A machine learning model for identifying patients at risk for wild-type transthyretin amyloid cardiomyopathy

Abstract

Transthyretin amyloid cardiomyopathy, an often unrecognized cause of heart failure, is now treatable with a transthyretin stabilizer. It is therefore important to identify at-risk patients who can undergo targeted testing for earlier diagnosis and treatment, prior to the development of irreversible heart failure. Here we show that a random forest machine learning model can identify potential wild-type transthyretin amyloid cardiomyopathy using medical claims data. We derive a machine learning model in 1071 cases and 1071 non-amyloid heart failure controls and validate the model in three nationally representative cohorts (9412 cases, 9412 matched controls), and a large, single-center electronic health record-based cohort (261 cases, 39393 controls). We show that the machine learning model performs well in identifying patients with cardiac amyloidosis in the derivation cohort and all four validation cohorts, thereby providing a systematic framework to increase the suspicion of transthyretin cardiac amyloidosis in patients with heart failure.

Artificial intelligence model comparison for risk factor analysis of patent ductus arteriosus in nationwide very low birth weight infants cohort

“…The study population was divided into a training set and a test set at a ratio of 80:20 using stratified random sampling 21 . To avoid AA methods overfitting the test set, we applied fivefold cross-validation to the training set in validating procedure, and the area under the receiver operating curve (AUC) was averaged over all the data fold sets 22 , 23 . The stratified cross-validation method ensures that each training and test fold has a similar distribution of outcomes with the entire dataset to reduce bias in the training and evaluating processes.…”

High-Throughput Precision Phenotyping of Left Ventricular Hypertrophy With Cardiovascular Deep Learning

“…Systematic screening paradigms, including through imaging and automated medical record feature review, have shown the opportunity to identify patients with underdiagnosed diseases that are increasingly recognized as more prevalent than was previously thought. 5 , 6 , 7 , 8 , 9 The ability to reliably distinguish between cardiac disease types with similar morphologic features but different causes would also enhance specificity for linking genetic risk variants and determining mechanisms.…”

Updates for the diagnosis and management of cardiac amyloidosis

“…The application of artificial intelligence for screening CA has been reported recently. 51,52 The authors showed that machine learning models well identified patients with CA.…”

“…the adverse effects of these agents (hypotension, kidney failure, conduction disorders) overwhelm benefits. 52 The use of calcium antagonists, particularly non-dihydropyridine calcium channel blocker and digoxin is also avoided. 53 The 2 nd area of treatment is reserved for hematologists and concerns AL-CA.…”