Dataset Information

Improved Quantification of Myocardium Scar in Late Gadolinium Enhancement Images: Deep Learning Based Image Fusion Approach.

ABSTRACT:

Background

Quantification of myocardium scarring in late gadolinium enhanced (LGE) cardiac magnetic resonance imaging can be challenging due to low scar-to-background contrast and low image quality. To resolve ambiguous LGE regions, experienced readers often use conventional cine sequences to accurately identify the myocardium borders.

Purpose

To develop a deep learning model for combining LGE and cine images to improve the robustness and accuracy of LGE scar quantification.

Study type

Retrospective.

Population

A total of 191 hypertrophic cardiomyopathy patients: 1) 162 patients from two sites randomly split into training (50%; 81 patients), validation (25%, 40 patients), and testing (25%; 41 patients); and 2) an external testing dataset (29 patients) from a third site.

Field strength/sequence

1.5T, inversion-recovery segmented gradient-echo LGE and balanced steady-state free-precession cine sequences ASSESSMENT: Two convolutional neural networks (CNN) were trained for myocardium and scar segmentation, one with and one without LGE-Cine fusion. For CNN with fusion, the input was two aligned LGE and cine images at matched cardiac phase and anatomical location. For CNN without fusion, only LGE images were used as input. Manual segmentation of the datasets was used as reference standard.

Statistical tests

Manual and CNN-based quantifications of LGE scar burden and of myocardial volume were assessed using Pearson linear correlation coefficients (r) and Bland-Altman analysis.

Results

Both CNN models showed strong agreement with manual quantification of LGE scar burden and myocardium volume. CNN with LGE-Cine fusion was more robust than CNN without LGE-Cine fusion, allowing for successful segmentation of significantly more slices (603 [95%] vs. 562 (89%) of 635 slices; P < 0.001). Also, CNN with LGE-Cine fusion showed better agreement with manual quantification of LGE scar burden than CNN without LGE-Cine fusion (%Scar_LGE-cine = 0.82 × %Scar_manual , r = 0.84 vs. %Scar_LGE = 0.47 × %Scar_manual , r = 0.81) and myocardium volume (Volume_LGE-cine = 1.03 × Volume_manual , r = 0.96 vs. Volume_LGE = 0.91 × Volume_manual , r = 0.91).

Data conclusion

CNN based LGE-Cine fusion can improve the robustness and accuracy of automated scar quantification.

Level of evidence

3 TECHNICAL EFFICACY: 1.

SUBMITTER: Fahmy AS

PROVIDER: S-EPMC8359184 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesThe aim of this study was to assess the effect of a deep learning (DL)-based reconstruction algorithm on late gadolinium enhancement (LGE) image quality and to evaluate its influence on scar quantification.MethodsSixty patients (46 ± 17 years, 50% male) with suspected or known cardiomyopathy underwent CMR. Short-axis LGE images were reconstructed using the conventional reconstruction and a DL network (DLRecon) with tunable noise reduction (NR) levels from 0 to 100%. Image quality of standard LGE images and DLRecon images with 75% NR was scored using a 5-point scale (poor to excellent). In 30 patients with LGE, scar size was quantified using thresholding techniques with different standard deviations (SD) above remote myocardium, and using full width at half maximum (FWHM) technique in images with varying NR levels.ResultsDLRecon images were of higher quality than standard LGE images (subjective quality score 3.3 ± 0.5 vs. 3.6 ± 0.7, p < 0.001). Scar size increased with increasing NR levels using the SD methods. With 100% NR level, scar size increased 36%, 87%, and 138% using 2SD, 4SD, and 6SD quantification method, respectively, compared to standard LGE images (all p values < 0.001). However, with the FWHM method, no differences in scar size were found (p = 0.06).ConclusionsLGE image quality improved significantly using a DL-based reconstruction algorithm. However, this algorithm has an important impact on scar quantification depending on which quantification technique is used. The FWHM method is preferred because of its independency of NR. Clinicians should be aware of this impact on scar quantification, as DL-based reconstruction algorithms are being used.Key points• The image quality based on (subjective) visual assessment and image sharpness of late gadolinium enhancement images improved significantly using a deep learning-based reconstruction algorithm that aims to reconstruct high signal-to-noise images using a denoising technique. • Special care should be taken when scar size is quantified using thresholding techniques with different standard deviations above remote myocardium because of the large impact of these advanced image enhancement algorithms. • The full width at half maximum method is recommended to quantify scar size when deep learning algorithms based on noise reduction are used, as this method is the least sensitive to the level of noise and showed the best agreement with visual late gadolinium enhancement assessment.

