Application of Artificial Intelligence in Coronary Computed Tomography Angiography

A. Selvarajah, M. Bennamoun, D. Playford, B. J.W. Chow, Girish Dwivedi

Research output: Contribution to journalReview article

1 Citation (Scopus)

Abstract

Purpose of Review: This article summarizes the currently available published literature with regard to the applications of artificial intelligence in cardiac computed tomography angiography. Recent Findings: Recent studies and emerging data demonstrate feasibility of artificial intelligence-based high-level image analysis and interpretation tools that will likely enable medical practitioners to achieve more accurate diagnosis of coronary artery disease. Emerging artificial intelligence-based computational modeling methods will assist with pre-operative planning for valve disease. Finally, early but significant work is also being performed in relation to real-time assessment of myocardial perfusion and fractional flow reserve using machine learning. Summary: We anticipate that within the next 5 years, the level of artificial intelligence-driven automation for the analysis and interpretation of cardiac computed tomography angiography will be significantly higher than what is available today. It is also expected that the most productive applications of artificial intelligence in cardiac computed tomography angiography will involve deep learning, utilizing a combination of imaging data and adjunctive data mining to generate more accurate and personalized diagnoses and risk metrics.

Original languageEnglish
Article number12
JournalCurrent Cardiovascular Imaging Reports
Volume11
Issue number6
DOIs
Publication statusPublished - 1 Jun 2018

Fingerprint

Artificial Intelligence
Myocardial Fractional Flow Reserve
Data Mining
Automation
Coronary Artery Disease
Perfusion
Computed Tomography Angiography
Learning

Cite this

@article{49d0e9792f1f492cbd045c39032e6bf4,
title = "Application of Artificial Intelligence in Coronary Computed Tomography Angiography",
abstract = "Purpose of Review: This article summarizes the currently available published literature with regard to the applications of artificial intelligence in cardiac computed tomography angiography. Recent Findings: Recent studies and emerging data demonstrate feasibility of artificial intelligence-based high-level image analysis and interpretation tools that will likely enable medical practitioners to achieve more accurate diagnosis of coronary artery disease. Emerging artificial intelligence-based computational modeling methods will assist with pre-operative planning for valve disease. Finally, early but significant work is also being performed in relation to real-time assessment of myocardial perfusion and fractional flow reserve using machine learning. Summary: We anticipate that within the next 5 years, the level of artificial intelligence-driven automation for the analysis and interpretation of cardiac computed tomography angiography will be significantly higher than what is available today. It is also expected that the most productive applications of artificial intelligence in cardiac computed tomography angiography will involve deep learning, utilizing a combination of imaging data and adjunctive data mining to generate more accurate and personalized diagnoses and risk metrics.",
keywords = "Artificial intelligence, Computed tomography, Convolutional neural network, Deep learning, Machine learning",
author = "A. Selvarajah and M. Bennamoun and D. Playford and Chow, {B. J.W.} and Girish Dwivedi",
year = "2018",
month = "6",
day = "1",
doi = "10.1007/s12410-018-9453-5",
language = "English",
volume = "11",
journal = "Current Cardiovascular Imaging Reports",
issn = "1941-9066",
publisher = "Current Medicine Group",
number = "6",

}

Application of Artificial Intelligence in Coronary Computed Tomography Angiography. / Selvarajah, A.; Bennamoun, M.; Playford, D.; Chow, B. J.W.; Dwivedi, Girish.

In: Current Cardiovascular Imaging Reports, Vol. 11, No. 6, 12, 01.06.2018.

Research output: Contribution to journalReview article

TY - JOUR

T1 - Application of Artificial Intelligence in Coronary Computed Tomography Angiography

AU - Selvarajah, A.

AU - Bennamoun, M.

AU - Playford, D.

AU - Chow, B. J.W.

AU - Dwivedi, Girish

PY - 2018/6/1

Y1 - 2018/6/1

N2 - Purpose of Review: This article summarizes the currently available published literature with regard to the applications of artificial intelligence in cardiac computed tomography angiography. Recent Findings: Recent studies and emerging data demonstrate feasibility of artificial intelligence-based high-level image analysis and interpretation tools that will likely enable medical practitioners to achieve more accurate diagnosis of coronary artery disease. Emerging artificial intelligence-based computational modeling methods will assist with pre-operative planning for valve disease. Finally, early but significant work is also being performed in relation to real-time assessment of myocardial perfusion and fractional flow reserve using machine learning. Summary: We anticipate that within the next 5 years, the level of artificial intelligence-driven automation for the analysis and interpretation of cardiac computed tomography angiography will be significantly higher than what is available today. It is also expected that the most productive applications of artificial intelligence in cardiac computed tomography angiography will involve deep learning, utilizing a combination of imaging data and adjunctive data mining to generate more accurate and personalized diagnoses and risk metrics.

AB - Purpose of Review: This article summarizes the currently available published literature with regard to the applications of artificial intelligence in cardiac computed tomography angiography. Recent Findings: Recent studies and emerging data demonstrate feasibility of artificial intelligence-based high-level image analysis and interpretation tools that will likely enable medical practitioners to achieve more accurate diagnosis of coronary artery disease. Emerging artificial intelligence-based computational modeling methods will assist with pre-operative planning for valve disease. Finally, early but significant work is also being performed in relation to real-time assessment of myocardial perfusion and fractional flow reserve using machine learning. Summary: We anticipate that within the next 5 years, the level of artificial intelligence-driven automation for the analysis and interpretation of cardiac computed tomography angiography will be significantly higher than what is available today. It is also expected that the most productive applications of artificial intelligence in cardiac computed tomography angiography will involve deep learning, utilizing a combination of imaging data and adjunctive data mining to generate more accurate and personalized diagnoses and risk metrics.

KW - Artificial intelligence

KW - Computed tomography

KW - Convolutional neural network

KW - Deep learning

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85049312344&partnerID=8YFLogxK

U2 - 10.1007/s12410-018-9453-5

DO - 10.1007/s12410-018-9453-5

M3 - Review article

VL - 11

JO - Current Cardiovascular Imaging Reports

JF - Current Cardiovascular Imaging Reports

SN - 1941-9066

IS - 6

M1 - 12

ER -