Myocardial Infarction Quantification from Late Gadolinium Enhancement MRI Using Top-Hat Transforms and Neural Networks

Automatic detection of myocardial ischaemia using generalisable spatio-temporal hierarchical Bayesian modelling of DCE-MRI

Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) can be used as a non-invasive method for the assessment of myocardial perfusion. The acquired images can be utilised to analyse the spatial extent and severity of myocardial ischaemia (regions with impaired microvascular blood flow). In the present paper, we propose a novel generalisable spatio-temporal hierarchical Bayesian model (GST-HBM) to automate the detection of ischaemic lesions and improve the in silico prediction accuracy by systematically integrating spatio-temporal context information. We present a computational inference procedure with an adequate trade-off between accuracy and computational efficiency, whereby model parameters are sampled from the posterior distribution with Gibbs sampling, while lower-level hyperparameters are selected using model selection strategies based on the Watanabe Akaike information criterion (WAIC). We have assessed our method on both synthetic (in silico) data with known gold-standard and 12 sets of clinical first-pass myocardial perfusion DCE-MRI datasets. We have also carried out a comparative performance evaluation with four established alternative methods: Gaussian mixture model (GMM), opening and closing operations based on Gaussian mixture model (GMMC&Omax), Markov random field constrained Gaussian mixture model (GMM-MRF) and model-based hierarchical Bayesian model (M-HBM). Our results show that the proposed GST-HBM method achieves much higher in silico prediction accuracy than the established alternative methods. Furthermore, this method appears to provide a more robust delineation of ischaemic lesions in datasets affected by spatially variant noise.

Efficacy of catheter-based drug delivery in a hybrid in vitro model of cardiac microvascular obstruction with porcine microthrombi

Microvascular obstruction (MVO) often occurs in ST-elevation myocardial infarction (STEMI) patients after percutaneous coronary intervention (PCI). Diagnosis and treatment of MVO lack appropriate and established procedures. This study focused on two major points by using an in vitro multiscale flow model, which comprised an aortic root model with physiological blood flow and a microfluidic model of the microcirculation with vessel diameters down to 50 μm. First, the influence of porcine microthrombi (MT), injected into the fluidic microchip, on perfusion was investigated. We found that only 43 % of all injected MT were fully occlusive. Second, it could also be shown that the maximal concentration of a dye (representing therapeutic agent) during intracoronary infusion could be increased on average by 58 % , when proximally occluding the coronary artery by a balloon during drug infusion. The obtained results and insights enhance the understanding of perfusion in MVO-affected microcirculation and could lead to improved treatment methods for MVO patients.

Enhanced Drug Delivery for Cardiac Microvascular Obstruction with an Occlusion-Infusion-Catheter

Microvascular Obstruction (MVO) is a common consequence of acute myocardial infarction. MVO is underdiagnosed and treatment is often nonspecific and ineffective. A multi-scale in-vitro benchtop model was established to investigate drug perfusion in MVO affected microcirculation. The central element of the benchtop model was a fluidic microchip containing channels with diameters between [Formula: see text] and 50 μm representing [Formula: see text] of the microvascular tree fed by the left anterior descending artery (LAD). The outlets of the chip could be closed to mimic MVO. Two methods for intracoronary infusion of pharmacologic agents (simulated by dye) to regions with MVO were investigated using an occlusion-infusion catheter. The first case was a simple, bolus-like infusion into the LAD, whereas the second case consisted of infusion with concomitant proximal occlusion of the LAD phantom with a balloon. Results show that local dye concentration maxima in the chip with MVO were 2.2-3.2 times higher for the case with proximal balloon occlusion than for the conventional infusion method. The cumulated dose could be raised by a factor 4.6-5.2. These results suggest that drug infusion by catheter is more effective if the blood supply to the treated vascular bed is temporarily blocked by a balloon catheter.

Effect of Collateral Flow on Catheter-Based Assessment of Cardiac Microvascular Obstruction

Cardiac microvascular obstruction (MVO) associated with acute myocardial infarction (heart attack) is characterized by partial or complete elimination of perfusion in the myocardial microcirculation. A new catheter-based method (CoFI, Controlled Flow Infusion) has recently been developed to diagnose MVO in the catheterization laboratory during acute therapy of the heart attack. A porcine MVO model demonstrates that CoFI can accurately identify the increased hydraulic resistance of the affected microvascular bed. A benchtop microcirculation model was developed and tuned to reproduce in vivo MVO characteristics. The tuned benchtop model was then used to systematically study the effect of different levels of collateral flow. These experiments showed that measurements obtained in the catheter-based method were adversely affected such that collateral flow may be misinterpreted as MVO. Based on further analysis of the measured data, concepts to mitigate the adverse effects were formulated which allow discrimination between collateral flow and MVO.

An Improved 3D Deep Learning-Based Segmentation of Left Ventricular Myocardial Diseases from Delayed-Enhancement MRI with Inclusion and Classification Prior Information U-Net (ICPIU-Net)

Accurate segmentation of the myocardial scar may supply relevant advancements in predicting and controlling deadly ventricular arrhythmias in subjects with cardiovascular disease. In this paper, we propose the architecture of inclusion and classification of prior information U-Net (ICPIU-Net) to efficiently segment the left ventricle (LV) myocardium, myocardial infarction (MI), and microvascular-obstructed (MVO) tissues from late gadolinium enhancement magnetic resonance (LGE-MR) images. Our approach was developed using two subnets cascaded to first segment the LV cavity and myocardium. Then, we used inclusion and classification constraint networks to improve the resulting segmentation of the diseased regions within the pre-segmented LV myocardium. This network incorporates the inclusion and classification information of the LGE-MRI to maintain topological constraints of pathological areas. In the testing stage, the outputs of each segmentation network obtained with specific estimated parameters from training were fused using the majority voting technique for the final label prediction of each voxel in the LGE-MR image. The proposed method was validated by comparing its results to manual drawings by experts from 50 LGE-MR images. Importantly, compared to various deep learning-based methods participating in the EMIDEC challenge, the results of our approach have a more significant agreement with manual contouring in segmenting myocardial diseases.