Deep learning-based atherosclerotic coronary plaque segmentation on coronary CT angiography

A novel radiomics-based technique for identifying vulnerable coronary plaques: a follow-up study

BACKGROUND

Previous reports have suggested that coronary computed tomography angiography (CCTA)-based radiomics analysis is a potentially helpful tool for assessing vulnerable plaques. We aimed to investigate whether coronary radiomic analysis of CCTA images could identify vulnerable plaques in patients with stable angina pectoris.

METHODS

This retrospective study included patients initially diagnosed with stable angina pectoris. Patients were randomly divided into either the training or test dataset at an 8 : 2 ratio. Radiomics features were extracted from CCTA images. Radiomics models for predicting vulnerable plaques were developed using the support vector machine (SVM) algorithm. The model performance was assessed using the area under the curve (AUC); the accuracy, sensitivity, and specificity were calculated to compare the diagnostic performance using the two cohorts.

RESULTS

A total of 158 patients were included in the analysis. The SVM radiomics model performed well in predicting vulnerable plaques, with AUC values of 0.977 and 0.875 for the training and test cohorts, respectively. With optimal cutoff values, the radiomics model showed accuracies of 0.91 and 0.882 in the training and test cohorts, respectively.

CONCLUSION

Although further larger population studies are necessary, this novel CCTA radiomics model may identify vulnerable plaques in patients with stable angina pectoris.

Using artificial intelligence to study atherosclerosis from computed tomography imaging: A state-of-the-art review of the current literature

With the enormous progress in the field of cardiovascular imaging in recent years, computed tomography (CT) has become readily available to phenotype atherosclerotic coronary artery disease. New analytical methods using artificial intelligence (AI) enable the analysis of complex phenotypic information of atherosclerotic plaques. In particular, deep learning-based approaches using convolutional neural networks (CNNs) facilitate tasks such as lesion detection, segmentation, and classification. New radiotranscriptomic techniques even capture underlying bio-histochemical processes through higher-order structural analysis of voxels on CT images. In the near future, the international large-scale Oxford Risk Factors And Non-invasive Imaging (ORFAN) study will provide a powerful platform for testing and validating prognostic AI-based models. The goal is the transition of these new approaches from research settings into a clinical workflow. In this review, we present an overview of existing AI-based techniques with focus on imaging biomarkers to determine the degree of coronary inflammation, coronary plaques, and the associated risk. Further, current limitations using AI-based approaches as well as the priorities to address these challenges will be discussed. This will pave the way for an AI-enabled risk assessment tool to detect vulnerable atherosclerotic plaques and to guide treatment strategies for patients.

Patient-Specific Numerical Simulations of Coronary Artery Hemodynamics and Biomechanics: A Pathway to Clinical Use

PURPOSE

Numerical models that simulate the behaviors of the coronary arteries have been greatly improved by the addition of fluid-structure interaction (FSI) methods. Although computationally demanding, FSI models account for the movement of the arterial wall and more adequately describe the biomechanical conditions at and within the arterial wall. This offers greater physiological relevance over Computational Fluid Dynamics (CFD) models, which assume the walls do not move or deform. Numerical simulations of patient-specific cases have been greatly bolstered by the use of imaging modalities such as Computed Tomography Angiography (CTA), Magnetic Resonance Imaging (MRI), Optical Coherence Tomography (OCT), and Intravascular Ultrasound (IVUS) to reconstruct accurate 2D and 3D representations of artery geometries. The goal of this study was to conduct a comprehensive review on CFD and FSI models on coronary arteries, and evaluate their translational potential.

METHODS

This paper reviewed recent work on patient-specific numerical simulations of coronary arteries that describe the biomechanical conditions associated with atherosclerosis using CFD and FSI models. Imaging modality for geometry collection and clinical applications were also discussed.

RESULTS

Numerical models using CFD and FSI approaches are commonly used to study biomechanics within the vasculature. At high temporal and spatial resolution (compared to most cardiac imaging modalities), these numerical models can generate large amount of biomechanics data.

