Key result
Time-resolved nnU-Net accurately segments left atrium on 4D flow MRI with ~0.89 Dice score.
Why the study?
To design a deep learning pipeline for automated, time-resolved segmentation of the left atrium from 4D flow MRI data.
Does a deep learning-based nnU-Net pipeline accurately and robustly segment the left atrium in 4D flow MRI compared to expert manual annotation in patients with atrial fibrillation and healthy subjects?
Does a deep learning-based nnU-Net pipeline accurately and robustly segment the left atrium in 4D flow MRI compared to expert manual annotation in patients with atrial fibrillation and healthy subjects?
A deep learning-based nnU-Net pipeline provides accurate, automated, time-resolved segmentation of the left atrium from 4D flow MRI, enabling reliable quantification of LA volumes and hemodynamics in both healthy subjects and patients with atrial fibrillation.
Enables automated LA segmentation from 4D flow MRI; leaves open prospective validation before clinical adoption in AF.
PURPOSE: To design a deep learning pipeline for automated, time-resolved segmentation of the left atrium (LA) from 4D flow MRI data. METHODS: We studied 100 individuals including 65 patients with atrial fibrillation (AF) and 35 healthy subjects (HS) resulting in 2530 4D flow MRI time-volumes acquired in different centers and scanners. In AF patients, expert data annotation was initialized on a single frame of highly resolved computed tomography data and then registered (initialization frame), propagated through time and corrected manually on 4D flow MRI, while it was performed manually in HS. Segmentation pipeline was based on a two-stage nnU-Net and was evaluated through segmentation performance metrics, as well as concordance of LA volumes through time and inner LA hemodynamics between prediction and reference segmentations. RESULTS: Testing on 35 AF patients and 15 HS revealed high segmentation performances (overall Dice score = 0.89 ± 0.03 for the initialization frame and 0.86 ± 0.04 for all time frames; AF patient/HS- and vendor-based Dice scores ≥ 0.86 ± 0.03 for the initialization frame and ≥ 0.83 ± 0.04 for all time frames). Strong associations and low Bland-Altman biases were obtained between predicted and reference LA volumes (r ≥ 0.89), velocities (r ≥ 0.96), stasis (r ≥ 0.99), vorticity (r ≥ 0.97), and kinetic energy (r ≥ 0.94). Interestingly, associations remained strong when comparing prediction against 4D flow-independent cine SSFP-derived LA volumes (r ≥ 0.87). CONCLUSION: The proposed time-resolved nnU-Net pipeline robustly and accurately segments the LA from 4D flow MRI images demonstrating excellent segmentation performance, strong agreement with reference volumes and velocity-based indices, as well as reliable generalization across vendors and patient populations.
No takes yet. Share an insight, caveat, or question.
Leite et al. (2026) studied Atrial fibrillation and healthy subjects (n=100). Deep learning-based time-resolved nnU-Net segmentation pipeline vs. Expert manual/semi-automated reference segmentation was evaluated on Overall Dice similarity score for the initialization frame. The time-resolved nnU-Net pipeline accurately segmented the left atrium from 4D flow MRI images, achieving an overall Dice score of 0.89 for the initialization frame and strong agreement with reference volumes.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: