Automatic mitral valve inflow measurements from Doppler echocardiography

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.

Harnessing Machine Intelligence in Automatic Echocardiogram Analysis: Current Status, Limitations, and Future Directions

Echocardiography (echo) is a critical tool in diagnosing various cardiovascular diseases. Despite its diagnostic and prognostic value, interpretation and analysis of echo images are still widely performed manually by echocardiographers. A plethora of algorithms has been proposed to analyze medical ultrasound data using signal processing and machine learning techniques. These algorithms provided opportunities for developing automated echo analysis and interpretation systems. The automated approach can significantly assist in decreasing the variability and burden associated with manual image measurements. In this paper, we review the state-of-the-art automatic methods for analyzing echocardiography data. Particularly, we comprehensively and systematically review existing methods of four major tasks: echo quality assessment, view classification, boundary segmentation, and disease diagnosis. Our review covers three echo imaging modes, which are B-mode, M-mode, and Doppler. We also discuss the challenges and limitations of current methods and outline the most pressing directions for future research. In summary, this review presents the current status of automatic echo analysis and discusses the challenges that need to be addressed to obtain robust systems suitable for efficient use in clinical settings or point-of-care testing.

Automatic Recognition of Complete Atrioventricular Activity in Fetal Pulsed-Wave Doppler Signals

Echocardiography is the gold standard for antenatal cardiological assessment. However, the adoption of this technique is challenging, since it is intrinsically operator-dependent and because of the different confounding factors related to the fetal heart size, the fetal movements and the ultrasound artifacts. Among the different options, fetal echocardiography is widely used, concurring to an early diagnosis of several cardiac pathologies. In this work, a neural network-based algorithm targeted at the identification of the most important features of Doppler fetal echocardiography videos is presented and evaluated on real signals. Compared to other approaches, the proposed algorithm works on a couple of ID signals, representing the pulse-wave Doppler envelope extracted from the video, thus preserving a Iightweight approach. For the validation, a small dataset was created, including recordings from five voluntary pregnant women 21^st to 27^th gestational week), for a total of 20 records, 10 seconds each. The dataset was annotated by an expert cardiologist in order to identify the epochs of the signal where a complete readable cardiac cycle could be identified. The performance of the method was evaluated through a 5-fold cross-validation. An average accuracy up to 88% was obtained, confirming the validity of the proposed approach and paving the way to future improvements of the technique.

Extraction of Peak Velocity Profiles from Doppler Echocardiography Using Image Processing

The objective of this study is to extract positive and negative peak velocity profiles from Doppler echocardiographic images. These profiles are currently estimated using tedious manual approaches. Profiles can be used to establish realistic boundary conditions for computational hemodynamic studies and to estimate cardiac time intervals, which are of clinical utility. In the current study, digital image processing algorithms that rely on intensity calculations and two different thresholding methods were proposed and tested. Image intensity histograms were used to guide threshold choices, which were selected such that the resulting velocity profiles appropriately represent Doppler shift envelopes. The resulting peak velocity profiles contained artifacts in the form of sudden velocity changes and possible outliers. To reduce these artifacts, image smoothing using a moving average process was then implemented. Bland-Altman analysis suggested good agreement between the two thresholding methods. Artifacts decreased when image smoothing was performed. Results also suggested that one thresholding method tended to provide the lower limit (i.e., underestimate) of velocities, while the second tended to provide the velocity upper limit (i.e., overestimate). Combining estimates from both methods appeared to provide a smoother peak velocity profile estimate. The proposed automated approach may be useful for objective estimation of peak velocity profiles, which may be helpful for computational hemodynamic studies and may increase the efficiency of current clinical diagnostic tools.

Open-source, vendor-independent, automated multi-beat tissue Doppler echocardiography analysis

Current guidelines for measuring cardiac function by tissue Doppler recommend using multiple beats, but this has a time cost for human operators. We present an open-source, vendor-independent, drag-and-drop software capable of automating the measurement process. A database of ~8000 tissue Doppler beats (48 patients) from the septal and lateral annuli were analyzed by three expert echocardiographers. We developed an intensity- and gradient-based automated algorithm to measure tissue Doppler velocities. We tested its performance against manual measurements from the expert human operators. Our algorithm showed strong agreement with expert human operators. Performance was indistinguishable from a human operator: for algorithm, mean difference and SDD from the mean of human operators’ estimates 0.48 ± 1.12 cm/s (R² = 0.82); for the humans individually this was 0.43 ± 1.11 cm/s (R² = 0.84), −0.88 ± 1.12 cm/s (R² = 0.84) and 0.41 ± 1.30 cm/s (R² = 0.78). Agreement between operators and the automated algorithm was preserved when measuring at either the edge or middle of the trace. The algorithm was 10-fold quicker than manual measurements (p < 0.001). This open-source, vendor-independent, drag-and-drop software can make peak velocity measurements from pulsed wave tissue Doppler traces as accurately as human experts. This automation permits rapid, bias-resistant multi-beat analysis from spectral tissue Doppler images.

