New strategies in the treatment of coronary bifurcations

Improving convolutional neural network learning based on a hierarchical bezier generative model for stenosis detection in X-ray images

BACKGROUND AND OBJECTIVE

Automatic detection of stenosis on X-ray Coronary Angiography (XCA) images may help diagnose early coronary artery disease. Stenosis is manifested by a buildup of plaque in the arteries, decreasing the blood flow to the heart, increasing the risk of a heart attack. Convolutional Neural Networks (CNNs) have been successfully applied to identify pathological, regular, and featured tissues on rich and diverse medical image datasets. Nevertheless, CNNs find operative and performing limitations while working with small and poorly diversified databases. Transfer learning from large natural image datasets (such as ImageNet) has become a de-facto method to improve neural networks performance in the medical image domain.

METHODS

This paper proposes a novel Hierarchical Bezier-based Generative Model (HBGM) to improve the CNNs training process to detect stenosis. Herein, artificial image patches are generated to enlarge the original database, speeding up network convergence. The artificial dataset consists of 10,000 images containing 50% stenosis and 50% non-stenosis cases. Besides, a reliable Fréchet Inception Distance (FID) is used to evaluate the generated data quantitatively. Therefore, by using the proposed framework, the network is pre-trained with the artificial datasets and subsequently fine-tuned using the real XCA training dataset. The real dataset consists of 250 XCA image patches, selecting 125 images for stenosis and the remainder for non-stenosis cases. Furthermore, a Convolutional Block Attention Module (CBAM) was included in the network architecture as a self-attention mechanism to improve the efficiency of the network.

RESULTS

The results showed that the pre-trained networks using the proposed generative model outperformed the results concerning training from scratch. Particularly, an accuracy, precision, sensitivity, and F1-score of 0.8934, 0.9031, 0.8746, 0.8880, 0.9111, respectively, were achieved. The generated artificial dataset obtains a mean FID of 84.0886, with more realistic visual XCA images.

CONCLUSIONS

Different ResNet architectures for stenosis detection have been evaluated, including attention modules into the network. Numerical results demonstrated that by using the HBGM is obtained a higher performance than training from scratch, even outperforming the ImageNet pre-trained models.

Ostial lesion of the anterior descending coronary artery treated via Szabo technique supported by stent boost imaging: a case report

Background

Stenosis at the opening and bifurcation of the anterior descending branch and circumflex branch around the end of the left main trunk is difficult to repair. Accurate positioning of a stent is the key problem.

Case presentations

Here we report the case of a 61-year-old man who suffered from paroxysmal chest pain for 1 year, without history of diabetes or hypertension. The coronary computed tomography showed mixed plaques in the proximal part of the anterior descending artery, with stenosis severe at 80–90%. The emergency coronary angiography showed occlusion of the anterior descending artery. During percutaneous coronary intervention, a drug-eluting stent was implanted into the anterior descending artery using the Szabo technique, supported by stent boost (StentBoost) imaging to pinpoint the location of the lesion. The patient’s paroxysmal chest pain was relieved after the procedure.

Conclusion

We used StentBoost to verify the accuracy of stent placement and the Szabo technique to rectify long-term coronary stenosis, which achieved satisfactory results. Combining the Szabo technique with StentBoost imaging was helpful to accurately evaluate the area and locate the stent when treating this ostial lesion of the anterior descending coronary artery.

Appropriate use criteria for optical coherence tomography guidance in percutaneous coronary interventions : Recommendations of the working group of interventional cardiology of the Netherlands Society of Cardiology

Introduction

Optical coherence tomography (OCT) enables detailed imaging of the coronary wall, lumen and intracoronary implanted devices. Responding to the lack of specific appropriate use criteria (AUC) for this technique, we conducted a literature review and a procedure for appropriate use criteria.

Methods

Twenty-one of all 184 members of the Dutch Working Group on Interventional Cardiology agreed to evaluate 49 pre-specified cases. During a meeting, factual indications were established whereupon members individually rated indications on a 9-point scale, with the opportunity to substantiate their scoring.

Results

Twenty-six indications were rated ‘Appropriate’, eighteen indications ‘May be appropriate’, and five ‘Rarely appropriate’. Use of OCT was unanimously considered ‘Appropriate’ in stent thrombosis, and ‘Appropriate’ for guidance in PCI, especially in distal left main coronary artery and proximal left anterior descending coronary artery, unexplained angiographic abnormalities, and use of bioresorbable vascular scaffold (BVS). OCT was considered ‘Rarely Appropriate’ on top of fractional flow reserve (FFR) for treatment indication, assessment of strut coverage, bypass anastomoses or assessment of proximal left main coronary artery.

Conclusions

The use of OCT in stent thrombosis is unanimously considered ‘Appropriate’ by these experts. Varying degrees of consensus exists on the appropriate use of OCT in other settings.

