Fast acquisition of left and right ventricular function parameters applying cardiovascular magnetic resonance in clinical routine – validation of a 2-shot compressed sensing cine sequence

Abstract

Objectives. To evaluate if cine sequences accelerated by compressed sensing (CS) are feasible in clinical routine and yield equivalent cardiac morphology in less time. Design. We evaluated 155 consecutive patients with various cardiac diseases scanned during our clinical routine. LV and RV short axis (SAX) cine images were acquired by conventional and prototype 2-shot CS sequences on a 1.5 T CMR. The 2-shot prototype captures the entire heart over a period of 3 beats making the acquisition potentially even faster. Both scans were performed with identical slice parameters and positions. We compared LV and RV morphology with Bland-Altmann plots and weighted the results in relation to pre-defined tolerance intervals. Subjective and objective image quality was evaluated using a 4-point score and adapted standardized criteria. Scan times were evaluated for each sequence. Results. In total, no acquisitions were lost due to non-diagnostic image quality in the subjective image score. Objective image quality analysis showed no statistically significant differences. The scan time of the CS cines was significantly shorter (p < .001) with mean scan times of 178 ± 36 s compared to 313 ± 65 s for the conventional cine. All cardiac function parameters showed excellent correlation (r 0.978-0.996). Both sequences were considered equivalent for the assessment of LV and RV morphology. Conclusions. The 2-shot CS SAX cines can be used in clinical routine to acquire cardiac morphology in less time compared to the conventional method, with no total loss of acquisitions due to nondiagnostic quality.

Multilevel comparison of deep learning models for function quantification in cardiovascular magnetic resonance: On the redundancy of architectural variations

“…Segmentation proved more difficult for the right ventricle than for the left; and was more challenging for basal and apical slices than for midventricular slices. When tested for mean differences in clinical parameters, we found that both U-Net and FCN were within predefined published tolerance limits (17,30) based on intraobserver variability and thus did not show greater deviation from the expert than is acceptable for human readers, whereas the MultiResUNet showed intolerable mean differences for LVEDV, LVESV, and RVESV (Figure 6).…”

“… Equivalence testing for clinical parameters. The 95% confidence intervals of mean errors in quantitative clinical parameters are plotted against the tolerance intervals (blue) as defined by Zange et al ( 30 ) and Gröschel et al ( 17 ) based on intraobserver variability. Equivalence is assumed if the respective CI lies completely within the tolerance range.…”

“…To date, very limited literature has been published on deep learning methods for image segmentation applied to such data. While the sequence used does not significantly affect diagnostic image quality according to objective criteria, the images have been considered to be blurrier and prone to ghosting artifacts ( 17 ), which is why the data presents a particular challenge for image analysis algorithms. Clinical evaluation using the Compressed Sensing sequence was considered equivalent to the standard method in ( 17 ).…”

“…While the sequence used does not significantly affect diagnostic image quality according to objective criteria, the images have been considered to be blurrier and prone to ghosting artifacts ( 17 ), which is why the data presents a particular challenge for image analysis algorithms. Clinical evaluation using the Compressed Sensing sequence was considered equivalent to the standard method in ( 17 ). We would expect similar, arguably slightly improved results on a dataset acquired using a standard sequence.…”

“…The dataset consists of routine clinical magnetic resonance studies of 148 patients randomly split into 119 training cases (1,955 images) and 29 test cases (479 images). Full specification on the dataset is published by Gröschel et al ( 17 ). Seven patients were excluded due to technical limitations.…”