Novel interpretable Feature set extraction and classification for accurate atrial fibrillation detection from ECGs.

OBJECTIVE: We present a novel method for detecting atrial fibrillation (AFib) by analyzing Lead II electrocardiograms (ECGs) using a unique set of features. METHODS: For this purpose, we used specific signal processing techniques, such as proper orthogonal decomposition, continuous wavelet transforms, discrete cosine transform, and standard cross-correlation, to extract 48 features from the ECGs. Thus, our approach aims to more effectively capture AFib signatures, such as beat-to-beat variability and fibrillatory waves, than traditional metrics. Moreover, our features were designed to be physiologically interpretable. Subsequently, we incorporated an XGBoost-based ECG classifier to train and evaluate it using the publicly available 'Training' dataset of the 2017 PhysioNet Challenge, which includes 'Normal,' 'AFib,' 'Other,' and 'Noisy' ECG categories. RESULTS: Our method achieved an accuracy of 96 % and an F1-score of 0.83 for 'AFib' detection and 80 % accuracy and 0.85 F1-score for 'Normal' ECGs. Finally, we compared our method's performance with the top-classifiers from the 2017 PhysioNet Challenge, namely ENCASE, Random Forest, and Cascaded Binary. As reported by Clifford et al., 2017, these three best performing models scored 0.80, 0.83, 0.82, in terms of F1-score for 'AFib' detection, respectively. CONCLUSION: Despite using significantly fewer features than the competition's state-of-the-art ECG detection algorithms (48 vs. 150-622), our model achieved a comparable F1-score of 0.83 for successful 'AFib' detection. SIGNIFICANCE: The interpretable features specifically designed for 'AFib' detection results in our method's adaptability in clinical settings for real-time arrhythmia detection and is even effective with short ECGs (<10 heartbeats).

Rose Bengal Labeled Bovine Serum Albumin for Protein Transport Imaging in Subcutaneous Tissues Using Computed Tomography and Fluorescence Microscopy.

Subcutaneous (SC) injection of protein-based therapeutics is a convenient and clinically established drug delivery method. However, progress is needed to increase the bioavailability. Transport of low molecular weight (Mw) biotherapeutics such as insulin and small molecule contrast agents such as lipiodol has been studied using X-ray computed tomography (CT). This analysis, however, does not translate to the investigation of higher Mw therapeutics, such as monoclonal antibodies (mAbs), due to differences in molecular and formulation properties. In this study, an iodinated fluorescein analog rose bengal (RB) was used as a radiopaque and fluorescent label to track the distribution of bovine serum albumin (BSA) compared against unconjugated RB and sodium iodide (NaI) via CT and confocal microscopy following injection into ex vivo porcine SC tissue. Importantly, the high concentration BSA-RB exhibited viscosities more like that of viscous biologics than the small molecule contrast agents, suggesting that the labeled protein may serve as a more suitable formulation for the investigation of injection plumes. Three-dimensional (3D) renderings of the injection plumes showed that the BSA-RB distribution was markedly different from unconjugated RB and NaI, indicating the need for direct visualization of large protein therapeutics using conjugated tags rather than using small molecule tracers. Whereas this proof-of-concept study shows the novel use of RB as a label for tracking BSA distribution, our experimental approach may be applied to high Mw biologics, including mAbs. These studies could provide crucial information about diffusion in SC tissue and the influence of injection parameters on distribution, transport, and downstream bioavailability.

Automated Layer Identification Method for Skin Tissue Histology Images.

We present a novel automated tissue layer identification method for histology images. The method requires a single user input: the number of layers to be identified. The method incorporates a coarse boundary identification step followed by a refinement step. The coarse identification segments the image into 125 × 125 pixel sub-tiles, computes the histogram of each sub-tile, implements K-means clustering to label each sub-tile, and uses Dijkstra's algorithm to form the layer boundary. The refinement step identifies hair follicles, improves the detail and accuracy of the boundary, and segments the epidermis. The method only uses one color channel (blue). We test our proposed method using eight excised porcine tissue samples taken at different anatomical locations. The layer segmentations demonstrated that the dermis thickness increased, and the subcutaneous thickness decreased moving from breast to belly. Minimal variation in the thickness of the epidermis layer across anatomical locations was observed. Overall, these results highlight the importance of quantifying and assessing the tissue environment. Moreover, we demonstrate that our proposed method was robust across different histology stains and did not depend on color-specific information.

