Abstract
Cardiac magnetic resonance (CMR) images play a growing role in the diagnostic imaging of cardiovascular diseases. Full coverage of the left ventricle (LV), from base to apex, is a basic criterion for CMR image quality and necessary for accurate measurement of cardiac volume and functional assessment. Incomplete coverage of the LV is identified through visual inspection, which is time-consuming and usually done retrospectively in the assessment of large imaging cohorts. This paper proposes a novel automatic method for determining LV coverage from CMR images by using Fisher-discriminative three-dimensional (FD3D) convolutional neural networks (CNNs). In contrast to our previous method employing 2D CNNs, this approach utilizes spatial contextual information in CMR volumes, extracts more representative high-level features and enhances the discriminative capacity of the baseline 2D CNN learning framework, thus achieving superior detection accuracy. A two-stage framework is proposed to identify missing basal and apical slices in measurements of CMR volume. First, the FD3D CNN extracts high-level features from the CMR stacks. These image representations are then used to detect the missing basal and apical slices. Compared to the traditional 3D CNN strategy, the proposed FD3D CNN minimizes within-class scatter and maximizes between-class scatter. We performed extensive experiments to validate the proposed method on more than 5,000 independent volumetric CMR scans from the UK Biobank study, achieving low error rates for missing basal/apical slice detection (4.9%/4.6%). The proposed method can also be adopted for assessing LV coverage for other types of CMR image data.
Highlights
L EFT ventricular (LV) cardiac anatomy and function are widely used in the field of cardiac medicine for diagnosis and monitoring disease progression and for assessing the patient’s response to cardiac surgery and interventional procedures
We experiment to characterize the performance of our Fisher-discriminative 3D (FD3D) convolutional neural networks (CNNs) learning framework
Error rates of missing basal slice (MBS)/missing apical slices (MAS) representation learning are shown in Fig. 4, where our proposed method appears to achieve comparable results with less training data compared to the conventional 3D CNN
Summary
L EFT ventricular (LV) cardiac anatomy and function are widely used in the field of cardiac medicine for diagnosis and monitoring disease progression and for assessing the patient’s response to cardiac surgery and interventional procedures. CMR is a single technique that provides access to cardiac anatomy and non-invasive measurements of cardiac function [1]. In large population imaging studies or assessment of patient cohorts from large clinical trials, the quantification of LV anatomy and function requires automatic image quality assessment and tools for image analysis. One basic criterion for cardiac image quality is LV coverage and detection of missing apical and basal CMR slices [2]. Even if scout images are acquired, in order to centre the LV in view and minimize this issue, incomplete coverage may result at any point throughout the cardiac cycle because of changes in patient breathing and cardiac motion. Image quality assessment is traditionally performed by radiographers who ensure that patients do not leave the scanner without providing diagnostically interpretable data.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call