Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 5, 2025Open Access

Cardiac Function Assessment with Deep-Learning-Based Automatic Segmentation of Free-Running 4D Whole-Heart CMR

View Full Paper
Ask AI
Bookmark
Share

Authors

AOAugustin C. OgierSBSalomé BaupGIGorun Ilanjian

Discussion

Loading...

Member takes

Overview

This approach enables rapid segmentation of cardiac MRI images, improving clinical metrics and suggesting broader adoption in practice.

Key Points

  • The deep learning model achieved automatic segmentation of whole-heart cardiac MRI images within a minute.
  • High geometric accuracy was confirmed with a Dice similarity coefficient (DSC) reaching 0.94 for the left ventricular blood pool.
  • Validation showed strong agreement in clinical metrics with an intraclass correlation coefficient (ICC) greater than 0.98 for left ventricular measurements.
  • Training on all cardiac phases enhanced performance, reducing volume mismatch significantly from 4.0% to 2.6%.

Cite This Study

Ogier et al. (2025) studied this question.

synapsesocial.com/papers/689a0f93e6551bb0af8d134chttps://doi.org/10.1101/2025.07.15.25331281
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Enhancing Cardiac Functional Assessment with Deep-Learning-based Automatic Segmentation of Free-Running 4D Whole-Heart MR Images2025
  2. 2Filling the Gaps: Generating 4D Dense Cardiac Anatomy from Sparse CMR for Enhanced Tetralogy of Fallot Assessment2026
  3. 3A Deep Learning‐Based Fully Automated Cardiac MRI Segmentation Approach for Tetralogy of Fallot Patients2025 · 2 citations
  4. 4Automated whole-heart volumetrics and haemodynamics from 4D flow CMR magnitude images: development and validation of a deep learning model2026
  5. 5Automated Deep Learning Cine MRI Segmentation for Cardiac Function Assessment in Preclinical Models2025