Dataset Information

Open-Source Routines for Building Personalized Left Ventricular Models from Cardiac MRI Data.

ABSTRACT: Creating patient-specific models of the heart is a promising approach for predicting outcomes in response to congenital malformations, injury, or disease, as well as an important tool for developing and customizing therapies. However, integrating multi-modal imaging data to construct patient-specific models is a non-trivial task. Here, we propose an approach that employs a prolate spheroidal coordinate system to interpolate information from multiple imaging datasets and map those data onto a single geometric model of the left ventricle. We demonstrate the mapping of the location and transmural extent of post-infarction scar segmented from late gadolinium enhancement (LGE) MRI, as well as mechanical activation calculated from displacement encoding with stimulated echoes (DENSE) MRI. As a supplement to this article, we provide MATLAB and Python versions of the routines employed here for download from SimTK (https://simtk.org/projects/lvdatamap).

SUBMITTER: Phung TN

PROVIDER: S-EPMC7104752 | biostudies-literature | 2019 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Open-Source Routines for Building Personalized Left Ventricular Models From Cardiac Magnetic Resonance Imaging Data.

Phung Thien-Khoi N TN Waters Christopher D CD Holmes Jeffrey W JW

Journal of biomechanical engineering 20200201 2

Creating patient-specific models of the heart is a promising approach for predicting outcomes in response to congenital malformations, injury, or disease, as well as an important tool for developing and customizing therapies. However, integrating multimodal imaging data to construct patient-specific models is a nontrivial task. Here, we propose an approach that employs a prolate spheroidal coordinate system to interpolate information from multiple imaging datasets and map those data onto a singl ...[more]

PMID: 31141592

Similar Datasets

Project description:BackgroundCardiac magnetic resonance (CMR) has been used to diagnose and risk-stratify patients with left ventricular noncompaction (LVNC). The prognostic value of CMR parameters for LVNC, especially feature tracking (CMR-FT), is not well known in LVNC patients with left ventricular dysfunction. The present study aimed to investigate whether the combination of CMR-FT with traditional CMR parameters can increase the prognostic value of CMR for LVNC patients with reduced left ventricular ejection fraction (LVEF).MethodsA total of 123 candidates were retrospectively included in this multicenter study and 55 LVNC patients (mean age, 45.7 ± 16.2 years; 61.8% men) remained after applying the exclusion criteria. Clinical features, left ventricular (LV) function parameters, global and segment myocardial strain, and late gadolinium enhancement (LGE) were evaluated. The outcomes include the composite events of cardiovascular death, heart transplantation, hospitalization for heart failure, thromboembolic events, and ventricular arrhythmias.ResultsAfter a median follow-up of 5.17 years (interquartile range: 0.17 to 10.58 years), 24 (36.8%) patients experienced at least one major adverse cardiovascular event (MACE). The myocardial strain parameters of patients with events were lower than those of patients without events. In the univariable Cox analysis, LVEF, the presence of LGE, global longitudinal strain (GLS) and segmental strains, including longitudinal strain at the apical level and radial and circumferential strain at the basal level, were significantly associated with MACEs. In the multivariate analysis, LGE (hazard ratio (HR) 3.452, 95% CI 1.133 to 10.518, p = 0.029) was a strong predictor of MACEs and significantly improved the predictive value (chi-square of the model after adding LGE: 7.51 vs. 13.47, p = 0.009). However, myocardial strain parameters were not statistically significant for the prediction of MACEs after adjusting for age, body mass index, LVEF and the presence of LGE and did not increase the prognostic value (chi-square of the model after adding GLS: 13.47 vs. 14.14, p = 0.411) in the multivariate model.ConclusionsThe combination of CMR-FT with traditional CMR parameters may not increase the prognostic value of CMR in LVNC patients with reduced LVEF, while the presence of LGE was a strong independent predictor of MACEs and significantly improved the predictive value.

Project description:Background Cardiac resynchronization therapy (CRT) restores ventricular synchrony and induces left ventricular (LV) reverse remodeling in patients with heart failure (HF) and dyssynchrony. However, 30% of treated patients are non-responders despite all efforts. Cardiac magnetic resonance imaging (CMR) can be used to quantify regional contributions to stroke volume (SV) as potential CRT predictors. The aim of this study was to determine if LV longitudinal (SVlong%), lateral (SVlat%), and septal (SVsept%) contributions to SV differ from healthy controls and investigate if these parameters can predict CRT response. Methods Sixty-five patients (19 women, 67 ± 9 years) with symptomatic HF (LVEF ≤ 35%) and broadened QRS (≥ 120 ms) underwent CMR. SVlong% was calculated as the volume encompassed by the atrioventricular plane displacement (AVPD) from end diastole (ED) to end systole (ES) divided by total SV. SVlat%, and SVsept% were calculated as the volume encompassed by radial contraction from ED to ES. Twenty age- and sex-matched healthy volunteers were used as controls. The regional measures were compared to outcome response defined as ≥ 15% decrease in echocardiographic LV end-systolic volume (LVESV) from pre- to 6-months post CRT (delta, Δ). Results AVPD and SVlong% were lower in patients compared to controls (8.3 ± 3.2 mm vs 15.3 ± 1.6 mm, P < 0.001; and 53 ± 18% vs 64 ± 8%, P < 0.01). SVsept% was lower (0 ± 15% vs 10 ± 4%, P < 0.01) with a higher SVlat% in the patient group (42 ± 16% vs 29 ± 7%, P < 0.01). There were no differences between responders and non-responders in neither SVlong% (P = 0.87), SVlat% (P = 0.09), nor SVsept% (P = 0.65). In patients with septal net motion towards the right ventricle (n = 28) ΔLVESV was − 18 ± 22% and with septal net motion towards the LV (n = 37) ΔLVESV was − 19 ± 23% (P = 0.96). Conclusions Longitudinal function, expressed as AVPD and longitudinal contribution to SV, is decreased in patients with HF scheduled for CRT. A larger lateral contribution to SV compensates for the abnormal septal systolic net movement. However, LV reverse remodeling could not be predicted by these regional contributors to SV.

Dataset Information

Open-Source Routines for Building Personalized Left Ventricular Models from Cardiac MRI Data.

Publications

Open-Source Routines for Building Personalized Left Ventricular Models From Cardiac Magnetic Resonance Imaging Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets