Comparison of the within-reader and inter-vendor agreement of left ventricular circumferential strains and volume indices derived from cardiovascular magnetic resonance imaging

Mansell, Doyin S.; Frank, Evelyn; Kelly, Nathaniel; Hernandez, Bruno Agostinho; Fletcher, James; Bruno, Vito Domenico; Sammut, Eva; Chiribiri, Amedeo; Johnson, Tom; Ascione, Raimondo; Bartlett, Jonathan; Gill, Harinderjit; Fraser, Katharine; Cookson, Andrew

doi:10.1371/journal.pone.0242908

Cited by 7 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An example is shown in Figure 3A . Two smoothing approaches were implemented and evaluated, First, the concave areas created by the papillary muscles were “filled in” by using the binary “close” function with a disk of 10 pixels and then fitting a convex hull to the perimeter of the endocardial bloodpool for each frame ( 35 ). An example result of this approach is shown in Figure 3 .…”

Section: Methodsmentioning

confidence: 99%

Deep learning automates detection of wall motion abnormalities via measurement of longitudinal strain from ECG-gated CT images

Chen

Kahn

et al. 2022

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

Introduction4D cardiac CT (cineCT) is increasingly used to evaluate cardiac dynamics. While echocardiography and CMR have demonstrated the utility of longitudinal strain (LS) measures, measuring LS from cineCT currently requires reformatting the 4D dataset into long-axis imaging planes and delineating the endocardial boundary across time. In this work, we demonstrate the ability of a recently published deep learning framework to automatically and accurately measure LS for detection of wall motion abnormalities (WMA).MethodsOne hundred clinical cineCT studies were evaluated by three experienced cardiac CT readers to identify whether each AHA segment had a WMA. Fifty cases were used for method development and an independent group of 50 were used for testing. A previously developed convolutional neural network was used to automatically segment the LV bloodpool and to define the 2, 3, and 4 CH long-axis imaging planes. LS was measured as the perimeter of the bloodpool for each long-axis plane. Two smoothing approaches were developed to avoid artifacts due to papillary muscle insertion and texture of the endocardial surface. The impact of the smoothing was evaluated by comparison of LS estimates to LV ejection fraction and the fractional area change of the corresponding view.ResultsThe automated, DL approach successfully analyzed 48/50 patients in the training cohort and 47/50 in the testing cohort. The optimal LS cutoff for identification of WMA was −21.8, −15.4, and −16.6% for the 2-, 3-, and 4-CH views in the training cohort. This led to correct labeling of 85, 85, and 83% of 2-, 3-, and 4-CH views, respectively, in the testing cohort. Per-study accuracy was 83% (84% sensitivity and 82% specificity). Smoothing significantly improved agreement between LS and fractional area change (R2: 2 CH = 0.38 vs. 0.89 vs. 0.92).ConclusionAutomated LV blood pool segmentation and long-axis plane delineation via deep learning enables automatic LS assessment. LS values accurately identify regional wall motion abnormalities and may be used to complement standard visual assessments.

show abstract

Section: Methodsmentioning

confidence: 99%

Deep learning automates detection of wall motion abnormalities via measurement of longitudinal strain from ECG-gated CT images

Chen

Kahn

et al. 2022

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

show abstract

“…The strain-based metric used in this study was formulated to be robust and reproducible across sites/users/software, use a transparent, non-proprietary algorithm, and be sufficiently sensitive to characterise local ventricular function in a layer-wise manner, as recently reported by our group 21 . The methodology is summarised on the LV schematic in Fig.…”

Section: Methodsmentioning

confidence: 99%

Acute regional changes in myocardial strain may predict ventricular remodelling after myocardial infarction in a large animal model