Wall Shear Stress Estimation for 4D Flow MRI Using Navier-Stokes Equation Correction.

This study introduces a novel wall shear stress (WSS) estimation method for 4D flow MRI. The method improves the WSS accuracy by using the reconstructed pressure gradient and the flow-physics constraints to correct velocity gradient estimation. The method was tested on synthetic 4D flow data of analytical Womersley flow and flow in cerebral aneurysms and applied to in vivo 4D flow data acquired in cerebral aneurysms and aortas. The proposed method's performance was compared to the state-of-the-art method based on smooth-spline fitting of velocity profile and the WSS calculated from uncorrected velocity gradient. The proposed method improved the WSS accuracy by as much as 100% for the Womersley flow and reduced the underestimation of mean WSS by 39 to 50% for the synthetic aneurysmal flow. The predicted mean WSS from the in vivo aneurysmal data using the proposed method was 31 to 50% higher than the other methods. The predicted aortic WSS using the proposed method was 3 to 6 times higher than the other methods and was consistent with previous CFD studies and the results from recently developed methods that take into account the limited spatial resolution of 4D flow MRI. The proposed method improves the accuracy of WSS estimation from 4D flow MRI, which can help predict blood vessel remodeling and progression of cardiovascular diseases.

A method for direct estimation of left ventricular global longitudinal strain rate from echocardiograms.

We present a new method for measuring global longitudinal strain and global longitudinal strain rate from 2D echocardiograms using a logarithmic-transform correlation (LTC) method. Traditional echocardiography strain analysis depends on user inputs and chamber segmentation, which yield increased measurement variability. In contrast, our approach is automated and does not require cardiac chamber segmentation and regularization, thus eliminating these issues. The algorithm was benchmarked against two conventional strain analysis methods using synthetic left ventricle ultrasound images. Measurement error was assessed as a function of contrast-to-noise ratio (CNR) using mean absolute error and root-mean-square error. LTC showed better agreement to the ground truth strain [Formula: see text] and ground truth strain rate [Formula: see text] compared with agreement to ground truth for two block-matching speckle tracking algorithms (one based on sum of square difference and the other on Fourier transform correlation; strain [Formula: see text], strain rate [Formula: see text]). A 200% increase in strain measurement accuracy was observed compared to the conventional algorithms. Subsequently, we tested the method using a 53-subject clinical cohort (20 subjects diseased with cardiomyopathy, 33 healthy controls). Our method distinguished between normal and abnormal left ventricular function with an AUC = 0.89, a 5% improvement over the conventional GLS algorithms.

Modeling Bias Error in 4D Flow MRI Velocity Measurements.

We present a model to estimate the bias error of 4D flow magnetic resonance imaging (MRI) velocity measurements. The local instantaneous bias error is defined as the difference between the expectation of the voxel's measured velocity and actual velocity at the voxel center. The model accounts for bias error introduced by the intra-voxel velocity distribution and partial volume (PV) effects. We assess the intra-voxel velocity distribution using a 3D Taylor Series expansion. PV effects and numerical errors are considered using a Richardson extrapolation. The model is applied to synthetic Womersley flow and in vitro and in vivo 4D flow MRI measurements in a cerebral aneurysm. The bias error model is valid for measurements with at least 3.75 voxels across the vessel diameter and signal-to-noise ratio greater than 5. All test cases exceeded this diameter to voxel size ratio with diameters, isotropic voxel sizes, and velocity ranging from 3-15mm, 0.5-1mm, and 0-60cm/s, respectively. The model accurately estimates the bias error in voxels not affected by PV effects. In PV voxels, the bias error is an order of magnitude higher, and the accuracy of the bias error estimation in PV voxels ranges from 67.3% to 108% relative to the actual bias error. The bias error estimated for in vivo measurements increased two-fold at systole compared to diastole in partial volume and non-partial volume voxels, suggesting the bias error varies over the cardiac cycle. This bias error model quantifies 4D flow MRI measurement accuracy and can help plan 4D flow MRI scans.

