Back to all papers

Deep learning models for segmentation and quantification of left atrial appendage volume using noncontrast cardiac computed tomography.

November 1, 2025pubmed logopapers

Authors

Santos DAM,de Oliveira Teixeira L,Massago M,da Alvarez Silva S,Gurgel SJT,Rochitte CE,da Costa YMEG,de Andrade L

Affiliations (7)

  • Postgraduate Program in Health Sciences, State University of Maringa, Maringa, Brazil.
  • Postgraduate Program in Computer Science, State University of Maringa, Maringa, Brazil.
  • Department of Medicine, State University of Maringa, Maringa, Brazil.
  • Institute of Heart, State University of São Paulo, São Paulo, Brazil.
  • Department of Informatics, State University of Maringa, Maringa, Brazil.
  • Postgraduate Program in Health Sciences, State University of Maringa, Maringa, Brazil. [email protected].
  • Department of Medicine, State University of Maringa, Maringa, Brazil. [email protected].

Abstract

The left atrial appendage (LAA) is a critical but frequently overlooked site of thrombus formation, reinforcing the need for accurate identification in routine cardiac imaging. This process is related to pathological dilation associated with endothelial injury and a proinflammatory status. This study assesses the performance of deep learning architectures based on U-Net, specifically UNet3D, Residual-UNet3D, 3D Attention-UNet, and Res16-PAC-UNet, in the semiautomated segmentation and volume measurement of LAA. We retrospectively analyzed noncontrast cardiac computed tomography (NCCT) scans from 452 patients aged ≥ 60 years, acquired for chest pain evaluation, to compare the performance of four U-Net-based deep learning architectures (UNet3D, Residual-UNet3D, 3D Attention-UNet, and Res16-PAC-UNet) for semiautomated LAA segmentation and volume measurement. Segmentation accuracy was assessed with the Dice coefficient, and volumetric agreement with Pearson correlation and Bland-Altman analysis. Dice coefficients were 78.44 ± 1.93 for UNet3D, 78.97 ± 0.79 for Residual-UNet3D, 79.07 ± 1.43 for 3D Attention-UNet, and 77.68 ± 1.47 for Res16-PAC-UNet. All models showed strong correlations between predicted and manual volumes (P < 0.001), with the highest in 3D Attention-UNet (r = 0.800). Bland-Altman analysis indicated minimal bias and narrow limits of agreement for all architectures, confirming consistent reliability. Deep learning-based segmentation on NCCT enables accurate, reproducible LAA morphological and volumetric assessment without contrast, offering a rapid and reliable tool to support cardiovascular risk stratification and treatment planning.

Topics

Journal Article

Ready to Sharpen Your Edge?

Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.