João B Augusto1, Rhodri H Davies1, Anish N Bhuva1, Kristopher D Knott1, Andreas Seraphim1, Mashael Alfarih1, Clement Lau2, Rebecca K Hughes1, Luís R Lopes1, Hunain Shiwani3, Thomas A Treibel1, Bernhard L Gerber4, Christian Hamilton-Craig5, Ntobeko A B Ntusi6, Gianluca Pontone7, Milind Y Desai8, John P Greenwood9, Peter P Swoboda9, Gabriella Captur10, João Cavalcante11, Chiara Bucciarelli-Ducci12, Steffen E Petersen2, Erik Schelbert13, Charlotte Manisty1, James C Moon14. 1. Cardiac Imaging Department, Barts Heart Centre, St Bartholomew's Hospital, London, UK; Institute of Cardiovascular Science, University College London, London, UK. 2. Cardiac Imaging Department, Barts Heart Centre, St Bartholomew's Hospital, London, UK; William Harvey Research Institute, NIHR Barts Biomedical Research Centre, Queen Mary University of London, London, UK. 3. Cardiac Imaging Department, Barts Heart Centre, St Bartholomew's Hospital, London, UK. 4. Division of Cardiology, Department of Cardiovascular Diseases, Cliniques Universitaires St Luc UCL, Woluwe St Lambert, Belgium; Pôle de Recherche Cardiovasculaire, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium. 5. The Prince Charles Hospital, Brisbane, QLD, Australia; Centre for Advanced Imaging, University of Queensland and Griffith University School of Medicine, QLD, Australia. 6. Division of Cardiology, Department of Medicine, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa; Hatter Institute of Cardiovascular Research in Africa and Cape Universities Body Imaging Centre, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa. 7. Department of Cardiovascular Imaging, Centro Cardiologico Monzino IRCCS, Milan, Italy. 8. Heart and Vascular Institute Cleveland Clinic, Cleveland, OH, USA. 9. Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, and Leeds Teaching Hospitals NHS Trust, UK. 10. Institute of Cardiovascular Science, University College London, London, UK. 11. Minneapolis Heart Institute, Department of Cardiology, Abbott Northwestern Hospital, Minneapolis, MN, USA; Valve Science Center, Minneapolis Heart Institute Foundation, Minneapolis, MN, USA. 12. Bristol Heart Institute, Bristol National Institute of Health Research Biomedical Research Centre, University Hospitals Bristol NHS Trust and University of Bristol, Bristol, UK. 13. Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA; Cardiovascular Magnetic Resonance Center, UPMC Heart and Vascular Institute, University of Pittsburgh Medical Center, Pittsburgh, PA, USA. 14. Cardiac Imaging Department, Barts Heart Centre, St Bartholomew's Hospital, London, UK; Institute of Cardiovascular Science, University College London, London, UK. Electronic address: j.moon@ucl.ac.uk.
Abstract
BACKGROUND: Left ventricular maximum wall thickness (MWT) is central to diagnosis and risk stratification of hypertrophic cardiomyopathy, but human measurement is prone to variability. We developed an automated machine learning algorithm for MWT measurement and compared precision (reproducibility) with that of 11 international experts, using a dataset of patients with hypertrophic cardiomyopathy. METHODS: 60 adult patients with hypertrophic cardiomyopathy, including those carrying hypertrophic cardiomyopathy gene mutations, were recruited at three institutes in the UK from August, 2018, to September, 2019: Barts Heart Centre, University College London Hospital (The Heart Hospital), and Leeds Teaching Hospitals NHS Trust. Participants had two cardiovascular magnetic resonance scans (test and retest) on the same day, ensuring no biological variability, using four cardiac MRI scanner models represented across two manufacturers and two field strengths. End-diastolic short-axis MWT was measured in test and retest by 11 international experts (from nine centres in six countries) and an automated machine learning method, which was trained to segment endocardial and epicardial contours on an independent, multicentre, multidisease dataset of 1923 patients. Machine learning MWT measurement was done with a method based on solving Laplace's equation. To assess test-retest reproducibility, we estimated the absolute test-retest MWT difference (precision), the coefficient of variation (CoV) for duplicate measurements, and the number of patients reclassified between test and retest according to different thresholds (MWT >15 mm and >30 mm). We calculated the sample size required to detect a prespecified MWT change between pairs of scans for machine learning and each expert. FINDINGS: 1440 MWT measurements were analysed, corresponding to two scans from 60 participants by 12 observers (11 experts and machine learning). Experts differed in the MWT they