A multi-modality approach for enhancing 4D flow magnetic resonance imaging via sparse representation.

This work evaluates and applies a multi-modality approach to enhance blood flow measurements and haemodynamic analysis with phase-contrast magnetic resonance imaging (4D flow MRI) in cerebral aneurysms (CAs). Using a library of high-resolution velocity fields from patient-specific computational fluid dynamic simulations and in vitro particle tracking velocimetry measurements, the flow field of 4D flow MRI data is reconstructed as the sparse representation of the library. The method was evaluated with synthetic 4D flow MRI data in two CAs. The reconstruction enhanced the spatial resolution and velocity accuracy of the synthetic MRI data, leading to reliable pressure and wall shear stress (WSS) evaluation. The method was applied on in vivo 4D flow MRI data acquired in the same CAs. The reconstruction increased the velocity and WSS by 6-13% and 39-61%, respectively, suggesting that the accuracy of these quantities was improved since the raw MRI data underestimated the velocity and WSS by 10-20% and 40-50%, respectively. The computed pressure fields from the reconstructed data were consistent with the observed flow structures. The results suggest that using the sparse representation flow reconstruction with in vivo 4D flow MRI enhances blood flow measurement and haemodynamic analysis.

Automated Peak Prominence-Based Iterative Dijkstra's Algorithm for Segmentation of B-Mode Echocardiograms.

We present a user-initialized, automated left ventricle (LV) segmentation method for use with echocardiograms (echo). The method uses an iterative Dijkstra's algorithm, strategic node selection, and novel cost matrix formulation based on intensity peak prominence and is termed the "Prominence Iterative Dijkstra's" algorithm, or ProID. ProID is initialized with three user-input clicks per time-series scan. ProID was tested using artificial echos representing five different systems. Results showed accurate LV contours and volume estimations as compared to the ground-truth for all systems. Using the CAMUS dataset, we demonstrate ProID maintained similar Dice similarity scores (DSS) to other automated methods. ProID was then used to analyze a clinical cohort of 66 pediatric patients, including normal and diseased hearts. Output segmentations, LV volume, and ejection fraction were compared against manual segmentations from two expert readers. ProID maintained an average DSS of 0.93 when comparing against manual segmentation. Comparing the two expert readers, the manual segmentations maintained a DSS of 0.93 which increased to 0.95 when they used ProID. Thus, ProID reduced inter-operator variability across the expert readers. Overall, this work demonstrates ProID yields accurate boundaries across age groups, disease states, and echo platforms with low computational cost and no need for training data.

Divergence-Free Constrained Phase Unwrapping and Denoising for 4D Flow MRI Using Weighted Least-Squares.

A novel divergence-free constrained phase unwrapping method was proposed and evaluated for 4D flow MRI. The unwrapped phase field was obtained by integrating the phase variations estimated from the wrapped phase data using weighted least-squares. The divergence-free constraint for incompressible blood flow was incorporated to regulate and denoise the resulting phase field. The proposed method was tested on synthetic phase data of left ventricular flow and in vitro 4D flow measurement of Poiseuille flow. The method was additionally applied to in vivo 4D flow measurements in the thoracic aorta from 30 human subjects. The performance of the proposed method was compared to the state-of-the-art 4D single-step Laplacian algorithm. The synthetic phase data were completely unwrapped by the proposed method for all the cases with velocity encoding (venc) as low as 20% of the maximum velocity and signal-to-noise ratio as low as 5. The in vitro Poiseuille flow data were completely unwrapped with a 60% increase in the velocity-to-noise ratio. For the in-vivo aortic datasets with venc ratio less than 0.4, the proposed method significantly improved the success rate by as much as 40% and reduced the velocity error levels by a factor of 10 compared to the state-of-the-art method. The divergence-free constrained method exhibits reliability and robustness on phase unwrapping and shows improved accuracy of velocity and hemodynamic quantities by unwrapping the low-venc 4D flow MRI data.

