eduzhai > Applied Sciences > Engineering >

Deep Learning from Dual-Energy Information for Whole-Heart Segmentation in Dual-Energy and Single-Energy Non-Contrast-Enhanced Cardiac CT

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 31 pages

Abstract: Deep learning-based whole-heart segmentation in coronary CT angiography(CCTA) allows the extraction of quantitative imaging measures forcardiovascular risk prediction. Automatic extraction of these measures inpatients undergoing only non-contrast-enhanced CT (NCCT) scanning would bevaluable. In this work, we leverage information provided by a dual-layerdetector CT scanner to obtain a reference standard in virtual non-contrast(VNC) CT images mimicking NCCT images, and train a 3D convolutional neuralnetwork (CNN) for the segmentation of VNC as well as NCCT images.Contrast-enhanced acquisitions on a dual-layer detector CT scanner werereconstructed into a CCTA and a perfectly aligned VNC image. In each CCTAimage, manual reference segmentations of the left ventricular (LV) myocardium,LV cavity, right ventricle, left atrium, right atrium, ascending aorta, andpulmonary artery trunk were obtained and propagated to the corresponding VNCimage. These VNC images and reference segmentations were used to train 3D CNNsfor automatic segmentation in either VNC images or NCCT images. Automaticsegmentations in VNC images showed good agreement with reference segmentations,with an average Dice similarity coefficient of 0.897 pm 0.034 and an averagesymmetric surface distance of 1.42 pm 0.45 mm. Volume differences [95 confidence interval] between automatic NCCT and reference CCTA segmentationswere -19 [-67; 30] mL for LV myocardium, -25 [-78; 29] mL for LV cavity, -29[-73; 14] mL for right ventricle, -20 [-62; 21] mL for left atrium, and -19[-73; 34] mL for right atrium, respectively. In 214 (74 ) NCCT images from anindependent multi-vendor multi-center set, two observers agreed that theautomatic segmentation was mostly accurate or better. This method might enablequantification of additional cardiac measures from NCCT images for improvedcardiovascular risk prediction.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×