Roadmap on the use of artificial intelligence for imaging of vulnerable atherosclerotic plaque in coronary arteries

Bernhard Foellmer, Michelle C Williams, Damini Dey, Armin Arbab-Zadeh, Pál Maurovich-Horvat, Rick Volleberg, Daniel Rueckert, Julia A. Schnabel, David E Newby, Marc R Dweck, Giulio Guagliumi, Volkmar Falk, Aldo Javier Vázquez Mézquita, Federico Biavati, Ivana Išgum, Marc Dewey*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

Artificial intelligence (AI) is likely to revolutionise the way medical images are analysed and has the potential to improve identification and analysis of vulnerable, or high-risk, coronary plaque and may thus lead to advances in treating coronary artery disease. However, coronary plaque analysis is challenging due to cardiac and respiratory motion as well as the small size of cardiovascular structures. Moreover, analysing coronary imaging data is currently time-consuming, requires high reader expertise with dedicated cardiovascular imaging training, and is subject to considerable inter- and intra-reader variability. AI has the potential to improve the imaging of vulnerable coronary plaque, but requires robust development, testing, and validation. Combining human reader expertise with AI may provide reliable and valid interpretation of images obtained using computed tomography, magnetic resonance imaging, positron emission tomography, intravascular ultrasound, or optical coherence tomography. In this Roadmap article, we review existing evidence for AI for vulnerable coronary plaque imaging and provide consensus recommendations developed by an interdisciplinary group of experts in AI and non-invasive and invasive coronary imaging. This Roadmap article also outlines future requirements for AI technology to address bias, uncertainty, explainability, and generalisability. This is essential for the clinical acceptance of AI and its use in handling the anticipated growing volume of coronary imaging procedures.
Original languageEnglish
JournalNature Reviews Cardiology
Early online date18 Jul 2023
DOIs
Publication statusE-pub ahead of print - 18 Jul 2023

Fingerprint

Dive into the research topics of 'Roadmap on the use of artificial intelligence for imaging of vulnerable atherosclerotic plaque in coronary arteries'. Together they form a unique fingerprint.

Cite this