CONCLUSIONS

Physiologically relevant FSI models can more accurately describe atherosclerosis pathogenesis, and help to translate biomechanical assessment to clinical evaluation.

The Segmentation of Multiple Types of Uterine Lesions in Magnetic Resonance Images Using a Sequential Deep Learning Method with Image-Level Annotations

Fully supervised medical image segmentation methods use pixel-level labels to achieve good results, but obtaining such large-scale, high-quality labels is cumbersome and time consuming. This study aimed to develop a weakly supervised model that only used image-level labels to achieve automatic segmentation of four types of uterine lesions and three types of normal tissues on magnetic resonance images. The MRI data of the patients were retrospectively collected from the database of our institution, and the T2-weighted sequence images were selected and only image-level annotations were made. The proposed two-stage model can be divided into four sequential parts: the pixel correlation module, the class re-activation map module, the inter-pixel relation network module, and the Deeplab v3 + module. The dice similarity coefficient (DSC), the Hausdorff distance (HD), and the average symmetric surface distance (ASSD) were employed to evaluate the performance of the model. The original dataset consisted of 85,730 images from 316 patients with four different types of lesions (i.e., endometrial cancer, uterine leiomyoma, endometrial polyps, and atypical hyperplasia of endometrium). A total number of 196, 57, and 63 patients were randomly selected for model training, validation, and testing. After being trained from scratch, the proposed model showed a good segmentation performance with an average DSC of 83.5%, HD of 29.3 mm, and ASSD of 8.83 mm, respectively. As far as the weakly supervised methods using only image-level labels are concerned, the performance of the proposed model is equivalent to the state-of-the-art weakly supervised methods.

Cardiovascular Computed Tomography in the Diagnosis of Cardiovascular Disease: Beyond Lumen Assessment

Cardiovascular CT is being widely used in the diagnosis of cardiovascular disease due to the rapid technological advancements in CT scanning techniques. These advancements include the development of multi-slice CT, from early generation to the latest models, which has the capability of acquiring images with high spatial and temporal resolution. The recent emergence of photon-counting CT has further enhanced CT performance in clinical applications, providing improved spatial and contrast resolution. CT-derived fractional flow reserve is superior to standard CT-based anatomical assessment for the detection of lesion-specific myocardial ischemia. CT-derived 3D-printed patient-specific models are also superior to standard CT, offering advantages in terms of educational value, surgical planning, and the simulation of cardiovascular disease treatment, as well as enhancing doctor-patient communication. Three-dimensional visualization tools including virtual reality, augmented reality, and mixed reality are further advancing the clinical value of cardiovascular CT in cardiovascular disease. With the widespread use of artificial intelligence, machine learning, and deep learning in cardiovascular disease, the diagnostic performance of cardiovascular CT has significantly improved, with promising results being presented in terms of both disease diagnosis and prediction. This review article provides an overview of the applications of cardiovascular CT, covering its performance from the perspective of its diagnostic value based on traditional lumen assessment to the identification of vulnerable lesions for the prediction of disease outcomes with the use of these advanced technologies. The limitations and future prospects of these technologies are also discussed.

Comparison of Two Contemporary Quantitative Atherosclerotic Plaque Assessment Tools for Coronary Computed Tomography Angiography: Single-Center Analysis and Multi-Center Patient Cohort Validation

BACKGROUND

Coronary computed tomography angiography (CCTA) provides non-invasive quantitative assessments of plaque burden and composition. The quantitative assessment of plaque components requires the use of analysis software that provides reproducible semi-automated plaque detection and analysis. However, commercially available plaque analysis software can vary widely in the degree of automation, resulting in differences in terms of reproducibility and time spent.

AIM

To compare the reproducibility and time spent of two CCTA analysis software tools using different algorithms for the quantitative assessment of coronary plaque volumes and composition in two independent patient cohorts.

