Fully Automatic Coronary Calcium Score Software Empowered by Artificial Intelligence Technology: Validation Study Using Three CT Cohorts

Abstract

Objective

This study aimed to validate a deep learning-based fully automatic calcium scoring (coronary artery calcium [CAC]_auto) system using previously published cardiac computed tomography (CT) cohort data with the manually segmented coronary calcium scoring (CAC_hand) system as the reference standard.

Materials and Methods

We developed the CAC_auto system using 100 co-registered, non-enhanced and contrast-enhanced CT scans. For the validation of the CAC_auto system, three previously published CT cohorts (n = 2985) were chosen to represent different clinical scenarios (i.e., 2647 asymptomatic, 220 symptomatic, 118 valve disease) and four CT models. The performance of the CAC_auto system in detecting coronary calcium was determined. The reliability of the system in measuring the Agatston score as compared with CAC_hand was also evaluated per vessel and per patient using intraclass correlation coefficients (ICCs) and Bland-Altman analysis. The agreement between CAC_auto and CAC_hand based on the cardiovascular risk stratification categories (Agatston score: 0, 1–10, 11–100, 101–400, > 400) was evaluated.

Results

In 2985 patients, 6218 coronary calcium lesions were identified using CAC_hand. The per-lesion sensitivity and false-positive rate of the CAC_auto system in detecting coronary calcium were 93.3% (5800 of 6218) and 0.11 false-positive lesions per patient, respectively. The CAC_auto system, in measuring the Agatston score, yielded ICCs of 0.99 for all the vessels (left main 0.91, left anterior descending 0.99, left circumflex 0.96, right coronary 0.99). The limits of agreement between CAC_auto and CAC_hand were 1.6 ± 52.2. The linearly weighted kappa value for the Agatston score categorization was 0.94. The main causes of false-positive results were image noise (29.1%, 97/333 lesions), aortic wall calcification (25.5%, 85/333 lesions), and pericardial calcification (24.3%, 81/333 lesions).

Conclusion

The atlas-based CAC_auto empowered by deep learning provided accurate calcium score measurement as compared with manual method and risk category classification, which co

Related collections

Artificial Intelligence in Medicine

Most cited references29

• Record: found
• Abstract: found
• Article: not found

elastix: a toolbox for intensity-based medical image registration.

Stefan Klein, Marius Staring, Keelin Murphy … (2010)

Medical image registration is an important task in medical image processing. It refers to the process of aligning data sets, possibly from different modalities (e.g., magnetic resonance and computed tomography), different time points (e.g., follow-up scans), and/or different subjects (in case of population studies). A large number of methods for image registration are described in the literature. Unfortunately, there is not one method that works for all applications. We have therefore developed elastix, a publicly available computer program for intensity-based medical image registration. The software consists of a collection of algorithms that are commonly used to solve medical image registration problems. The modular design of elastix allows the user to quickly configure, test, and compare different registration methods for a specific application. The command-line interface enables automated processing of large numbers of data sets, by means of scripting. The usage of elastix for comparing different registration methods is illustrated with three example experiments, in which individual components of the registration method are varied.

0 comments Cited 559 times – based on 0 reviews     Review now

Bookmark

• Record: found
• Abstract: not found
• Article: not found

Quantification of coronary artery calcium using ultrafast computed tomography

Arthur S Agatston, Warren Janowitz, Frank Hildner … (1990)

0 comments Cited 550 times – based on 0 reviews     Review now

Bookmark

• Record: found
• Abstract: found
• Article: not found

Quantification of coronary artery calcium using ultrafast computed tomography.

A Agatston, W Janowitz, F Hildner … (1990)

Ultrafast computed tomography was used to detect and quantify coronary artery calcium levels in 584 subjects (mean age 48 +/- 10 years) with (n = 109) and without (n = 475) clinical coronary artery disease. Fifty patients who underwent fluoroscopy and ultrafast computed tomography were also evaluated. Twenty contiguous 3 mm slices were obtained of the proximal coronary arteries. Total calcium scores were calculated based on the number, areas and peak Hounsfield computed tomographic numbers of the calcific lesions detected. In 88 subjects scored by two readers independently, interobserver agreement was excellent with identical total scores obtained in 70. Ultrafast computed tomography was more sensitive than fluoroscopy, detecting coronary calcium in 90% versus 52% of patients. There were significant differences (p less than 0.0001) in mean total calcium scores for those with versus those without clinical coronary artery disease by decade: 5 versus 132, age 30 to 39 years; 27 versus 291, age 40 to 49 years; 83 versus 462, age 50 to 59 years; and 187 versus 786, age 60 to 69 years. Sensitivity, specificity and predictive values for clinical coronary artery disease were calculated for several total calcium scores in each decade. For age groups 40 to 49 and 50 to 59 years, a total score of 50 resulted in a sensitivity of 71% and 74% and a specificity of 91% and 70%, respectively. For age group 60 to 69 years, a total score of 300 gave a sensitivity of 74% and a specificity of 81%. The negative predictive value of a 0 score was 98%, 94% and 100% for age groups 40 to 49, 50 to 59 and 60 to 69 years, respectively. Ultrafast computed tomography is an excellent tool for detecting and quantifying coronary artery calcium.

