Quantification of fibrous cap thickness in intracoronary optical coherence tomography with a contour segmentation method based on dynamic programming

Deep learning segmentation of fibrous cap in intravascular optical coherence tomography images

Thin-cap fibroatheroma (TCFA) is a prominent risk factor for plaque rupture. Intravascular optical coherence tomography (IVOCT) enables identification of fibrous cap (FC), measurement of FC thicknesses, and assessment of plaque vulnerability. We developed a fully-automated deep learning method for FC segmentation. This study included 32,531 images across 227 pullbacks from two registries (TRANSFORM-OCT and UHCMC). Images were semi-automatically labeled using our OCTOPUS with expert editing using established guidelines. We employed preprocessing including guidewire shadow detection, lumen segmentation, pixel-shifting, and Gaussian filtering on raw IVOCT (r,θ) images. Data were augmented in a natural way by changing θ in spiral acquisitions and by changing intensity and noise values. We used a modified SegResNet and comparison networks to segment FCs. We employed transfer learning from our existing much larger, fully-labeled calcification IVOCT dataset to reduce deep-learning training. Postprocessing with a morphological operation enhanced segmentation performance. Overall, our method consistently delivered better FC segmentation results (Dice: 0.837 ± 0.012) than other deep-learning methods. Transfer learning reduced training time by 84% and reduced the need for more training samples. Our method showed a high level of generalizability, evidenced by highly-consistent segmentations across five-fold cross-validation (sensitivity: 85.0 ± 0.3%, Dice: 0.846 ± 0.011) and the held-out test (sensitivity: 84.9%, Dice: 0.816) sets. In addition, we found excellent agreement of FC thickness with ground truth (2.95 ± 20.73 µm), giving clinically insignificant bias. There was excellent reproducibility in pre- and post-stenting pullbacks (average FC angle: 200.9 ± 128.0°/202.0 ± 121.1°). Our fully automated, deep-learning FC segmentation method demonstrated excellent performance, generalizability, and reproducibility on multi-center datasets. It will be useful for multiple research purposes and potentially for planning stent deployments that avoid placing a stent edge over an FC.

Symmetry and asymmetry in atherosclerosis

Atherosclerosis remains the main cause of death worldwide. Most important issues concerning atherosclerosis are hemodynamics and how it affects plaque prevalence and distribution, as well as the symmetry and asymmetry of vasculature and plaques. To present the symmetry in the vascular system an analysis of PubMed and MEDLINE databases was performed. As of February 21, 2023, the results were as follows: for "symmetry" AND "atherosclerosis" there were 47 results; for "symmetry" AND "atherosclerotic lesions" - 20 results; for "symmetry" AND "artery stenosis" - 82 results; for "asymmetry" AND "atherosclerosis" - 87 results. Not without meaning are preventive measures. In the light of the Fourth Industrial Revolution artificial intelligence (AI) solutions help to develop new tools outperforming already existing cardiovascular risk scales. The aim of this paper is to present a current view on symmetry within vasculature and atherosclerosis as well as present a new approach to assess individuals' cardiovascular risk in accordance with precision medicine assumptions. Symmetry and asymmetry within the human vascular system play a crucial role in understanding of arterial diseases, including atherosclerosis. Moreover, it is unavoidable to use AI in cardiovascular risk stratification. Int J Occup Med Environ Health. 2023;36(6):693-703.

Classifying breast cancer in ultrahigh-resolution optical coherence tomography images using convolutional neural networks

Optical coherence tomography (OCT) is being investigated in breast cancer diagnostics as a real-time histology evaluation tool. We present a customized deep convolutional neural network (CNN) for classification of breast tissues in OCT B-scans. Images of human breast samples from mastectomies and breast reductions were acquired using a custom ultrahigh-resolution OCT system with 2.72 µm axial resolution and 5.52 µm lateral resolution. The network achieved 96.7% accuracy, 92% sensitivity, and 99.7% specificity on a dataset of 23 patients. The usage of deep learning will be important for the practical integration of OCT into clinical practice.

Automated Coronary Optical Coherence Tomography Feature Extraction with Application to Three-Dimensional Reconstruction

Coronary optical coherence tomography (OCT) is an intravascular, near-infrared light-based imaging modality capable of reaching axial resolutions of 10-20 µm. This resolution allows for accurate determination of high-risk plaque features, such as thin cap fibroatheroma; however, visualization of morphological features alone still provides unreliable positive predictive capability for plaque progression or future major adverse cardiovascular events (MACE). Biomechanical simulation could assist in this prediction, but this requires extracting morphological features from intravascular imaging to construct accurate three-dimensional (3D) simulations of patients' arteries. Extracting these features is a laborious process, often carried out manually by trained experts. To address this challenge, numerous techniques have emerged to automate these processes while simultaneously overcoming difficulties associated with OCT imaging, such as its limited penetration depth. This systematic review summarizes advances in automated segmentation techniques from the past five years (2016-2021) with a focus on their application to the 3D reconstruction of vessels and their subsequent simulation. We discuss four categories based on the feature being processed, namely: coronary lumen; artery layers; plaque characteristics and subtypes; and stents. Areas for future innovation are also discussed as well as their potential for future translation.

