Related Experiment Video
Updated: Jul 4, 2025
Real-Time Cardiac Mapping with a Noninvasive Imageless Electrocardiographic Imaging System
Published on: April 11, 2025
Multi-network approach for image segmentation in non-contrast enhanced cardiac 3D MRI of arrhythmic patients
Ina Vernikouskaya1, Dagmar Bertsche1, Patrick Metze1
1Department of Internal Medicine II, Ulm University Medical Center, Ulm, Germany.
Insights
This study introduces an automated deep learning method for segmenting cardiac structures from non-contrast cardiac magnetic resonance images in patients with arrhythmias, improving stroke prevention strategies.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Cardiology
Background:
- Left atrial appendage (LAA) occlusion is crucial for stroke prevention in nonvalvular atrial fibrillation.
- Catheter-based LAA occlusion is challenging due to anatomical complexity and reliance on TEE and XR guidance.
- Cardiac magnetic resonance (CMR) offers radiation-free imaging but manual segmentation is difficult, especially with arrhythmias.
Purpose of the Study:
- To develop an automated image segmentation method for cardiac structures from non-contrast enhanced CMR images in arrhythmic patients.
- To address the limitations of manual segmentation, including its labor-intensive nature and susceptibility to image quality degradation from arrhythmias.
- To facilitate more efficient and accurate guidance for catheter-based LAA closure procedures.
Main Methods:
- Proposed a multi-stage pipeline approach utilizing fully-convolutional neural networks (CNNs), specifically U-Net architecture.
- Implemented a two-stage deep learning strategy: initial localization of cardiac structures followed by segmentation of cropped sub-regions.
- Trained and validated the model on non-contrast enhanced CMR images from arrhythmic patients.
Main Results:
- The proposed deep learning approach enables automatic segmentation of cardiac structures from challenging CMR datasets.
- The two-stage pipeline design enhances efficiency and effectiveness in automated cardiac segmentation.
- The method shows promise for improving image analysis in arrhythmic patients, overcoming limitations of manual delineation.
Conclusions:
- Automated segmentation of cardiac structures from non-contrast CMR using deep learning is feasible and effective, even in arrhythmic patients.
- This technique can aid in radiation-free cardiac imaging and potentially improve guidance for LAA occlusion procedures.
- Further development of automated segmentation methods is crucial for advancing cardiac imaging analysis and interventional cardiology.
Abstract:
Left atrial appendage (LAA) is the source of thrombi formation in more than 90% of strokes in patients with nonvalvular atrial fibrillation. Catheter-based LAA occlusion is being increasingly applied as a treatment strategy to prevent stroke. Anatomical complexity of LAA makes percutaneous occlusion commonly performed under transesophageal echocardiography (TEE) and X-ray (XR) guidance especially challenging. Image fusion techniques integrating 3D anatomical models derived from pre-procedural imaging into the live XR fluoroscopy can be applied to guide each step of the LAA closure. Cardiac magnetic resonance (CMR) imaging gains in importance for radiation-free evaluation of cardiac morphology as alternative to gold-standard TEE or computed tomography angiography (CTA). Manual delineation of cardiac structures from non-contrast enhanced CMR is, however, labor-intensive, tedious, and challenging due to the rather low contrast. Additionally, arrhythmia often impairs the image quality in ECG synchronized acquisitions causing blurring and motion artifacts. Thus, for cardiac segmentation in arrhythmic patients, there is a strong need for an automated image segmentation method. Deep learning-based methods have shown great promise in medical image analysis achieving superior performance in various imaging modalities and different clinical applications. Fully-convolutional neural networks (CNNs), especially U-Net, have become the method of choice for cardiac segmentation. In this paper, we propose an approach for automatic segmentation of cardiac structures from non-contrast enhanced CMR images of arrhythmic patients based on CNNs implemented in a multi-stage pipeline. Two-stage implementation allows subdividing the task into localization of the relevant cardiac structures and segmentation of these structures from the cropped sub-regions obtained from previous step leading to efficient and effective way of automated cardiac segmentation.