Project description:BackgroundAutomatic myocardial scar segmentation from late gadolinium enhancement (LGE) images using neural networks promises an alternative to time-consuming and observer-dependent semi-automatic approaches. However, alterations in data acquisition, reconstruction as well as post-processing may compromise network performance. The objective of the present work was to systematically assess network performance degradation due to a mismatch of point-spread function between training and testing data.MethodsThirty-six high-resolution (0.7×0.7×2.0 mm3) LGE k-space datasets were acquired post-mortem in porcine models of myocardial infarction. The in-plane point-spread function and hence in-plane resolution Δx was retrospectively degraded using k-space lowpass filtering, while field-of-view and matrix size were kept constant. Manual segmentation of the left ventricle (LV) and healthy remote myocardium was performed to quantify location and area (% of myocardium) of scar by thresholding (≥ SD5 above remote). Three standard U-Nets were trained on training resolutions Δxtrain = 0.7, 1.2 and 1.7 mm to predict endo- and epicardial borders of LV myocardium and scar. The scar prediction of the three networks for varying test resolutions (Δxtest = 0.7 to 1.7 mm) was compared against the reference SD5 thresholding at 0.7 mm. Finally, a fourth network trained on a combination of resolutions (Δxtrain = 0.7 to 1.7 mm) was tested.ResultsThe prediction of relative scar areas showed the highest precision when the resolution of the test data was identical to or close to the resolution used during training. The median fractional scar errors and precisions (IQR) from networks trained and tested on the same resolution were 0.0 percentage points (p.p.) (1.24 - 1.45), and - 0.5 - 0.0 p.p. (2.00 - 3.25) for networks trained and tested on the most differing resolutions, respectively. Deploying the network trained on multiple resolutions resulted in reduced resolution dependency with median scar errors and IQRs of 0.0 p.p. (1.24 - 1.69) for all investigated test resolutions.ConclusionA mismatch of the imaging point-spread function between training and test data can lead to degradation of scar segmentation when using current U-Net architectures as demonstrated on LGE porcine myocardial infarction data. Training networks on multi-resolution data can alleviate the resolution dependency.

