Cognitive Machine-Learning Algorithm for Cardiac Imaging

Abstract

Background Associating a patient’s profile with the memories of prototypical patients built through previous repeat clinical experience is a key process in clinical judgment. We hypothesized that a similar process using a cognitive computing tool would be well suited for learning and recalling multidimensional attributes of speckle tracking echocardiography (STE) data sets derived from patients with known constrictive pericarditis (CP) and restrictive cardiomyopathy (RCM). Methods and Results Clinical and echocardiographic data of 50 patients with CP and 44 with RCM were used for developing an associative memory classifier (AMC) based machine learning algorithm. The STE data was normalized in reference to 47 controls with no structural heart disease, and the diagnostic area under the receiver operating characteristic curve (AUC) of the AMC was evaluated for differentiating CP from RCM. Using only STE variables, AMC achieved a diagnostic AUC of 89·2%, which improved to 96·2% with addition of 4 echocardiographic variables. In comparison, the AUC of early diastolic mitral annular velocity and left ventricular longitudinal strain were 82.1% and 63·7%, respectively. Furthermore, AMC demonstrated greater accuracy and shorter learning curves than other machine learning approaches with accuracy asymptotically approaching 90% after a training fraction of 0·3 and remaining flat at higher training fractions. Conclusions This study demonstrates feasibility of a cognitive machine learning approach for learning and recalling patterns observed during echocardiographic evaluations. Incorporation of machine learning algorithms in cardiac imaging may aid standardized assessments and support the quality of interpretations, particularly for novice readers with limited experience.

Predictive Modeling of Hospital Readmission Rates Using Electronic Medical Record-Wide Machine Learning: A Case-Study Using Mount Sinai Heart Failure Cohort

“…In a recent study, we have created temporal models of disease trajectories that could potentially reveal how the population could cluster into subgroups based on age, gender, self-reported ancestry and comorbidities 34 . Further, we have shown that cognitive machine learning can be utilized for precise phenotyping of high volume echocardiography datasets 35 . We have also applied machine learning to understand various features driving patient satisfaction 36 .…”

Machine Learning and Deep Neural Networks in Thoracic and Cardiovascular Imaging

“…Ortiz et al 30 used neural networks to analyze cardiac contractility to predict 1-year mortality in patients with heart failure. Since this early work, supervised machine learning 26 RV, LV endocardium and epicardium CNN Tan et al 27 LV segmentation ANN Baessler et al 28 Myocardial scar detection Random forests Dawes et al 29 Pulmonary hypertension prognosis PCA ECHO Ortiz et al 30 HF prognosis ANN Narula et al 31 HCM vs athlete's heart SVM, Random forests, ANN Sengupta et al 32 Constrictive pericarditis vs restrictive cardiomyopathy AMC, random forest, k-NN, SVM Sengur 33 Valvular disease SVM Moghaddasi and Nourian 34 MR severity SVM Vidya et al 35 MI detection SVM CT Wolterink et al 36 CAC scoring CNN Isgum et al 37 CAC scoring k-NN, SVM Itu et al 38 FFR estimation deep neural network Motwani et al 39 Prognosis Logistic regression Mannil et al 40 MI detection Decision tree, k-NN, random forest, ANN 32 diagnose valvular heart disease, 33 grade severity of mitral valve regurgitation, 34 automate ejection fraction measurement, 53 and detect the presence of myocardial infarction. 35,54 Several machine learning applications have also been developed to assist in the interpretation of CT. For example, algorithms have been developed for the automation of coronary artery calcium scoring 36,37,55,56 and assessment of the functional significance of coronary lesions.…”

Exploration of English Composition Diagnosis System Based on Rule Matching

“…Murata and others as stated in [7] studied the machine learning and various textual features in language use process. In addition, the machine learning algorithm that is proposed by Sengupta and others as stated in [8] are fully considered. The further development of machine learning algorithm has changed the traditional way of research in the field of Natural Language Processing.…”