Cardiac MRI late gadolinium enhancement (LGE) enables non-invasive identification of left atrial (LA) scar, whose spatial distribution is strongly associated with atrial fibrillation (AF) severity and recurrence. However, automatic LA scar segmentation remains challenging due to low contrast, annotation variability, and the lack of anatomical constraints, often leading to non-reliable predictions. Accordingly, our aim was to propose a progressive learning strategy to segment LA scar from LGE images inspired from a clinical workflow. A 3-stage framework based on SwinUNETR was implemented, comprising: 1) a first LA cavity pre-learning model, 2) dual-task model which further learns spatial relationship between LA geometry and scar patterns, and 3) fine-tuning on precise segmentation of the scar. Furthermore, we introduced an anatomy-aware spatially weighted loss that incorporates prior clinical knowledge by constraining scar predictions to anatomically plausible LA wall regions while mitigating annotation bias. Our preliminary results obtained on validation LGE volumes from LASCARQS public dataset after 5-fold cross validation, LA segmentation had Dice score of 0.94, LA scar segmentation achieved Dice score of 0.50, Hausdorff Distance of 11.84 mm, Average Surface Distance of 1.80 mm, outperforming only a one-stage scar segmentation with 0.49, 13.02 mm, 1.96 mm, repectively. By explicitly embedding clinical anatomical priors and diagnostic reasoning into deep learning, the proposed approach improved the accuracy and reliability of LA scar segmentation from LGE, revealing the importance of clinically informed model design.
Figures & tables
Operation
Degree
P
Rationale
Tbias
Bias Field
[0.02, 0.06]
0.15
Magnetic field inhomogeneity
Telastic
3D Elastic
σ : [3, 5]; Δ∗
0.10
Anatomical deformation
Trot
Rotation
[-15 ∘ , 15 ∘ ]
0.50
Rigid transformation
Tzoom
Zoom
[0.9, 1.1]
0.30
Rigid transformation
Tblur
Gaussian Blurring
[0.5, 1.5]
0.20
Reconstruction effect
Tcontrast
Contrast
[0.8, 1.3]
0.30
Reconstruction effect
Table 1: Data augmentation operations along with rationale. Each operation has a probability (P) and range of variation.
Figure 1: Progressive learning for LA scar segmentation from LGE MRI. In Stage I, the model learns to identify the easy LA structure; in Stage II, the model inherited from Stage I, and learns to identify LA and scar at the same time with the use of two decoders as well as LA wall constraint for scar; in Stage III, the model inherited from Stage II, and keeps fine-tuning scar segmentation.
Figure 2: Examples of cropped axial images from LGE MRI volumes along with LA cavity (green) as well as scar (red) expert annotations from LASCARQS public dataset.
Figure 3: Examples of LA cavity and scar segmentation from LASCARQS public dataset. From left to right columns: input LGE MRI images, LA ground truth (GT), LA predicitons from Stage I+II. LA scar GT, scar predictions from Stage I+II and Stage I+II+III respectively. The Dice scores are indicated in red.
LA cavity segmentation ( n=12 )
Stage
Dsc (%)
HD (mm)
ASD (mm)
I
0.94±0.02
3.05±0.96
0.78±0.32
I+II
0.90±0.01
8.30±2.44
2.28±1.13
I+II+III
–
–
–
LA scar segmentation ( n=12 )
Stage
Dsc (%)
HD (mm)
ASD (mm)
Table 2: Performance for LA and scar segmentation on validation set in terms of Dice score (Dsc) ↑ , Hausdorff distance (HD) ↓ and Average Surface Distance (ASD) ↓ .
Baseline
Stage
Dsc (%) ↑
HD (mm) ↓
ASD (mm) ↓
nnUNET
–
0.50±0.09
16.10±24.33
1.51±3.63
B1
III
0.49±0.02
14.16±4.26
2.12±0.76
B2
I+III
0.50±0.02
11.67±2.48
1.72±0.34
B3 ∗
II
0.45±0.02
14.49±9.86
2.98±1.71
B4 ∗
II+III
0.49±0.01
11.25±0.80
1.84±0.81
Full
I+II+III
0.50±0.02
11.84±3.11
1.80±0.55
Table 3: Scar segmentation performances on validation set according to different baselines (B), in terms of Dsc, HD and ASD.