The ventricular volume variability study of the Pediatric Heart Network: study design and impact of beat averaging and variable type on the reproducibility of echocardiographic measurements in children with chronic dilated cardiomyopathy

BACKGROUND

Clinical trials often rely on echocardiographic measures of left ventricular size and function as surrogate end points. However, the quantitative impact of factors that affect the reproducibility of these measures is unknown. To address this issue, the National Heart, Lung, and Blood Institute-funded Pediatric Heart Network designed a longitudinal observational study of children with known or suspected dilated cardiomyopathy aged 0 to 22 years from eight pediatric clinical centers.

METHODS

Clinical data were collected together with 150 echocardiographic indices of left ventricular size and function. Separate observers performed duplicate echocardiographic imaging. Multiple observers performed measurements from three cardiac cycles to enable assessment of intraobserver and interobserver variability. The impacts of beat averaging (BA), observer type (local vs core), and variable type (areas, calculations, dimensions, slopes, time intervals, and velocities) on measurement reproducibility were studied. The outcome measure was percentage error (100 × difference/mean).

RESULTS

Of 173 enrolled subjects, 131 met criteria for dilated cardiomyopathy. BA, variable type and observer type all influenced percentage error (P < .0001). Core interobserver percentage error (medians, 11.4%, 10.2%, and 9.3% for BA using one, two, and three beats, respectively) was approximately twice the intraobserver percentage error (medians, 6.3%, 4.9%, and 4.2% for BA using one, two, and three beats, respectively). Slopes and calculated variables exhibited high percentage error despite BA. Chamber dimensions, areas, velocities, and time intervals exhibited low percentage error.

CONCLUSIONS

This comprehensive evaluation of quantitative echocardiographic methods will provide a valuable resource for the design of future pediatric studies. BA and a single core lab observer improve the reproducibility of echocardiographic measurements in children with dilated cardiomyopathy. Certain measurements are highly reproducible, while others, despite BA, are poorly reproducible.

A new automated system to identify a consistent sampling position to make tissue Doppler and transmitral Doppler measurements of E, E' and E/E'

Background

Transmitral pulse wave (PW) Doppler and annular tissue Doppler velocity measurements provide valuable diagnostic and prognostic information. However, they depend on an echocardiographer manually selecting positions to make the measurements. This is time-consuming and open to variability, especially by less experienced operators. We present a new, automated method to select consistent Doppler velocity sites to measure blood flow and muscle function.

Methods

Our automated algorithm combines speckle tracking and colour flow mapping to locate the septal and lateral mitral valve annuli (to measure peak early diastolic velocity, E′) and the mitral valve inflow (to measure peak inflow velocity, E). We also automate peak velocity measurements from resulting PW Doppler traces. The algorithm-selected locations and time taken to identify them were compared against a panel of echo specialists — the current “gold standard”.

Results

The algorithm identified positions to measure Doppler velocities within 3.6 ± 2.2 mm (mitral inflow), 3.2 ± 1.8 mm (septal annulus) and 3.8 ± 1.5 mm (lateral annulus) of the consensus of 3 specialists. This was less than the average 4 mm fidelity with which the specialists could themselves identify the points. The automated algorithm could potentially reduce the time taken to make these measurements by 60 ± 15%.

Conclusions

Our automated algorithm identified sampling positions for measurement of mitral flow, septal and lateral tissue velocities as reliably as specialists. It provides a rapid, easy method for new specialists and potentially non-specialists to make automated measurements of key cardiac physiological indices. This could help support decision-making, without introducing delay and extend availability of echocardiography to more patients.

Shape-based Similarity Retrieval of Doppler Images for Clinical Decision Support

Flow Doppler imaging has become an integral part of an echocardiographic exam. Automated interpretation of flow doppler imaging has so far been restricted to obtaining hemodynamic information from velocity-time profiles depicted in these images. In this paper we exploit the shape patterns in Doppler images to infer the similarity in valvular disease labels for purposes of automated clinical decision support. Specifically, we model the similarity in appearance of Doppler images from the same disease class as a constrained non-rigid translation transform of the velocity envelopes embedded in these images. The shape similarity between two Doppler images is then judged by recovering the alignment transform using a variant of dynamic shape warping. Results of similarity retrieval of doppler images for cardiac decision support on a large database of images are presented.