Publications
53 results found
Labs RB, Vrettos A, Loo J, et al., 2022, Automated assessment of transthoracic echocardiogram image quality using deep neural networks, Intelligent Medicine, ISSN: 2667-1026
Ribeiro HDM, Arnold A, Howard JP, et al., 2022, ECG-based real-time arrhythmia monitoring using quantized deep neural networks: A feasibility study, COMPUTERS IN BIOLOGY AND MEDICINE, Vol: 143, ISSN: 0010-4825
- Author Web Link
- Cite
- Citations: 8
Stowell C, Howard J, Demetrescu C, et al., 2021, Fully automated global longitudinal strain assessment using artificial intelligence developed and validated by a UK-wide echocardiography expert collaborative, Publisher: OXFORD UNIV PRESS, Pages: 2-2, ISSN: 0195-668X
Howard J, Stowell C, Cole G, et al., 2021, Automated left ventricular dimension assessment using artificial intelligence developed and validated by a UK-wide collaborative, Circulation: Cardiovascular Imaging, Vol: 14, Pages: 405-415, ISSN: 1941-9651
Background: Echocardiography artificial intelligence (AI) requires training and validation to standards expected of humans. We developed an online platform and established the Unity Collaborative to build a dataset of expertise from 17 hospitals for training, validation, and standardisation of such techniques. Methods: The training dataset were 2056individual frames drawn at random from 1265parasternal long-axis video-loops of patients undergoing clinical echocardiography in 2015-2016. Nine experts labelled these images using our online platform. From this, we trained a convolutional neural network to identify key points. Subsequently, 13 experts labelled a validation dataset of the end-systolic and end-diastolic frame from100new video-loops, twice each. The 26-opinionconsensus was used as the reference standard. The primary outcome was “precision SD”, the standard deviation of difference between AI measurement and expert consensus. Results: In the validation dataset, the AI’s precision SD for left ventricular internal dimension was 3.5mm. For context, precision SD of individual expert measurements against the expert consensus was 4.4mm. Intraclass correlation coefficient (ICC) between AI and expert consensus was 0.926 (95% CI 0.904–0.944), compared with 0.817 (0.778–0.954) between individual experts and expert consensus. For interventricular septum thickness, precision SD was 1.8mm for AI (ICC 0.809; 0.729–0.967), versus 2.0 for individuals (ICC 0.641; 0.568–0.716). For posterior wall thickness, precision SD was 1.4mm for AI (ICC 0.535; 95% CI 0.379–0.661), versus 2.2mm for individuals(0.366; 0.288 to 0.462).We present all images and annotations. This highlights challenging cases, including poor image quality, tapered ventricles, and indistinct boundaries. Conclusions: Experts at multiple institutions successfully cooperated to build a collaborative AI. This performed as well as individual experts. Future echocardiogr
Azarmehr N, Ye X, Howard JP, et al., 2021, Neural architecture search of echocardiography view classifiers, JOURNAL OF MEDICAL IMAGING, Vol: 8, ISSN: 2329-4302
- Author Web Link
- Cite
- Citations: 2
Lane ES, Azarmehr N, Jevsikov J, et al., 2021, Multibeat echocardiographic phase detection using deep neural networks, COMPUTERS IN BIOLOGY AND MEDICINE, Vol: 133, ISSN: 0010-4825
- Author Web Link
- Cite
- Citations: 4
Labs RB, Zolgharni M, Loo JP, 2021, Echocardiographic Image Quality Assessment Using Deep Neural Networks, 25th Conference on Medical Image Understanding and Analysis (MIUA), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 488-502, ISSN: 0302-9743
Ghatwary N, Zolgharni M, Janan F, et al., 2021, Learning Spatiotemporal Features for Esophageal Abnormality Detection From Endoscopic Videos, IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, Vol: 25, Pages: 131-142, ISSN: 2168-2194
- Author Web Link
- Cite
- Citations: 8
Azarmehr N, Ye X, Howes JD, et al., 2020, An optimisation-based iterative approach for speckle tracking echocardiography, MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING, Vol: 58, Pages: 1309-1323, ISSN: 0140-0118
Howard JP, Tan J, Shun-Shin MJ, et al., 2020, Improving ultrasound video classification: an evaluation of novel deep learning methods in echocardiography., J Med Artif Intell, Vol: 3
Echocardiography is the commonest medical ultrasound examination, but automated interpretation is challenging and hinges on correct recognition of the 'view' (imaging plane and orientation). Current state-of-the-art methods for identifying the view computationally involve 2-dimensional convolutional neural networks (CNNs), but these merely classify individual frames of a video in isolation, and ignore information describing the movement of structures throughout the cardiac cycle. Here we explore the efficacy of novel CNN architectures, including time-distributed networks and two-stream networks, which are inspired by advances in human action recognition. We demonstrate that these new architectures more than halve the error rate of traditional CNNs from 8.1% to 3.9%. These advances in accuracy may be due to these networks' ability to track the movement of specific structures such as heart valves throughout the cardiac cycle. Finally, we show the accuracies of these new state-of-the-art networks are approaching expert agreement (3.6% discordance), with a similar pattern of discordance between views.
