Typical and atypical imaging features of cardiac amyloidosis

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Cardiac amyloidosis is an uncommon restrictive cardiomyopathy featuring an unregulated amyloid protein deposition that impairs organic function. Early cardiac amyloidosis diagnosis is generally delayed by indistinguishable clinical findings of more frequent hypertrophic diseases. Furthermore, amyloidosis is divided into various groups, according to a generally accepted taxonomy, based on the proteins that make up the amyloid deposits; a careful differentiation between the various forms of amyloidosis is necessary to undertake an adequate therapeutic treatment. Thus, cardiac amyloidosis is thought to be underdiagnosed, which delays necessary therapeutic procedures, diminishing quality of life and impairing clinical prognosis. The diagnostic work-up for cardiac amyloidosis begins with the identification of clinical features, electrocardiographic and imaging findings suggestive or compatible with cardiac amyloidosis, and often requires the histological demonstration of amyloid deposition. One approach to overcome the difficulty of an early diagnosis is the use of automated diagnostic algorithms. Machine learning enables the automatic extraction of salient information from “raw data” without the need for pre-processing methods based on the a priori knowledge of the human operator. This review attempts to assess the various diagnostic approaches and artificial intelligence computational techniques in the detection of cardiac amyloidosis.

Prevalence and Clinical Outcomes of Transthyretin Amyloidosis: A Systematic Review and Meta-analysis

BACKGROUND

Systematic evidence on the prevalence and clinical outcome of transthyretin amyloidosis (ATTR) is missing. We explored: a) the prevalence of cardiac amyloidosis in various patient subgroups, b) survival estimates for ATTR subtypes and c) the effects of novel therapeutics on the natural course of disease.

METHODS

A systematic review of literature published in Medline before 31/12/2021 was performed for the prevalence of cardiac amyloidosis & all-cause mortality of ATTR patients. Extracted data included sample size, age, sex, and all-cause mortality at 1, 2 and 5-years. Subgroup analyses were performed for ATTR subtype i.e., wild type ATTR (wtATTR) vs. hereditary ATTR (htATTR), htATTR genotypes and treatment subgroups.

RESULTS

We identified a total of 62 studies (n=277,882 individuals) reporting the prevalence of cardiac amyloidosis, which was high among patients with a hypertrophic cardiomyopathy phenotype, HFpEF, and elderly with aortic stenosis. Data on ATTR mortality were extracted from 95 studies (n=18,238 ATTR patients). Patients with wtATTR were older (p=7x10^-10 ) and more frequently male (p=5x10^-20 ) vs. htATTR. The 2-year survival of ATTR was 73.3% (95%CI 71.6-76.2); for non-subtyped ATTR 70.4% (95%CI 66.9-73.9), for wtATTR (76.0%, 95%CI: 73.0-78.9) and for htATTR (77.2%, 95%CI: 74.0-80.4); in meta-regression analysis wtATTR was associated with higher survival after adjusting for confounders. There was an interaction between survival and htATTR genotypes (p=10^-15 , Val30Met having the lowest and Val122Ile/Thr60Ala the highest mortality). ATTR 2-year survival was higher on tafamidis/patisiran compared to natural disease course (79.9%, 95%CI: 74.4-85.3 vs. 72.4%, 95%CI 69.8-74.9, p<0.05).

CONCLUSIONS

We report the prevalence of ATTR in various population subgroups and provide survival estimates for the natural course of disease and the effects of novel therapeutics. Important gaps in worldwide epidemiology research in ATTR were identified. This article is protected by copyright. All rights reserved.

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

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.