Atherosclerotic plaque characterization in Optical Coherence Tomography images

Artificial Intelligence in Cardiovascular Atherosclerosis Imaging

At present, artificial intelligence (AI) has already been applied in cardiovascular imaging (e.g., image segmentation, automated measurements, and eventually, automated diagnosis) and it has been propelled to the forefront of cardiovascular medical imaging research. In this review, we presented the current status of artificial intelligence applied to image analysis of coronary atherosclerotic plaques, covering multiple areas from plaque component analysis (e.g., identification of plaque properties, identification of vulnerable plaque, detection of myocardial function, and risk prediction) to risk prediction. Additionally, we discuss the current evidence, strengths, limitations, and future directions for AI in cardiac imaging of atherosclerotic plaques, as well as lessons that can be learned from other areas. The continuous development of computer science and technology may further promote the development of this field.

Recent Trends in Artificial Intelligence-Assisted Coronary Atherosclerotic Plaque Characterization

Coronary artery disease is a major cause of morbidity and mortality worldwide. Its underlying histopathology is the atherosclerotic plaque, which comprises lipid, fibrous and—when chronic—calcium components. Intravascular ultrasound (IVUS) and intravascular optical coherence tomography (IVOCT) performed during invasive coronary angiography are reference standards for characterizing the atherosclerotic plaque. Fine image spatial resolution attainable with contemporary coronary computed tomographic angiography (CCTA) has enabled noninvasive plaque assessment, including identifying features associated with vulnerable plaques known to presage acute coronary events. Manual interpretation of IVUS, IVOCT and CCTA images demands scarce physician expertise and high time cost. This has motivated recent research into and development of artificial intelligence (AI)-assisted methods for image processing, feature extraction, plaque identification and characterization. We performed parallel searches of the medical and technical literature from 1995 to 2021 focusing respectively on human plaque characterization using various imaging modalities and the use of AI-assisted computer aided diagnosis (CAD) to detect and classify atherosclerotic plaques, including their composition and the presence of high-risk features denoting vulnerable plaques. A total of 122 publications were selected for evaluation and the analysis was summarized in terms of data sources, methods—machine versus deep learning—and performance metrics. Trends in AI-assisted plaque characterization are detailed and prospective research challenges discussed. Future directions for the development of accurate and efficient CAD systems to characterize plaque noninvasively using CCTA are proposed.

Plaque burden can be assessed using intravascular optical coherence tomography and a dedicated automated processing algorithm: a comparison study with intravascular ultrasound

AIMS

Plaque burden (PB) measurement using intravascular optical coherence tomography (IVOCT) is currently thought to be inferior to intravascular ultrasound (IVUS). We developed an automated IVOCT image processing algorithm to enhance the external elastic lamina (EEL) contour. Thus, we investigated the accuracies of standard IVOCT and an IVOCT enhancement algorithm for measuring PB using IVUS as the reference standard.

METHODS AND RESULTS

The EEL-enhancement algorithm combined adaptive attenuation compensation, exponentiation, angular registration, and image averaging using three sequential frames. In two different laboratories with intravascular imaging expertise, PB was quantified on 200 randomized, matched IVOCT and IVUS images by four independent observers. Fibroatheroma, fibrocalcific plaque, fibrous plaque, pathological intimal thickening (PIT), and mixed plaque were included in each set. Pearson's correlation coefficients between IVUS and standard IVOCT measurements of PB were 0.61, 0.67, 0.76, 0.78, and 0.87 for fibroatheromas, mixed plaques, fibrocalcific plaques, fibrous plaques, and PIT plaques, respectively. Pearson's correlation coefficients increased to 0.81, 0.83, 0.83, 0.84, and 0.90 when using the EEL-enhanced images (P = 0.003, P = 0.004, P = 0.08, P = 0.12, and P = 0.23, respectively). EEL-enhanced IVOCT analysis was associated with a lower EEL-area measurement absolute error for fibroatheromas, mixed plaques, and all pooled plaques (P = 0.006, P = 0.02, and P < 0.001, respectively). Compared with standard IVOCT, the EEL-enhanced IVOCT images had a higher sensitivity (79% vs. 28%, P < 0.001) and specificity (98% vs. 85%, P = 0.03) for plaques with an IVUS PB ≥70%.

CONCLUSION

EEL-enhanced IVOCT can be used to reliably measure PB in all types of coronary atherosclerotic lesions, including fibroatheromas and mixed plaques.

Automated plaque characterization using deep learning on coronary intravascular optical coherence tomographic images

Accurate identification of coronary plaque is very important for cardiologists when treating patients with advanced atherosclerosis. We developed fully-automated semantic segmentation of plaque in intravascular OCT images. We trained/tested a deep learning model on a folded, large, manually annotated clinical dataset. The sensitivities/specificities were 87.4%/89.5% and 85.1%/94.2% for pixel-wise classification of lipidous and calcified plaque, respectively. Automated clinical lesion metrics, potentially useful for treatment planning and research, compared favorably (<4%) with those derived from ground-truth labels. When we converted the results to A-line classification, they were significantly better (p < 0.05) than those obtained previously by using deep learning classifications of A-lines.

Position Paper Computational Cardiology

Computational cardiology is the scientific field devoted to the development of methodologies that enhance our mechanistic understanding, diagnosis and treatment of cardiovascular disease. In this regard, the field embraces the extraordinary pace of discovery in imaging, computational modeling, and cardiovascular informatics at the intersection of atherogenesis and vascular biology. This paper highlights existing methods, practices, and computational models and proposes new strategies to support a multidisciplinary effort in this space. We focus on the means by that to leverage and coalesce these multiple disciplines to advance translational science and computational cardiology. Analyzing the scientific trends and understanding the current needs we present our perspective for the future of cardiovascular treatment.

Automatic atherosclerotic heart disease detection in intracoronary optical coherence tomography images

Intracoronary optical coherence tomography (OCT) is a new invasive imaging system which produces high-resolution images of coronary arteries. Preliminary data suggests that the atherosclerotic disease can be detected from the intracoronary OCT images. However, manual assessment of the intracoronary OCT images is time-consuming and subjective. In this work, we present an automatic atherosclerotic disease detection system on intracoronary OCT images. In the system, a preprocessing scheme is first applied to remove speckle noise and artifacts caused by catheter. Intensity, Histograms of Oriented Gradients (HOG), and Local Binary Patterns (LBP) are then extracted to represent the OCT image. Finally a linear SVM classifier is employed to detect the unhealthy subject. Four-fold cross-validation process is conducted to evaluate the proposed system; and a dataset with 200 images from healthy subjects and 200 images from unhealthy subjects is built to evaluate the system. The mean accuracy is 0.90 and standard deviation is 0.0427, which indicates that the proposed system is accurate and stable.

Fully automated calcium detection using optical coherence tomography

Optical Coherence Tomography (OCT) is a new invasive technology for performing high-resolution cross-sectional imaging of the coronary arteries. In OCT images only Calcified plaque (CA) components can be accurately depicted as light penetrates hard tissue. In this work we present an automated method for detecting CA in OCT images. The method is fully automated as no user intervention is needed and includes three steps. In the first step the region between the lumen and the maximum penetration depth of OCT from the lumen border is determined. In the second step the region is classified into 3 clusters using the K-means algorithm. CA is identified using the results of k-means. The method was validated using experts' annotations on 27 images. The sensitivity of the method is 83% with Positive predictive value (PVV) 74 %.