measured, ranging from 14·9 mm (SD 4·2) to 19·0 mm (4·7; p<0·0001 for trend). Machine learning-measured mean MWT was 16·8 mm (4·1). Machine learning precision was superior, with a test-retest difference of 0·7 mm (0·6) compared with experts, who ranged from 1·1 mm (0·9) to 3·7 mm (2·0; p values for machine learning vs expert comparison ranging from <0·0001 to 0·0073) and a significantly lower CoV than for all experts (4·3% [95% CI 3·3-5·1] vs 5·7-12·1% across experts). On average, 38 (64%) patients were designated as having MWT greater than 15 mm by machine learning compared with 27 (45%) to 50 (83%) patients by experts; five (8%) patients were reclassified in test-retest by machine learning compared with four (7%) to 12 (20%) by experts. With a cutoff point of more than 30 mm for implantable cardioverter-defibrillator, three experts would have changed recommendations between tests a total of four times, but machine learning was consistent. Using machine learning, a clinical trial to detect a 2 mm MWT change would need 2·3 times (range 1·6-4·6) fewer patients. INTERPRETATION: In this preliminary study, machine learning MWT measurement in hypertrophic cardiomyopathy is superior to human experts with potential implications for diagnosis, risk stratification, and clinical trials. FUNDING: European Regional Development Fund and Barts Charity.
BACKGROUND: Left ventricular maximum wall thickness (MWT) is central to diagnosis and risk stratification of hypertrophic cardiomyopathy, but human measurement is prone to variability. We developed an automated machine learning algorithm for MWT measurement and compared precision (reproducibility) with that of 11 international experts, using a dataset of patients with hypertrophic cardiomyopathy. METHODS: 60 adult patients with hypertrophic cardiomyopathy, including those carrying hypertrophic cardiomyopathy gene mutations, were recruited at three institutes in the UK from August, 2018, to September, 2019: Barts Heart Centre, University College London Hospital (The Heart Hospital), and Leeds Teaching Hospitals NHS Trust. Participants had two cardiovascular magnetic resonance scans (test and retest) on the same day, ensuring no biological variability, using four cardiac MRI scanner models represented across two manufacturers and two field strengths. End-diastolic short-axis MWT was measured in test and retest by 11 international experts (from nine centres in six countries) and an automated machine learning method, which was trained to segment endocardial and epicardial contours on an independent, multicentre, multidisease dataset of 1923 patients. Machine learning MWT measurement was done with a method based on solving Laplace's equation. To assess test-retest reproducibility, we estimated the absolute test-retest MWT difference (precision), the coefficient of variation (CoV) for duplicate measurements, and the number of patients reclassified between test and retest according to different thresholds (MWT >15 mm and >30 mm). We calculated the sample size required to detect a prespecified MWT change between pairs of scans for machine learning and each expert. FINDINGS: 1440 MWT measurements were analysed, corresponding to two scans from 60 participants by 12 observers (11 experts and machine learning). Experts differed in the MWT they measured, ranging from 14·9 mm (SD 4·2) to 19·0 mm (4·7; p<0·0001 for trend). Machine learning-measured mean MWT was 16·8 mm (4·1). Machine learning precision was superior, with a test-retest difference of 0·7 mm (0·6) compared with experts, who ranged from 1·1 mm (0·9) to 3·7 mm (2·0; p values for machine learning vs expert comparison ranging from <0·0001 to 0·0073) and a significantly lower CoV than for all experts (4·3% [95% CI 3·3-5·1] vs 5·7-12·1% across experts). On average, 38 (64%) patients were designated as having MWT greater than 15 mm by machine learning compared with 27 (45%) to 50 (83%) patients by experts; five (8%) patients were reclassified in test-retest by machine learning compared with four (7%) to 12 (20%) by experts. With a cutoff point of more than 30 mm for implantable cardioverter-defibrillator, three experts would have changed recommendations between tests a total of four times, but machine learning was consistent. Using machine learning, a clinical trial to detect a 2 mm MWT change would need 2·3 times (range 1·6-4·6) fewer patients. INTERPRETATION: In this preliminary study, machine learning MWT measurement in hypertrophic cardiomyopathy is superior to human experts with potential implications for diagnosis, risk stratification, and clinical trials. FUNDING: European Regional Development Fund and Barts Charity.
