Direct head-to-head comparison of convolutional long short-term memory and transformer networks for artificial Intelligence-based quantification of atherosclerotic plaque and stenosis from coronary CT angiography

Nipun Manral, Andrew Lin, Caroline Park, Priscilla McElhinney, Aditya Killekar, Hidenari Matsumoto, Jacek Kwiecinski, Konrad Pieszko, Aryabod Razipour, Kajetan Grodecki, Mhairi Doris, Alan C. Kwan, Donghee Han, Keiichiro Kuronuma, Guadalupe Flores Tomasino, Evangelos Tzolos, Aakash Shanbhag, Markus Goeller, Mohamed Marwan, Sebastien CadetStephan Achenbach, Stephen J. Nicholls, Dennis T. Wong, Daniel S. Berman, Marc Dweck, David E. Newby, Michelle C. Williams, Piotr J. Slomka, Damini Dey

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Background: Coronary computed tomography angiography (CCTA) enables non-invasive assessment of luminal stenosis and coronary atherosclerotic plaque. We aimed to directly compare the performance of 2 novel deep learning networks—convolutional long short-term memory and transformer network—for artificial intelligence-based quantification of plaque volume and stenosis severity from CCTA. Methods: This was an international multicenter study of patients undergoing CCTA at 11 sites. The deep learning (DL) convolutional neural networks were trained to segment coronary plaque in 921 patients (5,045 lesions). The training dataset was further split temporally into training (80%) and internal validation (20%) datasets. The primary DL architecture was a hierarchical convolutional long short- term memory (ConvLSTM) network. This was compared against a TransUNet network, which combines the abilities of Vision Transformer with U-Net, enabling the capture of in-depth localization information while modeling long-range dependencies. Following training and internal validation, the both DL networks were applied to an external validation cohort of 162 patients (1,468 lesions) from the SCOT-HEART trial. Results: In the external validation cohort, agreement between DL and expert reader measurements was stronger when using the ConvLSTM network than with TransUNet, for both per-lesion total plaque volume (ICC 0·953 vs 0.830) and percent diameter stenosis (ICC 0·882 vs 0.735; both p<0.001). The ConvLSTM network showed higher per-cross-section overlap with expert reader segmentations (as measured by the Dice coefficient) compared to TransUnet, for vessel wall (0.947 vs 0.946), lumen (0.93 vs 0.92), and calcified plaque (0.87 vs 0.86; p<0.0001 for all), with similar execution times. Conclusions: In a direct comparison with external validation, the ConvLSTM network yielded higher agreement with expert readers for quantification of total plaque volume and stenosis severity compared to TransUnet, with faster execution times.

Original languageEnglish
Title of host publicationMedical Imaging 2023
Subtitle of host publicationImage Processing
EditorsOlivier Colliot, Ivana Isgum
PublisherSPIE
ISBN (Electronic)9781510660335
DOIs
Publication statusPublished - 2023
Externally publishedYes
EventMedical Imaging 2023: Image Processing - San Diego, United States
Duration: 19 Feb 202323 Feb 2023

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume12464
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2023: Image Processing
Country/TerritoryUnited States
CitySan Diego
Period19/02/2323/02/23

Keywords

  • artificial intelligence
  • atherosclerosis
  • convolutional long short-term memory network
  • coronary computed tomography angiography
  • deep learning
  • transformers

Fingerprint

Dive into the research topics of 'Direct head-to-head comparison of convolutional long short-term memory and transformer networks for artificial Intelligence-based quantification of atherosclerotic plaque and stenosis from coronary CT angiography'. Together they form a unique fingerprint.

Cite this