Electronic supplementary material

The online version of this article (10.1007/s12471-018-1143-z) contains supplementary material, which is available to authorized users.

Evaluation of Different Meshing Techniques for the Case of a Stented Artery

The formation and progression of in-stent restenosis (ISR) in bifurcated vessels may vary depending on the technique used for stenting. This study evaluates the effect of a variety of mesh styles on the accuracy and reliability of computational fluid dynamics (CFD) models in predicting these regions, using an idealized stented nonbifurcated model. The wall shear stress (WSS) and the near-stent recirculating vortices are used as determinants. The meshes comprise unstructured tetrahedral and polyhedral elements. The effects of local refinement, as well as higher-order elements such as prismatic inflation layers and internal hexahedral core, have also been examined. The uncertainty associated with individual mesh style was assessed through verification of calculations using the grid convergence index (GCI) method. The results obtained show that the only condition which allows the reliable comparison of uncertainty estimation between different meshing styles is that the monotonic convergence of grid solutions is in the asymptotic range. Comparisons show the superiority of a flow-adaptive polyhedral mesh over the commonly used adaptive and nonadaptive tetrahedral meshes in terms of resolving the near-stent flow features, GCI value, and prediction of WSS. More accurate estimation of hemodynamic factors was obtained using higher-order elements, such as hexahedral or prismatic grids. Incorporating these higher-order elements, however, was shown to introduce some degrees of numerical diffusion at the transitional area between the two meshes, not necessarily translating into high GCI value. Our data also confirmed the key role of local refinement in improving the performance and accuracy of nonadaptive mesh in predicting flow parameters in models of stented artery. The results of this study can provide a guideline for modeling biofluid domain in complex bifurcated arteries stented in regards to various stenting techniques.

Progress in interventional cardiology: challenges for the future

Cardiovascular disease is the leading cause of death in the western and developing countries. Percutaneous transluminal coronary interventions have become the most prevalent treatment option for coronary artery disease; however, due to serious complications, such as stent thrombosis and in-stent restenosis (ISR), the efficacy and safety of the procedure remain important issues to address. Strategies to overcome these aspects are under extensive investigation. In this review, we summarise relevant milestones during the time to overcome these limitations of coronary stents, such as the development of polymer-free drug-eluting stents (DES) to avoid pro-inflammatory response due to the polymer coating or the developement of stents with cell-directing drugs to, simultaneously, improve re-endothelialisation and inhibit ISR amongst other techniques most recently developed, which have not fully entered the clinical stage. Also the novel concept of fully biodegradable DES featured by the lack of a permanent foreign body promises to be a beneficial and applicable tool to restore a natural vessel with maintained vasomotion and to enable optional subsequent surgical revascularisation.

Unprotected left main coronary artery disease. Surgical therapy

Coronary artery bypass grafting (CABG) has been established as a safe concept in the treatment of unprotected left main coronary artery (ULMCA) disease and is considered the first-line treatment in current interdisciplinary guidelines. Regardless of the complexity of the lesion and the concomitant diseases, CABG is associated with excellent long-term results. Randomized controlled trials investigating the role of percutaneous coronary intervention in the setting of ULMCA lesions are frequently underpowered and suggest that, when restricted to less complex lesions, percutaneous coronary intervention is associated with results comparable to surgery. Thus, there is an urgent need for further randomized controlled trials RCTs with all-comers design to supply precise data under real-life conditions representative of the left main stem anatomy. Until then, an interdisciplinary approach to patients with ULMCA lesions is mandatory in view of the existing guidelines.

Optical coherence tomography for guidance of distal cell recrossing in bifurcation stenting: choosing the right cell matters

AIMS

The aim of this study was to assess the ability of optical coherence tomography (OCT) to guide recrossing during percutaneous coronary interventions in bifurcations and to reduce strut malapposition.

METHODS AND RESULTS

Fifty-two patients undergoing elective treatment of bifurcation lesions using provisional stenting as default strategy were included in the study. Patients were divided into two groups: OCT-guided stent recrossing (group 1, n=12), and angiography-guided recrossing (group 2, n=40). Malapposition in the various bifurcation segments was compared in the two groups, using propensity score analysis to correct for confounders. In 4/12 patients (33%) of the OCT-guided group after the first attempt to recross the stent towards the SB the wire was found to have crossed in a proximal cell, requiring a second and in one case a third attempt to successfully cross through a distal cell. Patients who were treated using OCT-guided recrossing had a significantly lower number of malapposed stent struts, especially in the quadrants towards the SB ostium (9.5%[7.5-17.4%] vs 42.3%[31.2-54.7%] in the angiography-guided group, p<0.0001).

CONCLUSIONS

The rate of strut malapposition was significantly reduced when OCT was used to confirm that wire recrossing was performed in a distal cell of the SB ostium.