Azarmehr N, Ye X, Sacchi S, et al., 2020, Segmentation of Left Ventricle in 2D Echocardiography Using Deep Learning, Editors: Zheng, Williams, Chen, Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 497-504, ISBN: 978-3-030-39342-7
- Author Web Link
- Cite
- Citations: 5
Amer A, Ye X, Zolgharni M, et al., 2020, ResDUnet: Residual Dilated UNet for Left Ventricle Segmentation from Echocardiographic Images, 42nd Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC), Publisher: IEEE, Pages: 2019-2022, ISSN: 1557-170X
- Author Web Link
- Cite
- Citations: 11
Ghatwary N, Zolgharni M, Ye X, 2019, Early esophageal adenocarcinoma detection using deep learning methods, INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY, Vol: 14, Pages: 611-621, ISSN: 1861-6410
- Author Web Link
- Cite
- Citations: 39
Ghatwary N, Zolgharni M, Ye X, 2019, GFD Faster R-CNN: Gabor Fractal DenseNet Faster R-CNN for Automatic Detection of Esophageal Abnormalities in Endoscopic Images, 10th International Workshop on Machine Learning in Medical Imaging (MLMI) / 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 89-97, ISSN: 0302-9743
- Author Web Link
- Cite
- Citations: 2
Ghatwary N, Ye X, Zolgharni M, 2019, Esophageal Abnormality Detection Using DenseNet Based Faster R-CNN With Gabor Features, IEEE ACCESS, Vol: 7, Pages: 84374-84385, ISSN: 2169-3536
- Author Web Link
- Cite
- Citations: 26
Sacchi S, Dhutia N, Shun-Shin MJ, et al., 2018, Doppler assessment of aortic stenosis: a 25-operator study demonstrating why reading the peak velocity is superior to velocity time integral, EHJ Cardiovascular Imaging / European Heart Journal - Cardiovascular Imaging, Vol: 19, Pages: 1380-1389, ISSN: 2047-2412
Aims Measurements with superior reproducibility are useful clinically and research purposes. Previous reproducibilitystudies of Doppler assessment of aortic stenosis (AS) have compared only a pair of observers and have notexplored the mechanism by which disagreement between operators occurs. Using custom-designed software whichstored operators’ traces, we investigated the reproducibility of peak and velocity time integral (VTI) measurementsacross a much larger group of operators and explored the mechanisms by which disagreement arose. ...................................................................................................................................................................................................Methodsand resultsTwenty-five observers reviewed continuous wave (CW) aortic valve (AV) and pulsed wave (PW) left ventricularoutflow tract (LVOT) Doppler traces from 20 sequential cases of AS in random order. Each operator unknowinglymeasured each peak velocity and VTI twice. VTI tracings were stored for comparison. Measuring the peak is muchmore reproducible than VTI for both PW (coefficient of variation 10.1 vs. 18.0%; P < 0.001) and CW traces (coeffi-cient of variation 4.0 vs. 10.2%; P < 0.001). VTI is inferior because the steep early and late parts of the envelope aredifficult to trace reproducibly. Dimensionless index improves reproducibility because operators tended to consistentlyover-read or under-read on LVOT and AV traces from the same patient (coefficient of variation 9.3 vs.17.1%; P < 0.001). ...................................................................................................................................................................................................Conclusion It is far more reproducible to measure the peak of a Doppler trace than the VTI, a strategy that reduces measurementvariance by approximately six-fold. Peak measurements are superior to VTI because tracing the steep slopesin th
Shun-Shin M, Cole G, Dhutia N, et al., 2017, The development of automated methods for the reproducible assessment of aortic stenosis, Publisher: OXFORD UNIV PRESS, Pages: 489-489, ISSN: 0195-668X
Zolgharni M, Negoita M, Dhutia NM, et al., 2017, Automatic detection of end-diastolic and end-systolic frames in 2D echocardiography, ECHOCARDIOGRAPHY-A JOURNAL OF CARDIOVASCULAR ULTRASOUND AND ALLIED TECHNIQUES, Vol: 34, Pages: 956-967, ISSN: 0742-2822