Mansell

Bruno

Sammut

et al. 2021

Sci Rep

View full text Add to dashboard Cite

To identify predictors of left ventricular remodelling (LVR) post-myocardial infarction (MI) and related molecular signatures, a porcine model of closed-chest balloon MI was used along with serial cardiac magnetic resonance imaging (CMRI) up to 5–6 weeks post-MI. Changes in myocardial strain and strain rates were derived from CMRI data. Tissue proteomics was compared between infarcted and non-infarcted territories. Peak values of left ventricular (LV) apical circumferential strain (ACS) changed over time together with peak global circumferential strain (GCS) while peak GLS epicardial strains or strain rates did not change over time. Early LVR post-MI enhanced abundance of 39 proteins in infarcted LV territories, 21 of which correlated with LV equatorial circumferential strain rate. The strongest associations were observed for D-3-phosphoglycerate dehydrogenase (D-3PGDH), cysteine and glycine-rich protein-2, and secreted frizzled-related protein 1 (sFRP1). This study shows that early changes in regional peak ACS persist at 5–6 weeks post-MI, when early LVR is observed along with increased tissue levels of D-3PGDH and sFRP1. More studies are needed to ascertain if the observed increase in tissue levels of D-3PGDH and sFRP1 might be casually involved in the pathogenesis of adverse LV remodelling.

show abstract

“…The novel method for calculating strains proposed by Mansell et al (Mansell et al, 2020;Mansell et al, 2021) was used here. Endocardial circumferential strains, n , were calculated from the perimeter length data extracted from the segmentations produced by CVI42, using the definition in Equation 1:…”

Section: Deformation Analysismentioning

confidence: 99%

“…Global circumferential strain (GCS) was also calculated as the mean of the circumferential strains from all slices (Mansell et al, 2020).…”

Section: Deformation Analysismentioning

confidence: 99%

“…In previous works (Mansell et al, 2020;Mansell et al, 2021) a localised, layer-specific, circumferential LV strain metric was developed, using porcine imaging data. The reproducibility of this strain metric was demonstrated across different users and software packages and compared favourably to that of standard clinical metrics such as global strains and ejection fraction (Mansell et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MRI‐based strain measurements reflect morphological changes following myocardial infarction: A study on the UK Biobank cohort

et al. 2022

View full text Add to dashboard Cite

In a porcine experimental model of myocardial infarction, a localised, layer‐specific, circumferential left ventricular strain metric has been shown to indicate chronic changes in ventricular function post‐infarction more strongly than ejection fraction. This novel strain metric might therefore provide useful prognostic information clinically. In this study, existing clinical volume indices, global strains, and the novel, layer‐specific strain were calculated for a large human cohort to assess variations in ventricular function and morphology with age, sex, and health status. Imaging and health data from the UK Biobank were obtained, including healthy volunteers and those with a history of cardiovascular illness. In total, 710 individuals were analysed and stratified by age, sex and health. Significant differences in all strain metrics were found between healthy and unhealthy populations, as well as between males and females. Significant differences in basal circumferential strain and global circumferential strain were found between healthy males and females, with males having smaller absolute values for both (all p≤ 0.001). There were significant differences in the functional variables left ventricular ejection fraction, end‐systolic volume, end‐systolic volume index and mid‐ventricular circumferential strain between healthy and unhealthy male cohorts aged 65–74 (all p≤ 0.001). These results suggest that whilst regional circumferential strains may be useful clinically for assessing cardiovascular health, care must be taken to ensure critical values are indexed correctly to age and sex, due to the differences in these values observed here.

show abstract

Comparison of the within-reader and inter-vendor agreement of left ventricular circumferential strains and volume indices derived from cardiovascular magnetic resonance imaging

Cited by 7 publications

References 24 publications

Deep learning automates detection of wall motion abnormalities via measurement of longitudinal strain from ECG-gated CT images

Deep learning automates detection of wall motion abnormalities via measurement of longitudinal strain from ECG-gated CT images

Acute regional changes in myocardial strain may predict ventricular remodelling after myocardial infarction in a large animal model

MRI‐based strain measurements reflect morphological changes following myocardial infarction: A study on the UK Biobank cohort

Contact Info

Product

Resources

About