Machine learning of native T1 mapping radiomics for classification of hypertrophic cardiomyopathy phenotypes

Abstract

We explored whether radiomic features from T1 maps by cardiac magnetic resonance (CMR) could enhance the diagnostic value of T1 mapping in distinguishing health from disease and classifying cardiac disease phenotypes. A total of 149 patients (n = 30 with no heart disease, n = 30 with LVH, n = 61 with hypertrophic cardiomyopathy (HCM) and n = 28 with cardiac amyloidosis) undergoing a CMR scan were included in this study. We extracted a total of 850 radiomic features and explored their value in disease classification. We applied principal component analysis and unsupervised clustering in exploratory analysis, and then machine learning for feature selection of the best radiomic features that maximized the diagnostic value for cardiac disease classification. The first three principal components of the T1 radiomics were distinctively correlated with cardiac disease type. Unsupervised hierarchical clustering of the population by myocardial T1 radiomics was significantly associated with myocardial disease type (chi2 = 55.98, p < 0.0001). After feature selection, internal validation and external testing, a model of T1 radiomics had good diagnostic performance (AUC 0.753) for multinomial classification of disease phenotype (normal vs. LVH vs. HCM vs. cardiac amyloid). A subset of six radiomic features outperformed mean native T1 values for classification between myocardial health vs. disease and HCM phenocopies (AUC of T1 vs. radiomics model, for normal: 0.549 vs. 0.888; for LVH: 0.645 vs. 0.790; for HCM 0.541 vs. 0.638; and for cardiac amyloid 0.769 vs. 0.840). We show that myocardial texture assessed by native T1 maps is linked to features of cardiac disease. Myocardial radiomic phenotyping could enhance the diagnostic yield of T1 mapping for myocardial disease detection and classification.

Multimodality imaging in cardiac amyloidosis: State‐of‐the‐art review

“…Diffusion tensor parameters were found to be linked to extracellular amyloid deposition assessed by ECV 90 . On the contrary, machine learning showed to add diagnostic value, on top of native T1, for distinction between healthy myocardium, hypertrophic cardiomyopathy and CA 91 …”

“…90 On the contrary, machine learning showed to add diagnostic value, on top of native T1, for distinction between healthy myocardium, hypertrophic cardiomyopathy and CA. 91 diagnosis, Table 2 summarizes the main diagnostic algorithms published in the last few years. 1,7,51,[92][93][94][95][96][97] Overall, they emphasize the importance of clinical red flags, together with laboratory results, ECG, echocardiography and in most cases CMR for the initial screening.…”

The Road Toward Reproducibility of Parametric Mapping of the Heart: A Technical Review

“…Although, this goes beyond current clinical applications of parametric mapping, it may very well lead to significant knowledge discovery in a broad range of cardiac diseases. The first radiomic studies based on T 1 mapping have shown to provide high diagnostic accuracy for the detection of microvascular obstruction (114) and hypertrophic cardiomyopathy phenotypes (115). However, it should be kept in mind that parametric mapping can only be included in radiomic analyses if the mapping techniques are consistent and reproducible in all patients; this may preclude its current use in multi-centric radiomic studies (116).…”

Big Data in Cardiology: State-of-Art and Future Prospects

“…In the field of cardiological research, radiomics, and radiogenomics can be utilized for the characterization, profiling/phenotyping, and risk stratification of coronary heart disease (CHD), hypertrophic cardiomyopathy, ischemic heart disease, and cerebrovascular disease (70)(71)(72)(73), among others. However, also given its recency, still too much has to be explored in this field.…”