METHODS

The study population included 100 patients from two different cohorts: 50 patients from a single-center (Siemens Healthineers, SOMATOM Force (DSCT)) and another 50 patients from a multi-center study (5 different > 64 slice CT scanner types). Quantitative measurements of total calcified and non-calcified plaque volume of the right coronary artery (RCA), left anterior descending (LAD), and left circumflex coronary artery (LCX) were performed on a total of 300 coronaries by two independent readers, using two different CCTA analysis software tools (Tool #1: Siemens Healthineers, syngo.via Frontier CT Coronary Plaque Analysis and Tool #2: Siemens Healthineers, successor CT Coronary Plaque Analysis prototype). In addition, the total time spent for the analysis was recorded with both programs.

RESULTS

The patients in cohorts 1 and 2 were 62.8 ± 10.2 and 70.9 ± 11.7 years old, respectively, 10 (20.0%) and 35 (70.0%) were female and 34 (68.0%) and 20 (40.0%), respectively, had hyperlipidemia. In Cohort #1, the inter- and intra-observer variabilities for the assessment of plaque volumes per patient for Tool #1 versus Tool #2 were 22.8%, 22.0%, and 26.0% versus 2.3%, 3.9%, and 2.5% and 19.7%, 21.4%, and 22.1% versus 0.2%, 0.1%, and 0.3%, respectively, for total, noncalcified, and calcified lesions (p < 0.001 for all between Tools #1 and 2 both for inter- and intra-observer). The inter- and intra-observer variabilities using Tool #2 remained low at 2.9%, 2.7%, and 3.0% and 3.8%, 3.7%, and 4.0%, respectively, for total, non-calcified, and calcified lesions in Cohort #2. For each dataset, the median processing time was higher for Tool #1 versus Tool #2 (459.5 s IQR = 348.0-627.0 versus 208.5 s; IQR = 198.0-216.0) (p < 0.001).

CONCLUSION

The plaque analysis Tool #2 (CT-guided PCI) encompassing a higher degree of automated support required less manual editing, was more time-efficient, and showed a higher intra- and inter-observer reproducibility for the quantitative assessment of plaque volumes both in a representative single-center and in a multi-center validation cohort.

Deep Learning Paradigm and Its Bias for Coronary Artery Wall Segmentation in Intravascular Ultrasound Scans: A Closer Look

BACKGROUND AND MOTIVATION

Coronary artery disease (CAD) has the highest mortality rate; therefore, its diagnosis is vital. Intravascular ultrasound (IVUS) is a high-resolution imaging solution that can image coronary arteries, but the diagnosis software via wall segmentation and quantification has been evolving. In this study, a deep learning (DL) paradigm was explored along with its bias.

METHODS

Using a PRISMA model, 145 best UNet-based and non-UNet-based methods for wall segmentation were selected and analyzed for their characteristics and scientific and clinical validation. This study computed the coronary wall thickness by estimating the inner and outer borders of the coronary artery IVUS cross-sectional scans. Further, the review explored the bias in the DL system for the first time when it comes to wall segmentation in IVUS scans. Three bias methods, namely (i) ranking, (ii) radial, and (iii) regional area, were applied and compared using a Venn diagram. Finally, the study presented explainable AI (XAI) paradigms in the DL framework.

FINDINGS AND CONCLUSIONS

UNet provides a powerful paradigm for the segmentation of coronary walls in IVUS scans due to its ability to extract automated features at different scales in encoders, reconstruct the segmented image using decoders, and embed the variants in skip connections. Most of the research was hampered by a lack of motivation for XAI and pruned AI (PAI) models. None of the UNet models met the criteria for bias-free design. For clinical assessment and settings, it is necessary to move from a paper-to-practice approach.

Medicine 2032: The future of cardiovascular disease prevention with machine learning and digital health technology

Machine learning (ML) refers to computational algorithms that iteratively improve their ability to recognize patterns in data. The digitization of our healthcare infrastructure is generating an abundance of data from electronic health records, imaging, wearables, and sensors that can be analyzed by ML algorithms to generate personalized risk assessments and promote guideline-directed medical management. ML's strength in generating insights from complex medical data to guide clinical decisions must be balanced with the potential to adversely affect patient privacy, safety, health equity, and clinical interpretability. This review provides a primer on key advances in ML for cardiovascular disease prevention and how they may impact clinical practice.