CMRINet: Joint Groupwise Registration and Segmentation for Cardiac Function Quantification from Cine-MRI

2025-05-22

AI Summary

In conventional cardiac functional quantification, left ventricular ejection fraction (LVEF) and myocardial strain are typically computed via separate segmentation and registration pipelines, leading to inconsistent functional assessments, limited accuracy, and low computational efficiency. To address this, we propose the first anatomy-guided, end-to-end deep groupwise registration–segmentation framework that jointly performs intra-sequence groupwise registration and multiphase left ventricular segmentation on cine-MRI data, enabling direct, synergistic inference of LVEF and myocardial strain. Our method integrates anatomical prior embedding with multi-task joint optimization to enhance the reliability of load-independent contractility assessment. Evaluated on 374 four-chamber cine-MRI scans, our approach achieves significantly higher registration accuracy and segmentation Dice scores compared to elastix and two state-of-the-art deep learning methods, while drastically reducing inference time.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate and efficient quantification of cardiac function is essential for the estimation of prognosis of cardiovascular diseases (CVDs). One of the most commonly used metrics for evaluating cardiac pumping performance is left ventricular ejection fraction (LVEF). However, LVEF can be affected by factors such as inter-observer variability and varying pre-load and after-load conditions, which can reduce its reproducibility. Additionally, cardiac dysfunction may not always manifest as alterations in LVEF, such as in heart failure and cardiotoxicity diseases. An alternative measure that can provide a relatively load-independent quantitative assessment of myocardial contractility is myocardial strain and strain rate. By using LVEF in combination with myocardial strain, it is possible to obtain a thorough description of cardiac function. Automated estimation of LVEF and other volumetric measures from cine-MRI sequences can be achieved through segmentation models, while strain calculation requires the estimation of tissue displacement between sequential frames, which can be accomplished using registration models. These tasks are often performed separately, potentially limiting the assessment of cardiac function. To address this issue, in this study we propose an end-to-end deep learning (DL) model that jointly estimates groupwise (GW) registration and segmentation for cardiac cine-MRI images. The proposed anatomically-guided Deep GW network was trained and validated on a large dataset of 4-chamber view cine-MRI image series of 374 subjects. A quantitative comparison with conventional GW registration using elastix and two DL-based methods showed that the proposed model improved performance and substantially reduced computation time.

Problem

• Jointly estimates groupwise registration and segmentation for cardiac cine-MRI
• Improves accuracy and efficiency of cardiac function quantification
• Reduces computation time compared to conventional registration methods

Innovation

• Joint groupwise registration and segmentation model
• End-to-end deep learning for cine-MRI
• Anatomically-guided Deep GW network