Authors: James P Howard; Catherine C Stowell; Graham D Cole; Kajaluxy Ananthan; Camelia D Demetrescu; Keith Pearce; Ronak Rajani; Jobanpreet Sehmi; Kavitha Vimalesvaran; G Sunthar Kanaganayagam; Eleanor McPhail; Arjun K Ghosh; John B Chambers; Amar P Singh; Massoud Zolgharni; Bushra Rana; Darrel P Francis; Matthew J Shun-Shin Journal: Circ Cardiovasc Imaging Date: 2021-05-17 Impact factor: 7.792
Authors: Krit Dwivedi; Michael Sharkey; Robin Condliffe; Johanna M Uthoff; Samer Alabed; Peter Metherall; Haiping Lu; Jim M Wild; Eric A Hoffman; Andrew J Swift; David G Kiely Journal: Diagnostics (Basel) Date: 2021-04-09
Authors: Rob Van Der Geest; Andrew J Swift; Faisal Alandejani; Samer Alabed; Pankaj Garg; Ze Ming Goh; Kavita Karunasaagarar; Michael Sharkey; Mahan Salehi; Ziad Aldabbagh; Krit Dwivedi; Michail Mamalakis; Pete Metherall; Johanna Uthoff; Chris Johns; Alexander Rothman; Robin Condliffe; Abdul Hameed; Athanasios Charalampoplous; Haiping Lu; Sven Plein; John P Greenwood; Allan Lawrie; Jim M Wild; Patrick J H de Koning; David G Kiely Journal: J Cardiovasc Magn Reson Date: 2022-04-07 Impact factor: 6.903
Authors: Miroslawa Gorecka; Gerry P McCann; Colin Berry; Vanessa M Ferreira; James C Moon; Christopher A Miller; Amedeo Chiribiri; Sanjay Prasad; Marc R Dweck; Chiara Bucciarelli-Ducci; Dana Dawson; Marianna Fontana; Peter W Macfarlane; Alex McConnachie; Stefan Neubauer; John P Greenwood Journal: J Cardiovasc Magn Reson Date: 2021-06-10 Impact factor: 5.364
Authors: Antonio de Marvao; Kathryn A McGurk; Sean L Zheng; Marjola Thanaj; Wenjia Bai; Jinming Duan; Carlo Biffi; Francesco Mazzarotto; Ben Statton; Timothy J W Dawes; Nicolò Savioli; Brian P Halliday; Xiao Xu; Rachel J Buchan; A John Baksi; Marina Quinlan; Paweł Tokarczuk; Upasana Tayal; Catherine Francis; Nicola Whiffin; Pantazis I Theotokis; Xiaolei Zhang; Mikyung Jang; Alaine Berry; Antonis Pantazis; Paul J R Barton; Daniel Rueckert; Sanjay K Prasad; Roddy Walsh; Carolyn Y Ho; Stuart A Cook; James S Ware; Declan P O'Regan Journal: J Am Coll Cardiol Date: 2021-09-14 Impact factor: 24.094
Authors: Arka Das; Christopher Kelly; Irvin Teh; Christopher Nguyen; Louise A E Brown; Amrit Chowdhary; Nicholas Jex; Sharmaine Thirunavukarasu; Noor Sharrack; Miroslawa Gorecka; Peter P Swoboda; John P Greenwood; Peter Kellman; James C Moon; Rhodri H Davies; Luis R Lopes; George Joy; Sven Plein; Jürgen E Schneider; Erica Dall'Armellina Journal: Eur Heart J Cardiovasc Imaging Date: 2022-02-22 Impact factor: 6.875