Background:Correctly selecting the end-diastolic and end-systolic frames on a 2D echocardiogram is important and challenging, for both human experts and automated algorithms. Manual selection is time-consuming and subject to uncertainty, and may affect the results obtained, especially for advanced measurements such as myocardial strain.Methods and Results:We developed and evaluated algorithms which can automatically extract global and regional cardiac velocity, and identify end-diastolic and end-systolic frames. We acquired apical four-chamber 2D echocardiographic video recordings, each at least 10 heartbeats long, acquired twice at frame rates of 52 and 79 frames/s from 19 patients, yielding 38 recordings. Five experienced echocardiographers independently marked end-systolic and end-diastolic frames for the first 10 heartbeats of each recording. The automated algorithm also did this. Using the average of time points identified by five human operators as the reference gold standard, the individual operators had a root mean square difference from that gold standard of 46.5 ms. The algorithm had a root mean square difference from the human gold standard of 40.5 ms (P<.0001). Put another way, the algorithm-identified time point was an outlier in 122/564 heartbeats (21.6%), whereas the average human operator was an outlier in 254/564 heartbeats (45%).Conclusion:An automated algorithm can identify the end-systolic and end-diastolic frames with performance indistinguishable from that of human experts. This saves staff time, which could therefore be invested in assessing more beats, and reduces uncertainty about the reliability of the choice of frame.
Cole G, Sacchi S, Dhutia N, et al., 2017, DOPPLER ASSESSMENT OF AORTIC STENOSIS: READING THE PEAK VELOCITY IS SUPERIOR TO VELOCITY TIME INTEGRAL, Annual Conference of the British-Cardiovascular-Society (BCS), Publisher: BMJ PUBLISHING GROUP, Pages: A93-A93, ISSN: 1355-6037
Dhutia NM, Zolgharni M, Mielewczik M, et al., 2017, Open-source, vendor-independent, automated multi-beat tissue Doppler echocardiography analysis, International Journal of Cardiovascular Imaging, Vol: 33, Pages: 1135-1148, ISSN: 1569-5794
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.
Negoita M, Zolgharni M, Dadkho E, et al., 2016, Frame rate required for speckle tracking echocardiography: A quantitative clinical study with open-source, vendor-independent software, International Journal of Cardiology, Vol: 218, Pages: 31-36, ISSN: 1874-1754
ObjectivesTo determine the optimal frame rate at which reliable heart walls velocities can be assessed by speckle tracking.BackgroundAssessing left ventricular function with speckle tracking is useful in patient diagnosis but requires a temporal resolution that can follow myocardial motion. In this study we investigated the effect of different frame rates on the accuracy of speckle tracking results, highlighting the temporal resolution where reliable results can be obtained.Material and methods27 patients were scanned at two different frame rates at their resting heart rate. From all acquired loops, lower temporal resolution image sequences were generated by dropping frames, decreasing the frame rate by up to 10-fold.ResultsTissue velocities were estimated by automated speckle tracking. Above 40 frames/s the peak velocity was reliably measured. When frame rate was lower, the inter-frame interval containing the instant of highest velocity also contained lower velocities, and therefore the average velocity in that interval was an underestimate of the clinically desired instantaneous maximum velocity.ConclusionsThe higher the frame rate, the more accurately maximum velocities are identified by speckle tracking, until the frame rate drops below 40 frames/s, beyond which there is little increase in peak velocity. We provide in an online supplement the vendor-independent software we used for automatic speckle-tracked velocity assessment to help others working in this field.
Cole GD, Dhutia NM, Shun-Shin MJ, et al., 2015, Defining the real-world reproducibility of visual grading and visual estimation of left ventricular ejection fraction: impact of image quality, experience and accreditation., International Journal of Cardiovascular Imaging, Vol: 31, Pages: 1303-1314, ISSN: 1569-5794
Left ventricular function can be evaluated by qualitative grading and by eyeball estimation of ejection fraction (EF). We sought to define the reproducibility of these techniques, and how they are affected by image quality, experience and accreditation. Twenty apical four-chamber echocardiographic cine loops (Online Resource 1–20) of varying image quality and left ventricular function were anonymized and presented to 35 operators. Operators were asked to provide (1) a one-phrase grading of global systolic function (2) an “eyeball” EF estimate and (3) an image quality rating on a 0–100 visual analogue scale. Each observer viewed every loop twice unknowingly, a total of 1400 viewings. When grading LV function into five categories, an operator’s chance of agreement with another operator was 50 % and with themself on blinded re-presentation was 68 %. Blinded eyeball LVEF re-estimates by the same operator had standard deviation (SD) of difference of 7.6 EF units, with the SD across operators averaging 8.3 EF units. Image quality, defined as the average of all operators’ assessments, correlated with EF estimate variability (r = −0.616, p < 0.01) and visual grading agreement (r = 0.58, p < 0.01). However, operators’ own single quality assessments were not a useful forewarning of their estimate being an outlier, partly because individual quality assessments had poor within-operator reproducibility (SD of difference 17.8). Reproducibility of visual grading of LV function and LVEF estimation is dependent on image quality, but individuals cannot themselves identify when poor image quality is disrupting their LV function estimate. Clinicians should not assume that patients changing in grade or in visually estimated EF have had a genuine clinical change.
Saura Espin D, Caballero Jimenez L, Oliva Sandoval M, et al., 2014, Oral Abstract session: Demanding measurements: why bother? Thursday 4 December 2014, 16:30-18:00Location: Agora., Eur Heart J Cardiovasc Imaging, Vol: 15 Suppl 2, Pages: ii65-ii67
Dhutia NM, Cole GD, Zolgharni M, et al., 2014, Automated speckle tracking algorithm to aid on-axis imaging in echocardiography, Journal of Medical Imaging, Vol: 1, ISSN: 2329-4310
Obtaining a "correct" view in echocardiography is a subjective process in which an operator attempts to obtain images conforming to consensus standard views. Real-time objective quantification of image alignment may assist less experienced operators, but no reliable index yet exists. We present a fully automated algorithm for detecting incorrect medial/lateral translation of an ultrasound probe by image analysis. The ability of the algorithm to distinguish optimal from sub-optimal four-chamber images was compared to that of specialists-the current "gold-standard." The orientation assessments produced by the automated algorithm correlated well with consensus visual assessments of the specialists ([Formula: see text]) and compared favourably with the correlation between individual specialists and the consensus, [Formula: see text]. Each individual specialist's assessments were within the consensus of other specialists, [Formula: see text] of the time, and the algorithm's assessments were within the consensus of specialists 85% of the time. The mean discrepancy in probe translation values between individual specialists and their consensus was [Formula: see text], and between the automated algorithm and specialists' consensus was [Formula: see text]. This technology could be incorporated into hardware to provide real-time guidance for image optimisation-a potentially valuable tool both for training and quality control.
Dhutia NM, Zolgharni M, Willson K, et al., 2014, Guidance for accurate and consistent tissue Doppler velocity measurement: comparison of echocardiographic methods using a simple vendor-independent method for local validation, EUROPEAN HEART JOURNAL-CARDIOVASCULAR IMAGING, Vol: 15, Pages: 817-827, ISSN: 2047-2404
- Author Web Link
- Cite
- Citations: 13
Zolgharni M, Dhutia NM, Cole GD, et al., 2014, Automated Aortic Doppler Flow Tracing for Reproducible Research and Clinical Measurements, IEEE TRANSACTIONS ON MEDICAL IMAGING, Vol: 33, Pages: 1071-1082, ISSN: 0278-0062
- Author Web Link
- Cite
- Citations: 18
Zolgharni M, Dhutia NM, Cole GD, et al., 2014, Feasibility of using a reliable automated Doppler flow velocity measurements for research and clinical practices, Pages: 90401D-90401D
Dhutia NM, Zolgharni M, Willson K, et al., 2014, Calibration of echocardiographic tissue doppler velocity, using simple, universally-applicable methods, Pages: 90400G-90400G
Zolgharni M, Griffiths H, Ledger PD, 2010, Frequency-difference MIT imaging of cerebral haemorrhage with a hemispherical coil array: numerical modelling, PHYSIOLOGICAL MEASUREMENT, Vol: 31, Pages: S111-S125, ISSN: 0967-3334
- Author Web Link
- Cite
- Citations: 51
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.