0 comments Cited 262 times – based on 0 reviews     Review now

Bookmark

All references

Author and article information

Journal

Journal ID (nlm-ta): Korean J Radiol

Journal ID (iso-abbrev): Korean J Radiol

Journal ID (publisher-id): KJR

Title: Korean Journal of Radiology

Publisher: The Korean Society of Radiology

ISSN (Print): 1229-6929

ISSN (Electronic): 2005-8330

Publication date (Print): November 2021

Publication date (Electronic): 26 July 2021

Volume: 22

Issue: 11

Pages: 1764-1776

Affiliations

[1 ]Biomedical Engineering Research Center, Asan Institute for Life Sciences, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea.

[2 ]Department of Radiology and Research Institute of Radiology, Cardiac Imaging Center, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea.

[3 ]Divison of Cardiology, Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea.

Author notes

Corresponding author: Dong Hyun Yang, MD, PhD. Department of Radiology and Research Institute of Radiology, Cardiac Imaging Center, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea. donghyun.yang@ 123456gmail.com

Corresponding author: Young-Hak Kim, MD. Division of Cardiology, Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88 Olumpicro 43-gil, Songpa-gu, Seoul 05505, Korea. mdyhkim@ 123456amc.seoul.kr

^*These authors contributed equally to this work.

Author information

June-Goo Lee https://orcid.org/0000-0002-1380-6682

HeeSoo Kim https://orcid.org/0000-0002-6715-9628

Heejun Kang https://orcid.org/0000-0002-0396-2112

Hyun Jung Koo https://orcid.org/0000-0001-5640-3835

Joon-Won Kang https://orcid.org/0000-0001-6478-0390

Young-Hak Kim https://orcid.org/0000-0002-3610-486X

Dong Hyun Yang https://orcid.org/0000-0001-5477-558X

Article

DOI: 10.3348/kjr.2021.0148

PMC ID: 8546141

PubMed ID: 34402248

SO-VID: 8e8e770e-542e-469a-abe8-8f5630cb0c56

Copyright © Copyright © 2021 The Korean Society of Radiology

License:

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

History

Date received : 19 February 2021

Date revision received : 26 April 2021

Date accepted : 13 May 2021

Funding

Funded by: Korea Health Industry Development Institute, CrossRef https://doi.org/10.13039/501100003710;

Award ID: HR20C0026020021

Funded by: Institute for Information & Communications Technology Promotion, CrossRef https://doi.org/10.13039/501100010418;

Award ID: 2018-0-00861

Funded by: Asan Medical Center;

Award ID: 2017-7036

Categories

Subject: Cardiovascular Imaging

Subject: Original Article

ScienceOpen disciplines: Radiology & Imaging

Keywords: coronary artery calcium score,computed tomography,artificial intelligence,accuracy

Data availability:

ScienceOpen disciplines: Radiology & Imaging

Keywords: coronary artery calcium score, computed tomography, artificial intelligence, accuracy

Comments

Comment on this article

Sign in to comment

scite_

Similar content155

• Mining the Human Phenome Using Allelic Scores That Index Biological Intermediates

  Authors: David M Evans, Marie Jo A. Brion, Lavinia Paternoster …
• Feasibility of TEE-guided stroke risk assessment in atrial fibrillation—background, aims, design and baseline data of the TIARA pilot study

  Authors: T. T. N. Dinh, L. Baur, R. Pisters …
• Sex-specific differences in physical health and health services use among Canadian Veterans: a retrospective cohort study using healthcare administrative data

  Authors: Kate St Cyr, A B Aiken, H Cramm …

See all similar

Cited by17

• Association of Sinoatrial Node Radiation Dose With Atrial Fibrillation and Mortality in Patients With Lung Cancer

  Authors: Kyung Hwan Kim, Jaewon Oh, Gowoon Yang …
• Fully automated deep learning powered calcium scoring in patients undergoing myocardial perfusion imaging

  Authors: Thomas Sartoretti, Antonio G. Gennari, Elisabeth Sartoretti …
• Current and Future Applications of Artificial Intelligence in Coronary Artery Disease

  Authors: Nitesh Gautam, Prachi Saluja, Abdallah Malkawi …

See all cited by

Most referenced authors439

• al et
• YH Kim
• H. K. KIM

See all reference authors

• Version 1 - Current
  • Version 1