Identification and quantification of myocardial scar is important for diagnosis and prognosis of cardiovascular diseases. However, reliable scar segmentation from Late Gadolinium Enhancement Cardiac Magnetic Resonance (LGE-CMR) images remains a challenge due to variations in contrast enhancement across patients, suboptimal imaging conditions such as post contrast washout, and inconsistencies in ground truth annotations on diffuse scars caused by inter observer variability. In this work, we propose a curriculum learning-based framework designed to improve segmentation performance under these challenging conditions. The method introduces a progressive training strategy that guides the model from high-confidence, clearly defined scar regions to low confidence or visually ambiguous samples with limited scar burden. By structuring the learning process in this manner, the network develops robustness to uncertain labels and subtle scar appearances that are often underrepresented in conventional training pipelines. Experimental results show that the proposed approach enhances segmentation accuracy and consistency, particularly for cases with minimal or diffuse scar, outperforming standard training baselines. This strategy provides a principled way to leverage imperfect data for improved myocardial scar quantification in clinical applications. Our code is publicly available on GitHub.
Nivetha Jayakumar, Jonathan Pan, Shuo Wang +6
Department of Electrical and Computer Engineering, University of Virginia, USA · Division of Cardiovascular Medicine, University of Virginia Health System, USA · Department of Computer Science, University of Virginia, USA
Myocardial scar segmentation from single-stack late gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) imaging has been a longstanding and clinically important challenge, particularly in the presence of low tissue contrast, diffuse, and small scar regions. These challenges are further intensified by the limited availability of 3D spatial context. This paper presents CalcSeg, a Confidence-aware latent context curriculum learning framework that leverages fused 3D feature representations from single-stack 2D LGE-CMR images for robust scar segmentation. Specifically, we introduce a dynamic semi-supervised curriculum learning strategy that progressively expands training from easier to more challenging scar cases using a learned confidence-aware scoring function. Such a function integrates errors in the predicted scar maps with quantified epistemic uncertainty and scar burden estimation to automatically assess sample difficulty without requiring manual labels. To compensate for the limited spatial context in single-stack acquisitions, we then develop a latent slice-wise self-attention to capture inter-slice dependencies and infer 3D spatial representations from sparse 2D inputs. We evaluate CalcSeg on multi-center clinical LGE-CMR datasets and benchmark against existing scar segmentation networks. Experimental results show that CalcSeg consistently outperforms all competing methods, particularly with substantial improvements on clinically challenging cases. Our code is released on Github.
Nivetha Jayakumar, Hannah Kim, Amit R. Patel +1
School of Engineering and Applied Science, University of Virginia, USA · School of Medicine, University of Virginia, USA
Late Gadolinium Enhancement (LGE) on cardiac magnetic resonance is a key marker of myocardial scar, but its limited accessibility motivates routine ECG-based screening. We evaluated whether β-variational autoencoder (VAE)-derived ECG representations can discriminate LGE+ from LGE- cardiomyopathic patients in a local cohort of 300 subjects. We compared 32-dimensional features from the foundation ECGx.AI model with those from a shallower β-VAE trained on normal PTB-XL ECGs, evaluating downstream classification and Dynamic Time Warping (DTW)-based reconstruction errors. ECGx.AI reached an area under ROC of 0.686 with Random Forest, while the proposed β-VAE reached 0.577 with sensitivity of 0.775 with Gradient Boosting. Notably, DTW-reconstruction errors significantly differed between classes in 10 out of 12 leads according to Mann-Whitney U test and help in classification, leading to an area under ROC of 0.643 with Logistic Regression, supporting their potential as markers of scar-related ECG alterations.
Shayan Sharifi, Riccardo Treu, Ilaria Gandin +3
Department of Mathematics, Informatics, and Geosciences, University of Trieste, Italy · Centre of Diagnosis and Management of Cardiomyopathies; Azienda Sanitaria Universitaria Giuliano Isontina; University of Trieste, Italy; Member of Ern Guard-Heart · Department of Medical, Surgical and Health Sciences, University of Trieste, Italy