A Machine Learning Challenge: Detection of Cardiac Amyloidosis Based on Bi-Atrial and Right Ventricular Strain and Cardiac Function

Abstract

Background: This study challenges state-of-the-art cardiac amyloidosis (CA) diagnostics by feeding multi-chamber strain and cardiac function into supervised machine (SVM) learning algorithms. Methods: Forty-three CA (32 males; 79 years (IQR 71; 85)), 20 patients with hypertrophic cardiomyopathy (HCM, 10 males; 63.9 years (±7.4)) and 44 healthy controls (CTRL, 23 males; 56.3 years (IQR 52.5; 62.9)) received cardiovascular magnetic resonance imaging. Left atrial, right atrial and right ventricular strain parameters and cardiac function generated a 41-feature matrix for decision tree (DT), k-nearest neighbor (KNN), SVM linear and SVM radial basis function (RBF) kernel algorithm processing. A 10-feature principal component analysis (PCA) was conducted using SVM linear and RBF. Results: Forty-one features resulted in diagnostic accuracies of 87.9% (AUC = 0.960) for SVM linear, 90.9% (0.996; Precision = 94%; Sensitivity = 100%; F1-Score = 97%) using RBF kernel, 84.9% (0.970) for KNN, and 78.8% (0.787) for DT. The 10-feature PCA achieved 78.9% (0.962) via linear SVM and 81.8% (0.996) via RBF SVM. Explained variance presented bi-atrial longitudinal strain and left and right atrial ejection fraction as valuable CA predictors. Conclusion: SVM RBF kernel achieved competitive diagnostic accuracies under supervised conditions. Machine learning of multi-chamber cardiac strain and function may offer novel perspectives for non-contrast clinical decision-support systems in CA diagnostics.

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

“…By putting cardiac function and multi-chamber strain into supervised ML algorithms, the work challenges CA diagnoses [ 85 ]. CMR imaging was performed on CA patients, HCM participants, and healthy volunteers.…”

“…Using the strain values for the left, right, and right ventricular atriums as well as heart function, the decision trees (DT), k-nearest neighbor (KNN), SVM linear, and SVM radial basis function (RBF) kernel algorithms developed a 41-feature matrix. Using linear SVM and RBF, a 10-feature principal component analysis (PCA) was carried out [ 85 ]. As a result, under supervised conditions, the SVM RBF kernel obtained competitive diagnostic accuracies.…”

How Artificial Intelligence Can Enhance the Diagnosis of Cardiac Amyloidosis: A Review of Recent Advances and Challenges

“…Another retrospective single-center study by Eckstein et al investigated the diagnostic performance of CMR imaging in detecting CA using ML algorithms [59]. The study used multi-chamber strain and cardiac function as input parameters for a 41-feature matrix decision tree and supervised ML algorithms.…”

“…The study used multi-chamber strain and cardiac function as input parameters for a 41-feature matrix decision tree and supervised ML algorithms. Under supervised conditions, the support vector machine (SVM) algorithm demonstrated competitive diagnostic accuracies of 87.9% (AUC = 0.960) [59]. This suggests that ML of multi-chamber cardiac strain and function could provide innovative insights for non-contrast clinical decision-support systems in the diagnosis of CA.…”

Imaging findings of right cardiac amyloidosis: impact on prognosis and clinical course

“…The study of the right atrial strain and strain rate performed by CMR also proved to be useful in distinguishing amyloid from hypertrophic cardiomyopathy [ 57 ]. Similary, the strain evaluation of biatrial chambers and right ventricle confirmed to be promising in early CA detection [ 58 ].…”