Automatic lumen segmentation using uniqueness of vascular connected region for intravascular optical coherence tomography

We present an automatic lumen segmentation method using uniqueness of connected region for intravascular optical coherence tomography (IVOCT), which can effectively remove the effect on lumen segmentation caused by blood artifacts. Utilizing the uniqueness of vascular wall on A-lines, we detect the A-lines shared by multiple connected regions, identify connected regions generated by blood artifacts using traversal comparison of connected regions' location, shared ratio and area ratio and then remove all artifacts. We compare these three methods by 216 challenging images with severe blood artifacts selected from clinical 1076 IVOCT images. The metrics of the proposed method are evaluated including Dice index, Jaccard index and accuracy of 94.57%, 90.12%, 98.02%. Compared with automatic lumen segmentation based on the previous morphological feature method and widely used dynamic programming method, the metrics of the proposed method are significantly enhanced, especially in challenging images with severe blood artifacts.

Prognostic irrelevance of plaque vulnerability following plaque sealing in high-risk patients with type 2 diabetes: an optical coherence tomography study

Background

Type 2 diabetes mellitus (T2DM) is associated with an increased cardiovascular risk related at least in part to a more vulnerable plaque phenotype. However, patients with T2DM exhibit also an increased risk following percutaneous coronary intervention (PCI). It is unknown if plaque vulnerability of a treated lesion influences cardiovascular outcomes in patients with T2DM. In this study, we aimed to assess the association of plaque morphology as determined by optical coherence tomography (OCT) with cardiovascular outcome following PCI in high-risk patients with T2DM.

Methods

81 patients with T2DM and OCT-guided PCI were recruited. Pre-interventional OCT and systematic follow-up of median 66.0 (IQR = 8.0) months were performed.

Results

During follow-up, 24 patients (29.6%) died. The clinical parameters age (HR 1.16 per year, 95% CI 1.07–1.26, p < 0.001), diabetic polyneuropathy (HR 3.58, 95% CI 1.44–8.93, p = 0.006) and insulin therapy (HR 3.25, 95% CI 1.21–8.70, p = 0.019) predicted mortality in T2DM patients independently. Among OCT parameters only calcium-volume-index (HR 1.71 per 1000°*mm, 95% CI 1.21–2.41, p = 0.002) and lesion length (HR 1.93 per 10 mm, 95% CI 1.02–3.67, p = 0.044) as parameters describing atherosclerosis extent were significant independent predictors of mortality. However, classical features of plaque vulnerability, such as thickness of the fibrous cap, the extent of the necrotic lipid core and the presence of macrophages had no significant predictive value (all p = ns).

Conclusion

Clinical parameters including those describing diabetes severity as well as OCT-parameters characterizing atherosclerotic extent but not classical features of plaque vulnerability predict mortality in T2DM patients following PCI. These data suggest that PCI may provide effective plaque sealing resulting in limited importance of local target lesion vulnerability for future cardiovascular events in high-risk patients with T2DM.

Fully Automated Postlumpectomy Breast Margin Assessment Utilizing Convolutional Neural Network Based Optical Coherence Tomography Image Classification Method

BACKGROUND

The purpose of this study was to develop a deep learning classification approach to distinguish cancerous from noncancerous regions within optical coherence tomography (OCT) images of breast tissue for potential use in an intraoperative setting for margin assessment.

METHODS

A custom ultrahigh-resolution OCT (UHR-OCT) system with an axial resolution of 2.7 μm and a lateral resolution of 5.5 μm was used in this study. The algorithm used an A-scan-based classification scheme and the convolutional neural network (CNN) was implemented using an 11-layer architecture consisting of serial 3 × 3 convolution kernels. Four tissue types were classified, including adipose, stroma, ductal carcinoma in situ, and invasive ductal carcinoma.

RESULTS

The binary classification of cancer versus noncancer with the proposed CNN achieved 94% accuracy, 96% sensitivity, and 92% specificity. The mean five-fold validation F1 score was highest for invasive ductal carcinoma (mean standard deviation, 0.89 ± 0.09) and adipose (0.79 ± 0.17), followed by stroma (0.74 ± 0.18), and ductal carcinoma in situ (0.65 ± 0.15).

