TLM of the human arterial tree used by this simulation. The results are a display of the main functions and simulation results of this tool. This tool allows simulation of pulse wave propagation with capability to change the range of parameters, such as heart rate, mean flow and left ventricular ejection time, body height, arterial radius and wall thickness, arterial viscoelasticity, peripheral resistance and compliance. It also accounts for the nonlinear elasticity of arteries. The simulation results are displayed as 2D and 3D figures of blood pressure and flow waveforms, input impedance and pressure transfer function between aorta and femoral artery, including systolic blood pressure, diastolic blood pressure, pulse pressure, carotid-femoral pulse wave velocity, brachial-ankle pulse wave velocity and ankle-brachial index. It is a useful and interactive simulation tool of pulse wave propagation in the systematic arterial tree.

Effects of cardiac timing and peripheral resistance on measurement of pulse wave velocity for assessment of arterial stiffness

To investigate the effects of heart rate (HR), left ventricular ejection time (LVET) and wave reflection on arterial stiffness as assessed by pulse wave velocity (PWV), a pulse wave propagation simulation system (PWPSim) based on the transmission line model of the arterial tree was developed and was applied to investigate pulse wave propagation. HR, LVET, arterial elastic modulus and peripheral resistance were increased from 60 to 100 beats per minute (bpm), 0.1 to 0.45 seconds, 0.5 to 1.5 times and 0.5 to 1.5 times of the normal value, respectively. Carotid-femoral PWV (cfPWV) and brachial-ankle PWV (baPWV) were calculated by intersecting tangent method (cfPWV_tan and baPWV_tan), maximum slope (cfPWV_max and baPWV_max), and using the Moens-Korteweg equation ([Formula: see text] and [Formula: see text]). Results showed cfPWV and baPWV increased significantly with arterial elastic modulus but did not increase with HR when using a constant elastic modulus. However there were significant LVET dependencies of cfPWV_tan and baPWV_tan (0.17 ± 0.13 and 0.17 ± 0.08 m/s per 50 ms), and low peripheral resistance dependencies of cfPWV_tan, cfPWV_max, baPWV_tan and baPWV_max (0.04 ± 0.01, 0.06 ± 0.04, 0.06 ± 0.03 and 0.09 ± 0.07 m/s per 10% peripheral resistance), respectively. This study demonstrated that LVET dominates the effect on calculated PWV compared to HR and peripheral resistance when arterial elastic modulus is constant.

Arterial viscoelasticity: role in the dependency of pulse wave velocity on heart rate in conduit arteries

Experimental investigations have established that the stiffness of large arteries has a dependency on acute heart rate (HR) changes. However, the possible underlying mechanisms inherent in this HR dependency have not been well established. This study aimed to explore a plausible viscoelastic mechanism by which HR exerts an influence on arterial stiffness. A multisegment transmission line model of the human arterial tree incorporating fractional viscoelastic components in each segment was used to investigate the effect of varying fractional order parameter (α) of viscoelasticity on the dependence of aortic arch to femoral artery pulse wave velocity (afPWV) on HR. HR was varied from 60 to 100 beats/min at a fixed mean flow of 100 ml/s. PWV was calculated by intersecting tangent method (afPWVTan) and by phase velocity from the transfer function (afPWVTF) in the time and frequency domain, respectively. PWV was significantly and positively associated with HR for α ≥ 0.6; for α = 0.6, 0.8, and 1, HR-dependent changes in afPWVTan were 0.01 ± 0.02, 0.07 ± 0.04, and 0.22 ± 0.09 m/s per 5 beats/min; HR-dependent changes in afPWVTF were 0.02 ± 0.01, 0.12 ± 0.00, and 0.34 ± 0.01 m/s per 5 beats/min, respectively. This crosses the range of previous physiological studies where the dependence of PWV on HR was found to be between 0.08 and 0.10 m/s per 5 beats/min. Therefore, viscoelasticity of the arterial wall could contribute to mechanisms through which large artery stiffness changes with changing HR. Physiological studies are required to confirm this mechanism.

NEW & NOTEWORTHY This study used a transmission line model to elucidate the role of arterial viscoelasticity in the dependency of pulse wave velocity on heart rate. The model uses fractional viscoelasticity concepts, which provided novel insights into arterial hemodynamics. This study also provides a means of assessing the clinical manifestation of the association of pulse wave velocity and heart rate.

A new formula for predicting the position of severe arterial stenosis

Noninvasive location of an occlusion or a severe stenosis in the arterial system is of a great interest for surgical interventions. Here, we present a new method to determine the location of arterial 99% stenosis in the arterial (sub) system. The method requires a measurement of propagation constant and the instantaneous flow rate or velocity at two sites of an arterial tree. The method was successfully tested using Womersley's oscillatory flow theory and the data obtained by a simulation of Fluid structure interaction (FSI). The effect of noise has been investigated to simulate experimental conditions. The results demonstrate that location of 99% severe stenosis could be accurately obtained. The spatial resolution was approximately a few centimeters and the differences between exact and computed values didn't exceed 13%. However, the identifications of stenotic sites decreased with the distance. Further investigation of the developed method in vivo and in vitro is required.