Optimization of Doppler echocardiographic velocity measurements using an automatic contour detection method

Automated mitral inflow Doppler peak velocity measurement using deep learning

Doppler echocardiography is a widely utilised non-invasive imaging modality for assessing the functionality of heart valves, including the mitral valve. Manual assessments of Doppler traces by clinicians introduce variability, prompting the need for automated solutions. This study introduces an innovative deep learning model for automated detection of peak velocity measurements from mitral inflow Doppler images, independent from Electrocardiogram information. A dataset of Doppler images annotated by multiple expert cardiologists was established, serving as a robust benchmark. The model leverages heatmap regression networks, achieving 96% detection accuracy. The model discrepancy with the expert consensus falls comfortably within the range of inter- and intra-observer variability in measuring Doppler peak velocities. The dataset and models are open-source, fostering further research and clinical application.

Intra-and inter-observer variability of point of care ultrasound measurements to evaluate hemodynamic parameters in healthy volunteers

BACKGROUND

Point-of-care ultrasound (POCUS) is a valuable tool for assessing the hemodynamic status of acute patients. Even though POCUS often uses a qualitative approach, quantitative measurements have potential advantages in evaluating hemodynamic status. Several quantitative ultrasound parameters can be used to assess the hemodynamic status and cardiac function. However, only limited data on the feasibility and reliability of the quantitative hemodynamic measurements in the point-of-care setting are available. This study investigated the intra- and inter-observer variability of PoCUS measurements of quantitative hemodynamic parameters in healthy volunteers.

METHODS

In this prospective observational study, three sonographers performed three repeated measurements of eight different hemodynamic parameters in healthy subjects. An expert panel of two experienced sonographers evaluated the images' quality. The repeatability (intra-observer variability) was determined by calculating the coefficient of variation (CV) between the separate measurements for each observer. The reproducibility (inter-observer variability) was assessed by determining the intra-class correlation coefficient (ICC).

RESULTS

32 subjects were included in this study, on whom, in total, 1502 images were obtained for analysis. All parameters were in a normal physiological range. Stroke volume (SV), cardiac output (CO), and inferior vena cava diameter (IVC-D) showed high repeatability (CV under 10%) and substantial reproducibility (ICC 0.61-0.80). The other parameters had only moderate repeatability and reproducibility.

CONCLUSIONS

We demonstrated good inter-observer reproducibility and good intra-observer repeatability for CO, SV and IVC-D taken in healthy subjects by emergency care physicians.

Automated multi-beat tissue Doppler echocardiography analysis using deep neural networks

Tissue Doppler imaging is an essential echocardiographic technique for the non-invasive assessment of myocardial blood velocity. Image acquisition and interpretation are performed by trained operators who visually localise landmarks representing Doppler peak velocities. Current clinical guidelines recommend averaging measurements over several heartbeats. However, this manual process is both time-consuming and disruptive to workflow. An automated system for accurate beat isolation and landmark identification would be highly desirable. A dataset of tissue Doppler images was annotated by three cardiologist experts, providing a gold standard and allowing for observer variability comparisons. Deep neural networks were trained for fully automated predictions on multiple heartbeats and tested on tissue Doppler strips of arbitrary length. Automated measurements of peak Doppler velocities show good Bland-Altman agreement (average standard deviation of 0.40 cm/s) with consensus expert values; less than the inter-observer variability (0.65 cm/s). Performance is akin to individual experts (standard deviation of 0.40 to 0.75 cm/s). Our approach allows for > 26 times as many heartbeats to be analysed, compared to a manual approach. The proposed automated models can accurately and reliably make measurements on tissue Doppler images spanning several heartbeats, with performance indistinguishable from that of human experts, but with significantly shorter processing time. HIGHLIGHTS: • Novel approach successfully identifies heartbeats from Tissue Doppler Images • Accurately measures peak velocities on several heartbeats • Framework is fast and can make predictions on arbitrary length images • Patient dataset and models made public for future benchmark studies.

Application of AI in cardiovascular multimodality imaging

Technical advances in artificial intelligence (AI) in cardiac imaging are rapidly improving the reproducibility of this approach and the possibility to reduce time necessary to generate a report. In cardiac computed tomography angiography (CCTA) the main application of AI in clinical practice is focused on detection of stenosis, characterization of coronary plaques, and detection of myocardial ischemia. In cardiac magnetic resonance (CMR) the application of AI is focused on post-processing and particularly on the segmentation of cardiac chambers during late gadolinium enhancement. In echocardiography, the application of AI is focused on segmentation of cardiac chambers and is helpful for valvular function and wall motion abnormalities. The common thread represented by all of these techniques aims to shorten the time of interpretation without loss of information compared to the standard approach. In this review we provide an overview of AI applications in multimodality cardiac imaging.