CONCLUSION

It is feasible to use CNN based algorithm to accurately distinguish cancerous regions in OCT images. This fully automated method can overcome limitations of manual interpretation including interobserver variability and speed of interpretation and may enable real-time intraoperative margin assessment.

Skin Surface Detection in 3D Optoacoustic Mesoscopy Based on Dynamic Programming

Optoacoustic (photoacoustic) mesoscopy offers unique capabilities in skin imaging and resolves skin features associated with detection, diagnosis, and management of disease. A critical first step in the quantitative analysis of clinical optoacoustic images is to identify the skin surface in a rapid, reliable, and automated manner. Nevertheless, most common edge- and surface-detection algorithms cannot reliably detect the skin surface on 3D raster-scan optoacoustic mesoscopy (RSOM) images, due to discontinuities and diffuse interfaces in the image. We present herein a novel dynamic programming approach that extracts the skin boundary as a 2D surface in one single step, as opposed to consecutive extraction of several independent 1D contours. A domain-specific energy function is introduced, taking into account the properties of volumetric optoacoustic mesoscopy images. The accuracy of the proposed method is validated on scans of the volar forearm of 19 volunteers with different skin complexions, for which the skin surface has been traced manually to provide a reference. In addition, the robustness and the limitations of the method are demonstrated on data where the skin boundaries are low-contrast or ill-defined. The automatic skin surface detection method can improve the speed and accuracy in the analysis of quantitative features seen on the RSOM images and accelerate the clinical translation of the technique. Our method can likely be extended to identify other types of surfaces in the RSOM and other imaging modalities.

Multiple Wavelength Optical Coherence Tomography Assessments for Enhanced Ex Vivo Intra-Cochlear Microstructural Visualization

The precise identification of intra-cochlear microstructures is an essential otorhinolaryngological requirement to diagnose the progression of cochlea related diseases. Thus, we demonstrated an experimental procedure to investigate the most optimal wavelength range, which can enhance the visualization of ex vivo intra-cochlear microstructures using multiple wavelengths (i.e., 860 nm, 1060 nm, and 1300 nm) based optical coherence tomography (OCT) systems. The high-resolution tomograms, volumetric, and quantitative evaluations obtained from Basilar membrane, organ of Corti, and scala vestibule regions revealed complementary comparisons between the aforementioned three distinct wavelengths based OCT systems. Compared to 860 nm and 1300 nm wavelengths, 1060 nm wavelength OCT was discovered to be an appropriate wavelength range verifying the simultaneously obtainable high-resolution and reasonable depth range visualization of intra-cochlear microstructures. Therefore, the implementation of 1060 nm OCT can minimize the necessity of two distinct OCT systems. Moreover, the results suggest that the performed qualitative and quantitative analysis procedure can be used as a powerful tool to explore further anatomical structures of the cochlea for future studies in otorhinolaryngology.

A Survey on Coronary Atherosclerotic Plaque Tissue Characterization in Intravascular Optical Coherence Tomography

PURPOSE OF REVIEW

Atherosclerotic plaque deposition within the coronary vessel wall leads to arterial stenosis and severe catastrophic events over time. Identification of these atherosclerotic plaque components is essential to pre-estimate the risk of cardiovascular disease (CVD) and stratify them as a high or low risk. The characterization and quantification of coronary plaque components are not only vital but also a challenging task which can be possible using high-resolution imaging techniques.

RECENT FINDING

Atherosclerotic plaque components such as thin cap fibroatheroma (TCFA), fibrous cap, macrophage infiltration, large necrotic core, and thrombus are the microstructural plaque components that can be detected with only high-resolution imaging modalities such as intravascular ultrasound (IVUS) and optical coherence tomography (OCT). Light-based OCT provides better visualization of plaque tissue layers of coronary vessel walls as compared to IVUS. Three dominant paradigms have been identified to characterize atherosclerotic plaque components based on optical attenuation coefficients, machine learning algorithms, and deep learning techniques. This review (condensation of 126 papers after downloading 150 articles) presents a detailed comparison among various methodologies utilized for plaque tissue characterization, classification, and arterial measurements in OCT. Furthermore, this review presents the different ways to predict and stratify the risk associated with the CVD based on plaque characterization and measurements in OCT. Moreover, this review discovers three different paradigms for plaque characterization and their pros and cons. Among all of the techniques, a combination of machine learning and deep learning techniques is a best possible solution that provides improved OCT-based risk stratification.

