Login
Search
Search
0 Dates
2024
2023
2022
2021
2020
2019
2018
0 Events
CPC 2018
CPC 2019
Curso de Atualização em Medicina Cardiovascular 2019
Reunião Anual Conjunta dos Grupos de Estudo de Cirurgia Cardíaca, Doenças Valvulares e Ecocardiografia da SPC
CPC 2020
CPC 2021
CPC 2022
CPC 2023
CPC 2024
0 Topics
A. Basics
B. Imaging
C. Arrhythmias and Device Therapy
D. Heart Failure
E. Coronary Artery Disease, Acute Coronary Syndromes, Acute Cardiac Care
F. Valvular, Myocardial, Pericardial, Pulmonary, Congenital Heart Disease
G. Aortic Disease, Peripheral Vascular Disease, Stroke
H. Interventional Cardiology and Cardiovascular Surgery
I. Hypertension
J. Preventive Cardiology
K. Cardiovascular Disease In Special Populations
L. Cardiovascular Pharmacology
M. Cardiovascular Nursing
N. E-Cardiology / Digital Health, Public Health, Health Economics, Research Methodology
O. Basic Science
P. Other
0 Themes
01. History of Cardiology
02. Clinical Skills
03. Imaging
04. Arrhythmias, General
05. Atrial Fibrillation
06. Supraventricular Tachycardia (non-AF)
07. Syncope and Bradycardia
08. Ventricular Arrhythmias and Sudden Cardiac Death (SCD)
09. Device Therapy
10. Chronic Heart Failure
11. Acute Heart Failure
12. Coronary Artery Disease (Chronic)
13. Acute Coronary Syndromes
14. Acute Cardiac Care
15. Valvular Heart Disease
16. Infective Endocarditis
17. Myocardial Disease
18. Pericardial Disease
19. Tumors of the Heart
20. Congenital Heart Disease and Pediatric Cardiology
21. Pulmonary Circulation, Pulmonary Embolism, Right Heart Failure
22. Aortic Disease
23. Peripheral Vascular and Cerebrovascular Disease
24. Stroke
25. Interventional Cardiology
26. Cardiovascular Surgery
27. Hypertension
28. Risk Factors and Prevention
29. Rehabilitation and Sports Cardiology
30. Cardiovascular Disease in Special Populations
31. Pharmacology and Pharmacotherapy
32. Cardiovascular Nursing
33. e-Cardiology / Digital Health
34. Public Health and Health Economics
35. Research Methodology
36. Basic Science
37. Miscellanea
0 Resources
Abstract
Slides
Vídeo
Report
CLEAR FILTERS
Machine learning wall thickness measurement in Hypertrophic Cardiomyopathy exceeds performance of world experts
Session:
Prémio Jovem Investigador
Speaker:
João Bicho Augusto
Congress:
CPC 2020
Topic:
B. Imaging
Theme:
03. Imaging
Subtheme:
03.3 Cardiac Magnetic Resonance
Session Type:
Prémios
FP Number:
---
Authors:
João Bicho Augusto; Rhodri Davies; Anish Bhuva; Kristopher Knott; Mashael Al-Farih; Clement Lau; Rebecca Hughes; Andreas Seraphim; Luís Lopes; Gabriella Captur; Peter Kellman; Bernhard Gerber; Ntobeko Ntusi; Milind Y Desai; Christian Hamilton Craig; João Cavalcante; Gianluca Pontone; Erik Schelbert; Chiara Bucciarelli-Ducci; Steffen E Petersen; Charlotte Manisty; James C Moon
Abstract
<p><strong>Background: </strong>Maximum wall thickness (MWT) is essential in hypertrophic cardiomyopathy (HCM) diagnosis and risk stratification, but its measurement is not standardized. Cardiac MRI (CMR) is the modality of choice to measure MWT, but human observers still make mistakes: (1) they do not measure every single myocardium segment in every single slice, (2) it is difficult to calliper two non-parallel edges and find where MWT is truly maximum and (3) the length of the calliper can be confounded by discrepancies in edge detection (e.g. trabeculae, blood-myocardium interface). </p> <p>We developed a fully automated machine learning (ML) algorithm, optimized it for MWT measurement in HCM, and compared this to the human performance of 11 world experts in CMR and HCM, using precision (repeatability) applied to a dataset of patients scanned twice.</p> <p><strong>Methods:</strong> <br /> <em>Training dataset</em>: 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.</p> <p><em>Patients:</em> 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 5 other centres to allow generalizability. The protocol consisted of long axes 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.</p> <p><em>Wall thickness:</em> MWT was measured in the SAX cine stack in end-diastole (scans A and B) by 11 experts (from 6 countries). 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.</p> <p><strong>Results:</strong> The ML was more precise than experts across several metrics: (1) absolute MWT difference between test-retest for ML was under-millimetric (0.7±0.6mm) and significantly inferior to all other experts (p<0.05, Table), (2) Bland-Altman limits of agreement interval (upper minus lower limit) were narrower in ML (3.7 vs 7.7mm for experts average), (3) coefficient of variation was lower in ML (4.3 [3.3–5.1]%) than all experts (p<0.05 versus each expert, Table). A study designed to detect a 2mm change in MWT as an endpoint would need on average 2.3 times fewer patients (1.3 to 4 times) if analyzed by ML than by humans (90% power, alfa=0.05).</p> <p><strong>Conclusions:</strong> ML fully automated MWT measurement in HCM is feasible and is more precise than human experts.</p>
Video
Our mission: To reduce the burden of cardiovascular disease
Visit our site