Automated Analysis of Doppler Echocardiographic Videos as a Screening Tool for Valvular Heart Diseases

OBJECTIVES

This study sought to develop a deep learning (DL) framework to automatically analyze echocardiographic videos for the presence of valvular heart diseases (VHDs).

BACKGROUND

Although advances in DL have been applied to the interpretation of echocardiograms, such techniques have not been reported for interpretation of color Doppler videos for diagnosing VHDs.

METHODS

We developed a 3-stage DL framework for automatic screening of echocardiographic videos for mitral stenosis (MS), mitral regurgitation (MR), aortic stenosis (AS), and aortic regurgitation (AR) that classifies echocardiographic views, detects the presence of VHDs, and, when present, quantifies key metrics related to VHD severities. The algorithm was trained (n = 1,335), validated (n = 311), and tested (n = 434) using retrospectively selected studies from 5 hospitals. A prospectively collected set of 1,374 consecutive echocardiograms served as a real-world test data set.

RESULTS

Disease classification accuracy was high, with areas under the curve of 0.99 (95% CI: 0.97-0.99) for MS; 0.88 [95% CI: 0.86-0.90] for MR; 0.97 [95% CI: 0.95-0.99] for AS; and 0.90 [95% CI: 0.88-0.92]) for AR in the prospective test data set. The limits of agreement (LOA) between the DL algorithm and physician estimates of metrics of valve lesion severities compared to the LOAs between 2 experienced physicians spanned from -0.60 to 0.77 cm² vs -0.48 to 0.44 cm² for MV area; from -0.27 to 0.25 vs -0.23 to 0.08 for MR jet area/left atrial area; from -0.86 to 0.52 m/s vs -0.48 to 0.54 m/s for peak aortic valve blood flow velocity (Vmax); from -10.6 to 9.5 mm Hg vs -10.2 to 4.9 mm Hg for average peak aortic valve gradient; and from -0.39 to 0.32 vs -0.31 to 0.32 for AR jet width/left ventricular outflow tract diameter.

CONCLUSIONS

The proposed deep learning algorithm has the potential to automate and increase efficiency of the clinical workflow for screening echocardiographic images for the presence of VHDs and for quantifying metrics of disease severity.

Left ventricular longitudinal strain variations assessed by speckle-tracking echocardiography after a passive leg raising maneuver in patients with acute circulatory failure to predict fluid responsiveness: A prospective, observational study

Background

An association was reported between the left ventricular longitudinal strain (LV-LS) and preload. LV-LS reflects the left cardiac function curve as it is the ratio of shortening over diastolic dimension. The aim of this study was to determine the sensitivity and specificity of LV-LS variations after a passive leg raising (PLR) maneuver to predict fluid responsiveness in intensive care unit (ICU) patients with acute circulatory failure (ACF).

Methods

Patients with ACF were prospectively included. Preload-dependency was defined as a velocity time integral (VTI) variation greater than 10% between baseline (T0) and PLR (T1), distinguishing the preload-dependent (PLD+) group and the preload-independent (PLD-) group. A 7-cycles, 4-chamber echocardiography loop was registered at T0 and T1, and strain analysis was performed off-line by a blind clinician. A general linear model for repeated measures was used to compare the LV-LS variation (T0 to T1) between the two groups.

Results

From June 2018 to August 2019, 60 patients (PLD+ = 33, PLD- = 27) were consecutively enrolled. The VTI variations after PLR were +21% (±8) in the PLD+ group and -1% (±7) in the PLD- group (p<0.01). Mean baseline LV-LS was -11.3% (±4.2) in the PLD+ group and -13.0% (±4.2) in the PLD- group (p = 0.12). LV-LS increased in the whole population after PLR +16.0% (±4.0) (p = 0.04). The LV-LS variations after PLR were +19.0% (±31) (p = 0.05) in the PLD+ group and +11.0% (±38) (p = 0.25) in the PLD- group, with no significant difference between the two groups (p = 0.08). The area under the curve for the LV-LS variations between T0 and T1 was 0.63 [0.48–0.77].

Conclusion

Our study confirms that LV-LS is load-dependent; however, the variations in LV-LS after PLR is not a discriminating criterion to predict fluid responsiveness of ICU patients with ACF in this cohort.