Universality of vortex ring decay in the left ventricle.

We present clinical measurements and a theoretical model for the decay of the left ventricular (LV) vortex ring. Previous works have postulated that the formation of the vortex ring downstream of the mitral annulus is affected by LV diastolic impairment. However, no previous works have considered how the strength of the vortex ring will decay inside the ventricle after its formation. Although the vortex ring formation relates to the very initial stage of the filling, the decay process is governed by a large portion of the diastolic time and will be affected by the interaction of the ventricle walls and the vortex ring. Here we used in-vivo measurements and presented a mechanistic model to calculate the evolution of the vortex ring strength and predict the rate of vortex ring decay within the left ventricle. The results demonstrated the actual circulation decay rate was universal, remaining nearly unchanged across all subjects of varying LV geometry or diastolic function. Furthermore, using the model-predicted circulation decay rate, differentiation between normal and abnormal filling was observed.

4D Flow MRI Pressure Estimation Using Velocity Measurement-Error-Based Weighted Least-Squares.

This work introduces a 4D flow magnetic resonance imaging (MRI) pressure reconstruction method which employs weighted least-squares (WLS) for pressure integration. Pressure gradients are calculated from the velocity fields, and velocity errors are estimated from the velocity divergence for incompressible flow. Pressure gradient errors are estimated by propagating the velocity errors through Navier-Stokes momentum equation. A weight matrix is generated based on the pressure gradient errors, then employed for pressure reconstruction. The pressure reconstruction method was demonstrated and analyzed using synthetic velocity fields as well as Poiseuille flow measured using in vitro 4D flow MRI. Performance of the proposed WLS method was compared to the method of solving the pressure Poisson equation which has been the primary method used in the previous studies. Error analysis indicated that the proposed method is more robust to velocity measurement errors. Improvement on pressure results was found to be more significant for the cases with spatially-varying velocity error level, with reductions in error ranging from 50% to over 200%. Finally, the method was applied to flow in patient-specific cerebral aneurysms. Validation was performed with in vitro flow data collected using Particle Tracking Velocimetry (PTV) and in vivo flow measurement obtained using 4D flow MRI. Pressure calculated by WLS, as opposed to the Poisson equation, was more consistent with the flow structures and showed better agreement between the in vivo and in vitro data. These results suggest the utility of WLS method to obtain reliable pressure field from clinical flow measurement data.

Multi-modality cerebral aneurysm haemodynamic analysis: in vivo 4D flow MRI, in vitro volumetric particle velocimetry and in silico computational fluid dynamics.

Typical approaches to patient-specific haemodynamic studies of cerebral aneurysms use image-based computational fluid dynamics (CFD) and seek to statistically correlate parameters such as wall shear stress (WSS) and oscillatory shear index (OSI) to risk of growth and rupture. However, such studies have reported contradictory results, emphasizing the need for in-depth multi-modality haemodynamic metric evaluation. In this work, we used in vivo 4D flow MRI data to inform in vitro particle velocimetry and CFD modalities in two patient-specific cerebral aneurysm models (basilar tip and internal carotid artery). Pulsatile volumetric particle velocimetry experiments were conducted, and the particle images were processed using Shake-the-Box, a particle tracking method. Distributions of normalized WSS and relative residence time were shown to be highly yet inconsistently affected by minor flow field and spatial resolution variations across modalities, and specific relationships among these should be explored in future work. Conversely, OSI, a non-dimensional parameter, was shown to be more robust to the varying assumptions, limitations and spatial resolutions of each subject and modality. These results suggest a need for further multi-modality analysis as well as development of non-dimensional haemodynamic parameters and correlation of such metrics to aneurysm risk of growth and rupture.

Computational Fluid Dynamics of Vascular Disease in Animal Models.