Contour segmentation of the intima, media, and adventitia layers in intracoronary OCT images: application to fully automatic detection of healthy wall regions

PURPOSE

Quantitative and automatic analysis of intracoronary optical coherence tomography images is useful and time-saving to assess cardiovascular risk in the clinical arena.

METHODS

First, the interfaces of the intima, media, and adventitia layers are segmented, by means of an original front propagation scheme, running in a 4D multi-parametric space, to simultaneously extract three non-crossing contours in the initial cross-sectional image. Second, information resulting from the tentative contours is exploited by a machine learning approach to identify healthy and diseased regions of the arterial wall. The framework is fully automatic.

RESULTS

The method was applied to 40 patients from two different medical centers. The framework was trained on 140 images and validated on 260 other images. For the contour segmentation method, the average segmentation errors were [Formula: see text] for the intima-media interface, [Formula: see text] for the media-adventitia interface, and [Formula: see text] for the adventitia-periadventitia interface. The classification method demonstrated a good accuracy, with a median Dice coefficient equal to 0.93 and an interquartile range of (0.78-0.98).

CONCLUSION

The proposed framework demonstrated promising offline performances and could potentially be translated into a reliable tool for various clinical applications, such as quantification of tissue layer thickness and global summarization of healthy regions in entire pullbacks.

Dynamic Programming Based Segmentation in Biomedical Imaging

Many applications in biomedical imaging have a demand on automatic detection of lines, contours, or boundaries of bones, organs, vessels, and cells. Aim is to support expert decisions in interactive applications or to include it as part of a processing pipeline for automatic image analysis. Biomedical images often suffer from noisy data and fuzzy edges. Therefore, there is a need for robust methods for contour and line detection. Dynamic programming is a popular technique that satisfies these requirements in many ways. This work gives a brief overview over approaches and applications that utilize dynamic programming to solve problems in the challenging field of biomedical imaging.

Evaluation of a framework for the co-registration of intravascular ultrasound and optical coherence tomography coronary artery pullbacks

A growing number of studies have used a combination of intravascular ultrasound (IVUS) and optical coherence tomography (OCT) for the assessment of atherosclerotic plaques. Given their respective strengths these imaging modalities highly complement each other. Correlations of hemodynamics and coronary artery disease (CAD) have been extensively investigated with both modalities separately, though not concurrently due to challenges in image registration. Manual co-registration of these modalities is a time expensive task subject to human error, and the development of an automatic method has not been previously addressed. We developed a framework that uses dynamic time warping for the longitudinal co-registration and dynamic programming for the circumferential co-registration of images and evaluated the methodology in a cohort (n = 12) of patients with moderate CAD. Excellent correlation was seen between the algorithm and two expert readers for longitudinal co-registration (CCC = 0.9964, CCC = 0.9959) and circumferential co-registration (CCC = 0.9688, CCC = 0.9598). The mean error of the circumferential co-registration angle was found to be within 10%. A framework for the co-registration of IVUS and OCT pullbacks has been developed which provides a foundation for comprehensive studies of CAD biomechanics.

Prevalence and Predictors of Multiple Coronary Plaque Ruptures: In Vivo 3-Vessel Optical Coherence Tomography Imaging Study

OBJECTIVE

Plaque rupture may be the local expression of a widespread coronary instability. This study aimed to investigate: (1) the prevalence and characteristics of nonculprit plaque rupture; (2) the pancoronary atherosclerotic phenotype in patients with and without nonculprit plaque rupture; and (3) the prevalence and predictors of multiple plaque ruptures.

APPROACH AND RESULTS

Six hundred and seventy-five nonculprit plaques from 261 patients (34 acute myocardial infarction, 73 unstable angina pectoris, and 154 stable angina pectoris) were analyzed by 3-vessel optical coherence tomography. Nonculprit plaque ruptures were identified in 51 patients (20%). Patients with nonculprit plaque ruptures had higher prevalence of thin-cap fibroatheroma (51% versus 13%; P<0.001) in the 3 major epicardial coronary vessels. Multiple plaque ruptures were observed in 20% of patients (38% acute myocardial infarction versus 10% unstable angina pectoris versus 19% stable angina pectoris; P=0.042). Thin-cap fibroatheroma, intimal vasculature, and macrophages were independent morphological predictors of multiple plaque ruptures, whereas acute myocardial infarction and chronic kidney disease were independent clinical predictors. Patients with nonculprit plaque ruptures showed higher 1-year rates of nontarget lesion revascularization (11.8% versus 4.4%; P=0.039).

CONCLUSIONS

Nonculprit plaque ruptures were observed in 20% of patients with coronary artery disease and were associated with pancoronary vulnerability and higher 1-year revascularization rate.

