Estimating extraction fraction and blood flow by combining first-pass myocardial perfusion and T1 mapping results

Physics-informed neural networks for myocardial perfusion MRI quantification

Tracer-kinetic models allow for the quantification of kinetic parameters such as blood flow from dynamic contrast-enhanced magnetic resonance (MR) images. Fitting the observed data with multi-compartment exchange models is desirable, as they are physiologically plausible and resolve directly for blood flow and microvascular function. However, the reliability of model fitting is limited by the low signal-to-noise ratio, temporal resolution, and acquisition length. This may result in inaccurate parameter estimates. This study introduces physics-informed neural networks (PINNs) as a means to perform myocardial perfusion MR quantification, which provides a versatile scheme for the inference of kinetic parameters. These neural networks can be trained to fit the observed perfusion MR data while respecting the underlying physical conservation laws described by a multi-compartment exchange model. Here, we provide a framework for the implementation of PINNs in myocardial perfusion MR. The approach is validated both in silico and in vivo. In the in silico study, an overall decrease in mean-squared error with the ground-truth parameters was observed compared to a standard non-linear least squares fitting approach. The in vivo study demonstrates that the method produces parameter values comparable to those previously found in literature, as well as providing parameter maps which match the clinical diagnosis of patients.

Hierarchical Bayesian myocardial perfusion quantification

Myocardial blood flow can be quantified from dynamic contrast-enhanced magnetic resonance (MR) images through the fitting of tracer-kinetic models to the observed imaging data. The use of multi-compartment exchange models is desirable as they are physiologically motivated and resolve directly for both blood flow and microvascular function. However, the parameter estimates obtained with such models can be unreliable. This is due to the complexity of the models relative to the observed data which is limited by the low signal-to-noise ratio, the temporal resolution, the length of the acquisitions and other complex imaging artefacts. In this work, a Bayesian inference scheme is proposed which allows the reliable estimation of the parameters of the two-compartment exchange model from myocardial perfusion MR data. The Bayesian scheme allows the incorporation of prior knowledge on the physiological ranges of the model parameters and facilitates the use of the additional information that neighbouring voxels are likely to have similar kinetic parameter values. Hierarchical priors are used to avoid making a priori assumptions on the health of the patients. We provide both a theoretical introduction to Bayesian inference for tracer-kinetic modelling and specific implementation details for this application. This approach is validated in both in silico and in vivo settings. In silico, there was a significant reduction in mean-squared error with the ground-truth parameters using Bayesian inference as compared to using the standard non-linear least squares fitting. When applied to patient data the Bayesian inference scheme returns parameter values that are in-line with those previously reported in the literature, as well as giving parameter maps that match the independant clinical diagnosis of those patients.

Extended quantitative dynamic contrast-enhanced cardiac perfusion imaging in mice using accelerated data acquisition and spatially distributed, two-compartment exchange modeling

The objective of the present work was to improve data acquisition and quantification of dynamic contrast-enhanced perfusion imaging in the in vivo murine heart. Four-fold undersampled data were acquired in 14 mice and reconstructed using k-t SPARSE. A two-compartment exchange model was employed to provide additional characterization of myocardial tissue based on compartment volumes and the permeability surface area product. The feasibility of the proposed method was tested using compartment-based analysis of contrast-enhanced perfusion data acquired with intravascular and extracellular contrast agents. A significantly different permeability surface area product was measured for the intravascular versus extracellular contrast agent (0.13-0.15 ml/g/min vs 0.86-0.88 ml/g/min). The reduced extravasation also resulted in significantly smaller interstitial volumes of the intravascular versus extracellular agent (9.8-11% vs 45-47%). No difference was found for myocardial blood flow (6.5-7.2 ml/g/min vs 6.0-7.0 ml/g/min). The results presented here show that two-compartment exchange modeling in the in vivo murine heart is feasible and gives access to tissue parameters beyond myocardial blood flow.

Monitoring reperfused myocardial infarction with delayed left ventricular systolic dysfunction in rabbits by longitudinal imaging

Background

An experimental imaging platform for longitudinal monitoring and evaluation of cardiac morphology-function changes has been long desired. We sought to establish such a platform by using a rabbit model of reperfused myocardial infarction (MI) that develops chronic left ventricle systolic dysfunction (LVSD) within 7 weeks.

Methods

Fifty-five New Zeeland white (NZW) rabbits received sham-operated or 60-min left circumflex coronary artery (LCx) ligation followed by reperfusion. Cardiac magnetic resonance imaging (cMRI), transthoracic echocardiography (echo), and blood samples were collected at baseline, in acute (48 hours or 1 week) and chronic (7 weeks) stage subsequent to MI for in vivo assessment of infarct size, cardiac morphology, LV function, and myocardial enzymes. Seven weeks post MI, animals were sacrificed and heart tissues were processed for histopathological staining.

Results

The success rate of surgical operation was 87.27%. The animal mortality rates were 12.7% and 3.6% both in acute and chronic stage separately. Serum levels of the myocardial enzyme cardiac Troponin T (cTnT) were significantly increased in MI rabbits as compared with sham animals after 4 hours of operation (P<0.05). According to cardiac morphology and function changes, 4 groups could be distinguished: sham rabbits (n=12), and MI rabbits with no (MI_NO_LVSD; n=10), moderate (MI_M_LVSD; n=9) and severe (MI_S_LVSD; n=15) LVSD. No significant differences in cardiac function or wall thickening between sham and MI_NO_LVSD rabbits were observed at both stages using both cMRI and echo methods. cMRI data showed that MI_M_LVSD rabbits exhibited a reduction of ejection fraction (EF) and an increase in end-systolic volume (ESV) at the acute phase, while at the chronic stage these parameters did not change further. Moreover, in MI_S_LVSD animals, these observations were more striking at the acute stage followed by a further decline in EF and increase in ESV at the chronic stage. Lateral wall thickening determined by cMRI was significantly decreased in MI_M_LVSD versus MI_NO_LVSD animals at both stages (P<0.05). As for MI_S_LVSD versus MI_M_LVSD rabbits, the thickening of anterior, inferior and lateral walls was significantly more decreased at both stages (P<0.05). Echo confirmed the findings of cMRI. Furthermore, these in vivo outcomes including those from vivid cine cMRI could be supported by exactly matched ex vivo histomorphological evidences.

Conclusions

Our findings indicate that chronic LVSD developed over time after surgery-induced MI in rabbits can be longitudinally evaluated using non-invasive imaging techniques and confirmed by the entire-heart-slice histomorphology. This experimental LVSD platform in rabbits may interest researchers in the field of experimental cardiology and help strengthen drug development and translational research for the management of cardiovascular diseases.