Characterization of normal and scarred myocardium based on texture analysis of cardiac computed tomography images

Radiomics in Cardiac Computed Tomography

In recent years, there has been an increasing recognition of coronary computed tomographic angiography (CCTA) and gated non-contrast cardiac CT in the workup of coronary artery disease in patients with low and intermediate pretest probability, through the readjustment guidelines by medical societies. However, in routine clinical practice, these CT data sets are usually evaluated dominantly regarding relevant coronary artery stenosis and calcification. The implementation of radiomics analysis, which provides visually elusive quantitative information from digital images, has the potential to open a new era for cardiac CT that goes far beyond mere stenosis or calcification grade estimation. This review offers an overview of the results obtained from radiomics analyses in cardiac CT, including the evaluation of coronary plaques, pericoronary adipose tissue, and the myocardium itself. It also highlights the advantages and disadvantages of use in routine clinical practice.

Myocardial Function Prediction After Coronary Artery Bypass Grafting Using MRI Radiomic Features and Machine Learning Algorithms

The main aim of the present study was to predict myocardial function improvement in cardiac MR (LGE-CMR) images in patients after coronary artery bypass grafting (CABG) using radiomics and machine learning algorithms. Altogether, 43 patients who had visible scars on short-axis LGE-CMR images and were candidates for CABG surgery were selected and enrolled in this study. MR imaging was performed preoperatively using a 1.5-T MRI scanner. All images were segmented by two expert radiologists (in consensus). Prior to extraction of radiomics features, all MR images were resampled to an isotropic voxel size of 1.8 × 1.8 × 1.8 mm³. Subsequently, intensities were quantized to 64 discretized gray levels and a total of 93 features were extracted. The applied algorithms included a smoothly clipped absolute deviation (SCAD)-penalized support vector machine (SVM) and the recursive partitioning (RP) algorithm as a robust classifier for binary classification in this high-dimensional and non-sparse data. All models were validated with repeated fivefold cross-validation and 10,000 bootstrapping resamples. Ten and seven features were selected with SCAD-penalized SVM and RP algorithm, respectively, for CABG responder/non-responder classification. Considering univariate analysis, the GLSZM gray-level non-uniformity-normalized feature achieved the best performance (AUC: 0.62, 95% CI: 0.53-0.76) with SCAD-penalized SVM. Regarding multivariable modeling, SCAD-penalized SVM obtained an AUC of 0.784 (95% CI: 0.64-0.92), whereas the RP algorithm achieved an AUC of 0.654 (95% CI: 0.50-0.82). In conclusion, different radiomics texture features alone or combined in multivariate analysis using machine learning algorithms provide prognostic information regarding myocardial function in patients after CABG.

Cardiac computed tomography radiomics: a narrative review of current status and future directions

BACKGROUND AND OBJECTIVE

In an era of profound growth of medical data and rapid development of advanced imaging modalities, precision medicine increasingly requires further expansion of what can be interpreted from medical images. However, the current interpretation of cardiac computed tomography (CT) images mainly depends on subjective and qualitative analysis. Radiomics uses advanced image analysis to extract numerous quantitative features from digital images that are unrecognizable to the naked eye. Visualization of these features can reveal underlying connections between image phenotyping and biological characteristics and support clinical outcomes. Although research into radiomics on cardiovascular disease began only recently, several studies have indicated its potential clinical value in assessing future cardiac risk and guiding prevention and management strategies. Our review aimed to summarize the current applications of cardiac CT radiomics in the cardiovascular field and discuss its advantages, challenges, and future directions.

METHODS

We searched for English-language articles published between January 2010 and August 2021 in the databases of PubMed, Embase, and Google Scholar. The keywords used in the search included computed tomography or CT, radiomics, cardiovascular or cardiac.

KEY CONTENT AND FINDINGS

The current applications of radiomics in cardiac CT were found to mainly involve research into coronary plaques, perivascular adipose tissue (PVAT), myocardial tissue, and intracardiac lesions. Related findings on cardiac CT radiomics suggested the technique can assist the identification of vulnerable plaques or patients, improve cardiac risk prediction and stratification, discriminate myocardial pathology and etiologies behind intracardiac lesions, and offer new perspective and development prospects to personalized cardiovascular medicine.

