Article Text

Download PDFPDF

E Machine learning wall thickness measurement in hypertrophic cardiomyopathy exceeds performance of world experts
  1. João B Augusto1,2,
  2. Rhodri Davies1,2,
  3. Anish Bhuva1,2,
  4. Kristopher Knott1,2,
  5. Andreas Seraphim1,2,
  6. Mashael Al-Farih1,2,
  7. Clement Lau1,3,
  8. Rebecca Hughes1,2,
  9. Luís Lopes1,2,
  10. Hunain Shiwani1,
  11. Bernhard Gerber4,5,
  12. Christian H Craig6,7,
  13. Ntobeko Ntusi8,9,
  14. Gianluca Pontone10,
  15. Milind Y Desai11,
  16. John P Greenwood12,
  17. Peter P Swoboda12,
  18. Gabriella Captur2,
  19. João Cavalcante13,14,
  20. Chiara Bucciarelli-Ducci15,
  21. Steffen E Petersen1,3,
  22. Erik Schelbert16,17,
  23. Charlotte Manisty1,2,
  24. James C Moon1,2
  1. 1Cardiac Imaging Department, Barts Heart Centre, St Bartholomew’s Hospital, London, UK
  2. 2Institute for Cardiovascular Science, University College of London, London, UK
  3. 3William Harvey Research Institute, NIHR Barts Biomedical Research Centre, Queen Mary University of London, UK
  4. 4Division of Cardiology, Department of Cardiovascular Diseases, Cliniques Universitaires St. Luc UCL, Woluwe St. Lambert, Belgium
  5. 5Pôle de Recherche Cardiovasculaire (CARD), Institut de Recherche Expérimentale et Clinique (IREC), Université Catholique de Louvain, Brussels, Belgium
  6. 6The Prince Charles Hospital, Brisbane, Queensland, Australia
  7. 7University of Queensland and Griffith University School of Medicine, Queensland, Australia
  8. 8Division of Cardiology, Department of Medicine, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa
  9. 9Cardiovascular Magnetic Resonance Congress of South Africa, Cape Town, South Africa
  10. 10Cardiac Department, Centro Cardiologico Monzino, Milano, Italy
  11. 11Heart and Vascular Institute Cleveland Clinic Foundation Cleveland OH
  12. 12Multidisciplinary Cardiovascular Research Centre and Division of Biomedical Imaging, Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, UK
  13. 13Minneapolis Heart Institute, Department of Cardiology, Abbott Northwestern Hospital, Minneapolis, Minnesota
  14. 14Valve Science Center, Minneapolis Heart Institute Foundation, Minneapolis, Minnesota
  15. 15Bristol Heart Institute, Bristol National Institute of Health Research (NIHR) Biomedical Research Centre, University Hospitals Bristol NHS Trust and University of Bristol, Bristol, UK
  16. 16Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
  17. 17Cardiovascular Magnetic Resonance Center, UPMC Heart and Vascular Institute, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania


Background Left ventricular maximum wall thickness (MWT) is central to diagnosis and risk stratification of hypertrophic cardiomyopathy (HCM), but measurement has variation.

Objectives We developed a fully automated machine learning (ML) algorithm for MWT measurement and compared it to international experts using precision (repeatability) on a dataset of HCM patients scanned twice with cardiovascular magnetic resonance (CMR).

Methods Training dataset: Endo- and epicardial end-diastolic contours were derived using a fully-automated convolutional neural network trained on 1,923 independent multi-centre multi-disease cases (14 centres from 3 countries, 10 scanner models, 2 field strengths, with balanced pathologies - health, athletes, myocardial infarction, aortic stenosis, HCM, dilated cardiomyopathy, infiltrative diseases) all segmented by a single expert.

Patients: 60 HCM patients were scanned twice (scan:rescan) in the same session (no biological variability) at different field strengths and vendors (Siemens, GE, Philips) in 3 centres to allow generalizability. The protocol consisted of long axis cines and a short axis (SAX) bSSFP cine stack. Between scans, patients were brought out of the bore, repositioned on the table and re-isocentered.

Wall thickness: MWT was measured in the SAX cine stack in end-diastole (scans A and B) by 11 experts (from 4 continents, 6 countries, 9 centers). For ML performance, the contours were based on a repurposed algorithm used for brain cortical thickness measurement, applying the Laplace equation for all contour points – effectively creating nested smoothly deforming surfaces from endo- to epicardium. We created orthogonal field lines to connect endo-and epicardial points, measured these distances and took the maximum as MWT.

Results 1320 MWT measurements by experts were analyzed. Mean MWT varied significantly from 14.9 mm to 19.0 mm (Δ4.1 mm, p<0.05). MWT measured by ML fell in the middle of the experts (5 read higher, 4 lower, p<0.05). Experts had significantly different test:retest precision, ranging from 1.1±0.9 to 3.7±2.0 mm. ML precision performance surpassed all humans on all measures: precision 0.7±0.6 mm, p<0.05; Bland-Altman limits of agreement (ML 3.7 vs humans average 7.7 mm), and coefficient of variance (ML 4.3% vs experts 5.7–12.1%, p<0.05). Using ML, sudden cardiac death risk prediction would be 1.4 to 3.1 times more precise, and a clinical trial to detect a 2 mm MWT interval change would need 1.3 to 4 (mean 2.3) times fewer patients (beta=0.90, alfa=0.05).

Conclusions ML MWT measurement in HCM is superior to all international experts studied with implications for risk stratification and sample sizes for clinical trials.

  • Hypertrophic cardiomyopathy
  • artificial intelligence
  • cardiovascular magnetic resonance.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.