Optical coherence tomography attenuation imaging for lipid core detection: an ex-vivo validation study

Lipid-core atherosclerotic plaques are associated with disease progression, procedural complications, and cardiac events. Coronary plaque lipid can be quantified in optical coherence tomography (OCT) pullbacks by measurement of lipid arcs and lipid lengths; parameters frequently used in clinical research, but labor intensive and subjective to analyse. In this study, we investigated the ability of quantitative attenuation, derived from intravascular OCT, to detect plaque lipid. Lipid cores are associated with a high attenuation coefficient. We compared the index of plaque attenuation (IPA), a local quantitative measure of attenuation, to the manually measured lipid score (arc and length) on OCT images, and to the plaque characterization ex-vivo. We confirmed a correlation between the IPA and lipid scores (r² > 0.7). Comparison to histology shows that high attenuation is associated with fibroatheroma, but also with macrophage presence. IPA is a robust, reproducible, and user-independent measure that facilitates quantification of coronary lipid, a potential tool in clinical research and in guiding percutaneous coronary intervention.

Fusion of fibrous cap thickness and wall shear stress to assess plaque vulnerability in coronary arteries: a pilot study

Purpose

Identification of rupture-prone plaques in coronary arteries is a major clinical challenge. Fibrous cap thickness and wall shear stress are two relevant image-based risk factors, but these two parameters are generally computed and analyzed separately. Accordingly, combining these two parameters can potentially improve the identification of at-risk regions. Therefore, the purpose of this study is to investigate the feasibility of the fusion of wall shear stress and fibrous cap thickness of coronary arteries in patient data.

Methods

Fourteen patients were included in this pilot study. Imaging of the coronary arteries was performed with optical coherence tomography and with angiography. Fibrous cap thickness was automatically quantified from optical coherence tomography pullbacks using a contour segmentation approach based on fast marching. Wall shear stress was computed by applying computational fluid dynamics on the 3D volume reconstructed from two angiograms. The two parameters then were co-registered using anatomical landmarks such as side branches.

Results

The two image modalities were successfully co-registered, with a mean (±SD) error corresponding to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$8.6\,\pm \,6.7\,\%$$\end{document}8.6±6.7% of the length of the analyzed region. For all the analyzed participants, the average thinnest portion of each fibrous cap was \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$129\,\pm \,69\,\upmu \text {m}$$\end{document}129±69μm, and the average WSS value at the location of the fibrous cap was \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1.46\,\pm \,1.16\,\text {Pa}$$\end{document}1.46±1.16Pa. A unique index was finally generated for each patient via the fusion of fibrous cap thickness and wall shear stress measurements, to translate all the measured parameters into a single risk map.

Conclusion

The introduced risk map integrates two complementary parameters and has potential to provide valuable information about plaque vulnerability.

Optical Coherence Tomography For the Detection of the Vulnerable Plaque

Morphological characteristics of the atheromatous plaque have been associated with the development of plaque rupture and the pathogenesis of acute coronary syndromes (ACS). Plaques with a specific morphological phenotype that are at high risk of causing ACS are called vulnerable plaques, and can be identified in vivo through the use of intracoronary imaging. Optical coherence tomography (OCT) is a high-resolution intravascular imaging modality that enables detailed visualization of atheromatous plaques. Consequently, OCT is a valuable research tool for examining the role of morphological characteristics of atheromatous plaques in the progression of coronary artery disease and plaque destabilisation, which leads to the clinical manifestation of ACS. This article summarises the pathophysiological insights obtained by OCT imaging in the formation and rupture of the vulnerable plaque.

OCT demonstrating neoatherosclerosis as part of the continuous process of coronary artery disease

Although the advent of drug-eluting stents has reduced the rates of target vessel revascularization, there are observations of ongoing stent failure occurring very late after stent implantation and presenting as very late restenosis or as very late stent thrombosis. The de novo development of atherosclerosis within the neointimal region, called neoatherosclerosis, has been identified as one of the pathomechanisms of these observed late stent failures. The mechanisms of neoatherosclerosis development and its association with stent failure are currently the subject of intensive research. Optical coherence tomography (OCT) is an invasive imaging modality that allows us to visualize the micromorphology of coronary arteries with near-histological resolution, thus providing detailed assessment of the morphological characteristics of the neointima after stent implantation, including neoatherosclerosis. Several OCT studies have tried to provide in vivo insights in the mechanisms of neoatherosclerosis development and its association with late stent failure. This review summarizes the current insights into neoatherosclerosis obtained with OCT and discusses the association of neoatherosclerosis with late stent failure.