CONCLUSIONS

Cardiac CT radiomics can gather additional disease-related information at a microstructural level and establish a link between imaging phenotyping and tissue pathology or biology alone. Therefore, cardiac CT radiomics has significant clinical implications, including a contribution to clinical decision-making. Along with advancements in cardiac CT imaging, cardiac CT radiomics is expected to provide more precise phenotyping of cardiovascular disease for patients and doctors, which can improve diagnostic, prognostic, and therapeutic decision making in the future.

Radiomics and Machine Learning for Detecting Scar Tissue on CT Delayed Enhancement Imaging

Background

Delayed enhancement CT (CT-DE) has been evaluated as a tool for the detection of myocardial scar and compares well to the gold standard of MRI with late gadolinium enhancement (MRI-LGE). Prior work has established that high performance can be achieved with manual reading; however, few studies have looked at quantitative measures to differentiate scar and healthy myocardium on CT-DE or automated analysis.

Methods

Eighteen patients with clinically indicated MRI-LGE were recruited for CT-DE at multiple 80 and 100 kV post contrast imaging. Left ventricle segmentation was performed on both imaging modalities, along with scar segmentation on MRI-LGE. Segmentations were registered together and scar regions were estimated on CT-DE. 93 radiomic features were calculated and analysed for their ability to differentiate between scarred and non-scarred myocardium regions. Machine learning (ML) classifiers were trained using the strongest set of radiomic features to classify segments containing scar on CT-DE. Features and classifiers were compared across both tube voltages and combined-energy images.

Results

There were 59 and 51 statistically significant features in the 80 and 100 kV images respectively. Combined-energy imaging increased this to 63 with more features having area under the curve (AUC) above 0.9. The 10 highest AUC features for each image were used in the ML classifiers. The 100 kV images produced the best ML classifier, a support vector machine with an AUC of 0.88 (95% CI 0.87-0.90). Comparable performance was achieved with both the 80 kV and combined-energy images.

Conclusions

CT-DE can be quantitatively analyzed using radiomic feature calculations. These features may be suitable for ML classification techniques to prospectively identify AHA segments with performance comparable to previously reported manual reading. Future work on larger CT-DE datasets is warranted to establish optimum imaging parameters and features.

Quality assessment of radiomics research in cardiac CT: a systematic review

OBJECTIVES

To assess the quality of current radiomics research on cardiac CT using radiomics quality score (RQS) and Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) systems.

METHODS

Systematic searches of PubMed and EMBASE were performed to identify all potentially relevant original research articles about cardiac CT radiomics. Fifteen original research articles were selected. Two cardiac radiologists assessed the quality of the methodology adopted in those studies according to the RQS and TRIPOD guidelines. Basic adherence rates for the following six key domains were evaluated: image protocol and reproducibility, feature reduction and validation, biologic/clinical utility, performance index, high level of evidence, and open science.

RESULTS

Among the 15 included articles, six (40%) were about coronary artery disease and six (40%) were about myocardial infarction. The mean RQS was 9.9 ± 7.3 (27.4% of the ideal score of 36), and the basic adherence rate was 44.6%. Fourteen (93.3%) and nine (60%) studies performed feature selection and validation, but only two (13.3%) of them performed external validation. Two studies (13.3%) were prospective, and only one study (6.7%) conducted calibration analysis and stated the potential clinical utility. None of the studies conducted phantom study and cost-effective analysis. The overall adherence rate for TRIPOD was 63%.

CONCLUSION

The quality of radiomics studies in cardiac CT is currently insufficient. A higher level of evidence is required, and analysis of clinical utility and calibration of model performance need to be improved.

KEY POINTS

• The quality of science of radiomics studies in cardiac CT is currently insufficient. • No study conducted a phantom study or cost-effective analysis, with further limitations being demonstrated in a high level of evidence for radiomics studies. • Analysis of clinical utility and calibration of model performance need to be improved, and a higher level of evidence is required.