Project description:BackgroundCardiovascular magnetic resonance (CMR) imaging may be used to visualize post-ablation atrial scar (PAAS), and three-dimensional late gadolinium enhancement (3D LGE) is the most widely employed technique for imaging of chronic scar. Detection of PAAS provides a unique non-invasive insight into the effects of the ablation and may help guide further ablation procedures. However, there is evidence that PAAS is often not detected by CMR, implying a significant sensitivity problem, and imaging parameters vary between leading centres. Therefore, there is a need to establish the optimal imaging parameters to detect PAAS.MethodsForty subjects undergoing their first pulmonary vein isolation procedure for AF had detailed CMR assessment of atrial scar: one scan pre-ablation, and two scans post-ablation at 3 months (separated by 48 h). Each scan session included ECG- and respiratory-navigated 3D LGE acquisition at 10, 20 and 30 min post injection of a gadolinium-based contrast agent (GBCA). The first post-procedural scan was performed on a 1.5 T scanner with standard acquisition parameters, including double dose (0.2 mmol/kg) Gadovist and 4 mm slice thickness. Ten patients subsequently underwent identical scan as controls, and the other 30 underwent imaging with a reduced, single, dose GBCA (n = 10), half slice thickness (n = 10) or on a 3 T scanner (n = 10). Apparent signal-to-noise (aSNR), contrast-to-noise (aCNR) and imaging quality (Likert Scale, 3 independent observers) were assessed. PAAS location and area (%PAAS scar) were assessed following manual segmentation. Atrial shells with standardised %PAAS at each timepoint were then compared to ablation lesion locations to assess quality of scar delineation.ResultsA total of 271 3D acquisitions (out of maximum 280, 96.7%) were acquired. Likert scale of imaging quality had high interobserver and intraobserver intraclass correlation coefficients (0.89 and 0.96 respectively), and showed lower overall imaging quality on 3 T and at half-slice thickness. aCNR, and quality of scar delineation increased significantly with time. aCNR was higher with reduced, single, dose of GBCA (p = 0.005).Conclusion3D LGE CMR atrial scar imaging, as assessed qualitatively and quantitatively, improves with time from GBCA administration, with some indices continuing to improve from 20 to 30 min. Imaging should be performed at least 20 min post-GBCA injection, and a single dose of contrast should be considered.Trial registrationTrial registry- United Kingdom National Research Ethics Service 08/H0802/68 - 30th September 2008.

Project description:BackgroundCardiovascular magnetic resonance (CMR) imaging has been used to visualise post-ablation atrial scar (PAAS), generally employing a three-dimensional (3D) late gadolinium enhancement (LGE) technique. However the reproducibility of PAAS imaging has not been determined. This cross-over study is the first to investigate the reproducibility of the technique, crucial for both future research design and clinical implementation.MethodsForty subjects undergoing first time ablation for atrial fibrillation (AF) had detailed CMR assessment of PAAS. Following baseline pre-ablation scan, two scans (separated by 48 h) were performed at three months post-ablation. Each scan session included 3D LGE acquisition at 10, 20 and 30 min post administration of gadolinium-based contrast agent (GBCA). Subjects were allocated at second scan post-ablation to identical imaging parameters ('Repro', n = 10), 3 T scanner ('3 T', n = 10), half-slice thickness ('Half-slice', n = 10) or half GBCA dose ('Half-gad', n = 10). PAAS was compared to baseline scar and then reproducibility was assessed for two measures of thresholded scar (% left atrial (LA) occupied by PAAS (%LA PAAS) and Pulmonary Vein Encirclement (PVE)), and then four measures of non-thresholded scar (point-by-point assessment of PAAS, four normalisation methods). Thresholded measures of PAAS were evaluated against procedural outcome (AF recurrence).ResultsA total of 271 3D acquisitions (out of maximum 280, 96.7%) were acquired. At 20 and 30 min, inter-scan reproducibility was good to excellent (coefficient of variation at 20 min and 30 min: %LA PAAS 0.41 and 0.20; PVE 0.13 and 0.04 respectively for 'Repro' group). Changes in imaging parameters, especially reduced GBCA dose, reduced inter-scan reproducibility, but for most measures remained good to excellent (ICC for %LA PAAS 0.454-0.825, PVE 0.618-0.809 at 30 min). For non-thresholded scar, highest reproducibility was observed using blood pool z-score normalisation technique: inter-scan ICC 0.759 (absolute agreement, 'Repro' group). There was no significant relationship between indices of PAAS and AF recurrence.ConclusionPAAS imaging is a reproducible finding. Imaging should be performed at least 20 min post-GBCA injection, and a blood pool z-score should be considered for normalisation of signal intensities. The clinical implications of these findings remain to be established in the absence of a simple correlation with arrhythmia outcome.Trial registrationUnited Kingdom National Research Ethics Service 08/H0802/68 - 30th September 2008.