Recent applications of computational fluid dynamics (CFD) applied to the cardiovascular system have demonstrated its power in investigating the impact of hemodynamics on disease initiation, progression, and treatment outcomes. Flow metrics such as pressure distributions, wall shear stresses (WSS), and blood velocity profiles can be quantified to provide insight into observed pathologies, assist with surgical planning, or even predict disease progression. While numerous studies have performed simulations on clinical human patient data, it often lacks prediagnosis information and can be subject to large intersubject variability, limiting the generalizability of findings. Thus, animal models are often used to identify and manipulate specific factors contributing to vascular disease because they provide a more controlled environment. In this review, we explore the use of CFD in animal models in recent studies to investigate the initiating mechanisms, progression, and intervention effects of various vascular diseases. The first section provides a brief overview of the CFD theory and tools that are commonly used to study blood flow. The following sections are separated by anatomical region, with the abdominal, thoracic, and cerebral areas specifically highlighted. We discuss the associated benefits and obstacles to performing CFD modeling in each location. Finally, we highlight animal CFD studies focusing on common surgical treatments, including arteriovenous fistulas (AVF) and pulmonary artery grafts. The studies included in this review demonstrate the value of combining CFD with animal imaging and should encourage further research to optimize and expand upon these techniques for the study of vascular disease.

Density and Viscosity Matched Newtonian and non-Newtonian Blood-Analog Solutions with PDMS Refractive Index.

Optical imaging is commonly used to investigate biological flows and cardiovascular disease using compliant silicone polydimethysiloxane (PDMS) Sylgard 184 geometries. However, selecting the working fluid with blood density and viscosity, and PDMS index of refraction (RI) for such experiments is challenging. Currently, water-glycerol is commonly used and sodium iodide (NaI) often added to increase the index of refraction without changing fluid viscosity. But the resulting fluid density is well above blood. Moreover, NaI is expensive, has safety and material discoloration concerns, and has been reported to affect non-Newtonian fluid behavior. Here, we present a new blood analog alternative based on urea. Urea is approximately five to fifteen times less expensive than NaI, safe and easy to handle, optically clear, and causes no discoloration. Water-glycerol-urea solutions, unlike those with NaI, simultaneously matched the density and viscosity of blood and RI of PDMS. Water-xylitol and water-xylitol-urea solutions are also possible blood analog solutions. Xanthan gum (XG)-water-glycerol non-Newtonian solutions maintained similar viscoelastic properties throughout the range of weight percent (about 15-25%) of urea and NaI used here. The results showed that the XG weight percent affected viscoelastic properties more than the weight percent of urea or NaI tested in this study. Overall, we demonstrate urea is useful for PDMS blood analog experiments and should also be considered as an inexpensive additive, and an alternative to NaI.

Hemodynamics of Stent Implantation Procedures in Coronary Bifurcations: An In Vitro Study.

Stent implantation in coronary bifurcations presents unique challenges and currently there is no universally accepted stent deployment approach. Despite clinical and computational studies, the effect of each stent implantation method on the coronary artery hemodynamics is not well understood. In this study the hemodynamics of stented coronary bifurcations under pulsatile flow conditions were investigated experimentally. Three implantation methods, provisional side branch (PSB), culotte (CUL), and crush (CRU), were investigated using time-resolved particle image velocimetry to measure the velocity fields. Subsequently, hemodynamic parameters including wall shear stress, oscillatory shear index (OSI), and relative residence time (RRT) were calculated. The pressure field through the vessel was non-invasively quantified and pressure wave speeds were computed. The effects of each stented case were evaluated and compared against an un-stented case. CRU provided the lowest compliance mismatch, but demonstrated detrimental stent interactions. PSB, the clinically preferred method, and CUL maintained many normal flow conditions. However, PSB provided about a 300% increase in both OSI and RRT. CUL yielded a 10 and 85% increase in OSI and RRT, respectively. The results of this study support the concept that different bifurcation stenting techniques result in hemodynamic environments that deviate from that of un-stented bifurcations, to varying degrees.