The Emergence of Artificial Intelligence in Cardiology: Current and Future Applications

Artificial intelligence technology is emerging as a promising entity in cardiovascular medicine, potentially improving diagnosis and patient care. In this article, we review the literature on artificial intelligence and its utility in cardiology. We provide a detailed description of concepts of artificial intelligence tools like machine learning, deep learning, and cognitive computing. This review discusses the current evidence, application, prospects, and limitations of artificial intelligence in cardiology.

Artificial intelligence in medical imaging: A radiomic guide to precision phenotyping of cardiovascular disease

ABSTRACT

Rapid technological advances in non-invasive imaging, coupled with the availability of large data sets and the expansion of computational models and power, have revolutionized the role of imaging in medicine. Non-invasive imaging is the pillar of modern cardiovascular diagnostics, with modalities such as cardiac computed tomography (CT) now recognized as first-line options for cardiovascular risk stratification and the assessment of stable or even unstable patients. To date, cardiovascular imaging has lagged behind other fields, such as oncology, in the clinical translational of artificial intelligence (AI)-based approaches. We hereby review the current status of AI in non-invasive cardiovascular imaging, using cardiac CT as a running example of how novel machine learning (ML)-based radiomic approaches can improve clinical care. The integration of ML, deep learning, and radiomic methods has revealed direct links between tissue imaging phenotyping and tissue biology, with important clinical implications. More specifically, we discuss the current evidence, strengths, limitations, and future directions for AI in cardiac imaging and CT, as well as lessons that can be learned from other areas. Finally, we propose a scientific framework in order to ensure the clinical and scientific validity of future studies in this novel, yet highly promising field. Still in its infancy, AI-based cardiovascular imaging has a lot to offer to both the patients and their doctors as it catalyzes the transition towards a more precise phenotyping of cardiovascular disease.

Radiomics for Distinguishing Myocardial Infarction from Myocarditis at Late Gadolinium Enhancement at MRI: Comparison with Subjective Visual Analysis

Purpose

To evaluate whether radiomics features of late gadolinium enhancement (LGE) regions at cardiac MRI enable distinction between myocardial infarction (MI) and myocarditis and to compare radiomics with subjective visual analyses by readers with different experience levels.

Materials and Methods

In this retrospective, institutional review board-approved study, consecutive MRI examinations of 111 patients with MI and 62 patients with myocarditis showing LGE were included. By using open-source software, classification performances attained from two-dimensional (2D) and three-dimensional (3D) texture analysis, shape, and first-order descriptors were compared, applying five different machine learning algorithms. A nested, stratified 10-fold cross-validation was performed. Classification performances were compared through Wilcoxon signed-rank tests. Supervised and unsupervised feature selection techniques were tested; the effect of resampling MR images was analyzed. Subjective image analysis was performed on 2D and 3D image sets by two independent, blinded readers with different experience levels.

Results

When trained with recursive feature elimination (RFE), a support vector machine achieved the best results (accuracy: 88%) for 2D features, whereas linear discriminant analysis (LDA) showed the highest accuracy (85%) for 3D features (P <.05). When trained with principal component analysis (PCA), LDA attained the highest accuracy with both 2D (86%) and 3D (89%; P =.4) features. Results found for classifiers trained with spline resampling were less accurate than those achieved with one-dimensional (1D) nearest-neighbor interpolation (P <.05), whereas results for classifiers trained with 1D nearest-neighbor interpolation and without resampling were similar (P =.1). As compared with the radiomics approach, subjective visual analysis performance was lower for the less experienced and higher for the experienced reader for both 2D and 3D data.

Conclusion

Radiomics features of LGE permit the distinction between MI and myocarditis with high accuracy by using either 2D features and RFE or 3D features and PCA.© RSNA, 2019Supplemental material is available for this article.

Assessment of Remote Myocardium Heterogeneity in Patients with Ventricular Tachycardia Using Texture Analysis of Late Iodine Enhancement (LIE) Cardiac Computed Tomography (cCT) Images

Purpose

Diffuse remodeling of myocardial extra-cellular matrix is largely responsible for left ventricle (LV) dysfunction and arrhythmias. Our hypothesis is that the texture analysis of late iodine enhancement (LIE) cardiac computed tomography (cCT) images may improve characterization of the diffuse extra-cellular matrix changes. Our aim was to extract volumetric extracellular volume (ECV) and LIE texture features of non-scarred (remote) myocardium from cCT of patients with recurrent ventricular tachycardia (rVT), and to compare these radiomic features with LV-function, LV-remodeling, and underlying cardiac disease.

