Automated segmentation of metal stent and bioresorbable vascular scaffold in intravascular optical coherence tomography images using deep learning architectures

Percutaneous Coronary Intervention (PCI) with stent placement is a treatment effective for coronary artery diseases. Intravascular optical coherence tomography (OCT) with high resolution is used clinically to visualize stent deployment and restenosis, facilitating PCI operation and for complication inspection. Automated stent struts segmentation in OCT images is necessary as each pullback of OCT images could contain thousands of stent struts. In this paper, a deep learning framework is proposed and demonstrated for the automated segmentation of two major clinical stent types: metal stents and bioresorbable vascular scaffolds (BVS). U-Net, the current most prominent deep learning network in biomedical segmentation, was implemented for segmentation with cropped input. The architectures of MobileNetV2 and DenseNet121 were also adapted into U-Net for improvement in speed and accuracy. The results suggested that the proposed automated algorithm’s segmentation performance approaches the level of independent human observers and is feasible for both types of stents despite their distinct appearance. U-Net with DenseNet121 encoder (U-Dense) performed best with Dice’s coefficient of 0.86 for BVS segmentation, and precision/recall of 0.92/0.92 for metal stent segmentation under optimal crop window size of 256.

Prognostic value of characteristics of plaque combined with residual syntax score among patients with STEMI undergoing primary PCI: an intravascular optical coherence tomography study

Aim The present study aimed to explore these characteristics, particularly thin-cap fibroatheroma (TCFA), in relation to residual syntax score (rSS) in patients who presented with acute MI. Methods and outcomes A total of 434 consecutive patients with MI aged ≥18 years who had STEMI underwent primary PCI. Notably, compared with other subgroups, the presence of TCFA in culprit lesions and a higher level of rSS, were significantly associated with MACE. When rSS was divided into three groups, high rSS levels were associated with a higher incidence of MACE, in the subgroups of without TCFA (P = 0.005), plaque erosion (P = 0.045), macrophage infiltration (P = 0.026), and calcification (P = 0.002). AUC of ROC curve was 0.794 and 0.816, whereas the AUC of the survival ROC was 0.798 and 0.846. Conclusion The results of this study could be used in clinical practice to support risk stratification. Trial registration This study was registered at ClinicalTrials.gov as NCT03593928.

Association between apolipoprotein B/A1 ratio and coronary plaque vulnerability in patients with atherosclerotic cardiovascular disease: an intravascular optical coherence tomography study

Background Apolipoprotein (Apo) A1 and Apo B are strongly associated with the risk of atherosclerotic cardiovascular disease (ASCVD). However, the relationship between the Apo B/A1 ratio and the morphology of coronary vulnerable plaques has not been fully elucidated in patients with ASCVD. Methods A total of 320 patients with ASCVD undergoing percutaneous coronary intervention were enrolled and assigned into acute coronary syndrome (ACS) or chronic coronary syndrome (CCS) group. The morphology of culprit plaque was analyzed by intravascular optical coherence tomography. Association between the Apo B/A1 ratio and coronary vulnerable plaques were evaluated using logistic regression models and receiver operator characteristic (ROC) curve analyses. Results The Apo B/A1 ratio was higher in ACS patients than CCS patients (0.77 ± 0.28 vs. 0.64 ± 0.22, P < 0.001) and it was also higher in patients with plaque rupture, erosion or thrombus than those without culprit plaques. The high Apo B/A1 ratio was associated with high percent of vulnerable plaques compared with low ratio group. The Apo B/A1 ratio was negatively related to fibrous cap thickness in lipid-rich plaque (r = − 0.228, P = 0.043). Univariate and multivariate logistic regression analyses revealed that the Apo B/A1 ratio was an independent factor of plaque rupture, erosion, and thrombus. The area under the ROC curve of the Apo B/A1 ratio for plaque rupture, erosion, and thrombus were 0.632, 0.624, and 0.670 respectively (P < 0.001 for all), which were higher than that of low-density lipoprotein cholesterol. Conclusions The Apo B/A1 ratio is an independent predictor for plaque rupture, erosion, and thrombus in patients with ASCVD.

Automatic Classification of A-Lines in Intravascular OCT Images Using Deep Learning and Estimation of Attenuation Coefficients

Intravascular Optical Coherence Tomography (IVOCT) images provide important insight into every aspect of atherosclerosis. Specifically, the extent of plaque and its type, which are indicative of the patient’s condition, are better assessed by OCT images in comparison to other in vivo modalities. A large amount of imaging data per patient require automatic methods for rapid results. An effective step towards automatic plaque detection and plaque characterization is axial lines (A-lines) based classification into normal and various plaque types. In this work, a novel automatic method for A-line classification is proposed. The method employed convolutional neural networks (CNNs) for classification in its core and comprised the following pre-processing steps: arterial wall segmentation and an OCT-specific (depth-resolved) transformation and a post-processing step based on the majority of classifications. The important step was the OCT-specific transformation, which was based on the estimation of the attenuation coefficient in every pixel of the OCT image. The dataset used for training and testing consisted of 183 images from 33 patients. In these images, four different plaque types were delineated. The method was evaluated by cross-validation. The mean values of accuracy, sensitivity and specificity were 74.73%, 87.78%, and 61.45%, respectively, when classifying into plaque and normal A-lines. When plaque A-lines were classified into fibrolipidic and fibrocalcific, the overall accuracy was 83.47% for A-lines of OCT-specific transformed images and 74.94% for A-lines of original images. This large improvement in accuracy indicates the advantage of using attenuation coefficients when characterizing plaque types. The proposed automatic deep-learning pipeline constitutes a positive contribution to the accurate classification of A-lines in intravascular OCT images.

Prognostic Value of Characteristics of Plaque Combined With Residual Syntax Score Among Patients With STEMI Undergoing Primary PCI: an Intravascular Optical Coherence Tomography Study

AimThe present study aimed to explore these characteristics, particularly thin-cap fibroatheroma (TCFA), in relation to residual syntax score (rSS) in patients who presented with acute MI.Methods and outcomesA total of 434 consecutive patients with MI aged ≥18 years who had STEMI underwent primary PCI. Notably, compared with other subgroups, the presence of TCFA in culprit lesions and a higher level of rSS, were significantly associated with MACE. When rSS was divided into three groups, high rSS levels were associated with a higher incidence of MACE, in the subgroups of without TCFA (P= 0.005), plaque erosion (P= 0.045), macrophage infiltration (P= 0.026), and calcification (P= 0.002). AUC of ROC curve was 0.794 and 0.816, whereas the AUC of the survival ROC was 0.798 and 0.846.ConclusionThe results of this study could be used in clinical practice to support risk stratification.Trial registrationThis study was registered at ClinicalTrials.gov as NCT03593928.