Intraoperative Renal Resistive Index as an Acute Kidney Injury Biomarker: Development and Validation of an Automated Analysis Algorithm

OBJECTIVE

Intraoperative Doppler-determined renal resistive index (RRI) is a promising early acute kidney injury (AKI) biomarker. As RRI continues to be studied, its clinical usefulness and robustness in research settings will be linked to the ease, efficiency, and precision with which it can be interpreted. Therefore, the authors assessed the usefulness of computer vision technology as an approach to developing an automated RRI-estimating algorithm with equivalent reliability and reproducibility to human experts.

DESIGN

Retrospective.

SETTING

Single-center, university hospital.

PARTICIPANTS

Adult cardiac surgery patients from 7/1/2013 to 7/10/2014 with intraoperative transesophageal echocardiography-determined renal blood flow measurements.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Renal Doppler waveforms were obtained retrospectively and assessed by blinded human expert raters. Images (430) were divided evenly into development and validation cohorts. An algorithm for automated RRI analysis was built using computer vision techniques and tuned for alignment with experts using bootstrap resampling in the development cohort. This algorithm then was applied to the validation cohort for an unbiased assessment of agreement with human experts. Waveform analysis time per image averaged 0.144 seconds. Agreement was excellent by intraclass correlation coefficient (0.939; 95% confidence interval [CI] 0.921 to 0.953) and in Bland-Altman analysis (mean difference [human-algorithm] -0.0015; 95% CI -0.0054 to 0.0024), without evidence of systematic bias.

CONCLUSION

The authors confirmed the value of computer vision technology to develop an algorithm for RRI estimation from automatically processed intraoperative renal Doppler waveforms. This simple-to-use and efficient tool further adds to the clinical and research value of RRI, already the "earliest" among several early AKI biomarkers being assessed.

Automated aortic Doppler flow tracing for reproducible research and clinical measurements

In clinical practice, echocardiographers are often unkeen to make the significant time investment to make additional multiple measurements of Doppler velocity. Main hurdle to obtaining multiple measurements is the time required to manually trace a series of Doppler traces. To make it easier to analyze more beats, we present the description of an application system for automated aortic Doppler envelope quantification, compatible with a range of hardware platforms. It analyses long Doppler strips, spanning many heartbeats, and does not require electrocardiogram to separate individual beats. We tested its measurement of velocity-time-integral and peak-velocity against the reference standard defined as the average of three experts who each made three separate measurements. The automated measurements of velocity-time-integral showed strong correspondence (R(2) = 0.94) and good Bland-Altman agreement (SD = 1.39 cm) with the reference consensus expert values, and indeed performed as well as the individual experts ( R(2) = 0.90 to 0.96, SD = 1.05 to 1.53 cm). The same performance was observed for peak-velocities; ( R(2) = 0.98, SD = 3.07 cm/s) and ( R(2) = 0.93 to 0.98, SD = 2.96 to 5.18 cm/s). This automated technology allows > 10 times as many beats to be analyzed compared to the conventional manual approach. This would make clinical and research protocols more precise for the same operator effort.

Effect of velocity profile skewing on blood velocity and volume flow waveforms derived from maximum Doppler spectral velocity

Given evidence that fully developed axisymmetric flow may be the exception rather than the rule, even in nominally straight arteries, maximum velocity (V(max)) can lie outside the Doppler sample volume (SV). The link between V(max) and derived quantities, such as volume flow (Q), may therefore be more complex than commonly thought. We performed idealized virtual Doppler ultrasound on data from image-based computational fluid dynamics (CFD) models of the normal human carotid artery and investigated how velocity profile skewing and choice of sample volume affected V(max) waveforms and derived Q variables, considering common assumptions about velocity profile shape (i.e., Poiseuille or Womersley). Severe velocity profile skewing caused substantial errors in V(max) waveforms when using a small, centered SV, although peak V(max) was reliably detected; errors with a long SV covering the vessel diameter were orientation dependent but lower overall. Cycle-averaged Q calculated from V(max) was typically within ±15%, although substantial skewing and use of a small SV caused 10%-25% underestimation. Peak Q derived from Womersley's theory was generally accurate to within ±10%. V(max) pulsatility and resistance indexes differed from Q-based values, although the Q-based resistance index could be predicted reliably. Skewing introduced significant error into V(max)-derived Q waveforms, particularly during mid-to-late systole. Our findings suggest that errors in the V(max) and Q waveforms related to velocity profile skewing and use of a small SV, or orientation-dependent errors for a long SV, could limit their use in wave analysis or for constructing characteristic or patient-specific flow boundary conditions for model studies.