Procedures

Forty-eight patients suffering from rVT were prospectively enrolled: 5/48 with idiopathic VT (IVT), 23/48 with post-ischemic dilated cardiomyopathy (ICM), 9/48 with idiopathic dilated cardiomyopathy (IDCM), and 11/48 with scars from a previous healed myocarditis (MYO). All patients underwent echocardiography to assess LV systolic and diastolic function and cCT with pre-contrast, angiographic, and LIE scan to obtain end-diastolic volume (EDV), ECV, and first-order texture parameters of Hounsfield Unit (HU) of remote myocardium in LIE [energy, entropy, HU-mean, HU-median, standard deviation (SD), and mean absolute deviation (MAD)].

Results

Energy, HU mean, and HU median by cCT texture analysis correlated with ECV (rho = 0.5650, rho = 0.5741, rho = 0.5068; p < 0.0005). cCT-derived ECV, HU-mean, HU-median, SD, and MAD correlated directly to EDV by cCT and inversely to ejection fraction by echocardiography (p < 0.05). SD and MAD correlated with diastolic function by echocardiography (rho = 0.3837, p = 0.0071; rho = 0.3330, p = 0.0208). MYO and IVT patients were characterized by significantly lower values of SD and MAD when compared with ICM and IDCM patients, independently of LV-volume systolic and diastolic function.

Conclusions

Texture analysis of LIE may expand cCT capability of myocardial characterization. Myocardial heterogeneity (SD and MAD) was associated with LV dilatation, systolic and diastolic function, and is able to potentially identify the different patterns of structural remodeling characterizing patients with rVT of different etiology.

Texture analysis of acute myocardial infarction with CT: First experience study

Objective

To investigate the feasibility and accuracy of texture analysis to distinguish through objective and quantitative image information between healthy and infarcted myocardium with computed tomography (CT).

Materials and methods

Twenty patients (5 females; mean age 56±10years) with proven acute myocardial infarction (MI) and 20 patients (8 females; mean age 42±15years) with no cardiac abnormalities (hereafter termed controls) underwent contrast-enhanced cardiac CT. Short axis CT images of the left ventricle (LV) were reconstructed at the slice thicknesses 1mm, 2mm, and 5mm. Two independent, blinded readers segmented the LV in controls and patients. Texture analysis was performed yielding first-level features based on the histogram (variance, skewness, kurtosis, entropy), second-level features based on the gray-level co-occurrence matrix (GLCM) (contrast, correlation, energy and homogeneity), and third-level features based on the gray-level run-length matrix (GLRLM).

Results

Inter-and intrareader agreement was good to excellent for all histogram (intraclass correlation coefficient (ICC):0.70–0.93) and for all GLCM features (ICC:0.66–0.99), and was variable for the GLRLM features (ICC:-0.12–0.99). Univariate analysis showed significant differences between patients and controls for 2/4 histogram features, 3/4 GLCM and for 6/11 GLRLM features and all assessed slice thicknesses (all,p<0.05). In a multivariate logistic regression model, the single best variable from each level, determined by ROC analysis, was included stepwise. The best model included kurtosis (OR 0.08, 95%CI:0.01–0.65,P = 0.018) and short run high gray-level emphasis (SRHGE, OR 0.97, 95%CI:0.94–0.99,P = 0.007), with an area-under-the-curve (AUC) of 0.90 (95%CI:0.80–0.99). The best results for kurtosis and SRHGE (AUC = 0.78) were obtained at a 5mm slice thickness. A cut-off value of 14.4 for kurtosis+0.013*SRHGE predicted acute MI with a sensitivity of 95% (specificity 55%).

Conclusion

Our study illustrates the feasibility of texture analysis for distinguishing healthy from acutely infarcted myocardium with cardiac CT using objective, quantitative features, with most reproducible and accurate results at a short axis slice thickness of 5mm.