Corpus ID: 54440526

Multiple Instance Learning for ECG Risk Stratification

  title={Multiple Instance Learning for ECG Risk Stratification},
  author={Divya Shanmugam and Davis W. Blalock and Jen J. Gong and John V. Guttag},
Patients who suffer an acute coronary syndrome are at elevated risk for adverse cardiovascular events such as myocardial infarction and cardiovascular death. Accurate assessment of this risk is crucial to their course of care. We focus on estimating a patient's risk of cardiovascular death after an acute coronary syndrome based on a patient's raw electrocardiogram (ECG) signal. Learning from this signal is challenging for two reasons: 1) positive examples signifying a downstream cardiovascular… Expand
Deep Learning for Cardiologist-level Myocardial Infarction Detection in Electrocardiograms
This article studies the feasibility of using deep learning to identify suggestive electrocardiographic changes that may correctly classify heart conditions using the Physikalisch-Technische Bundesanstalt (PTB) database, and fine-tune the ConvNetQuake neural network model. Expand
Interpretable Factorization for Neural Network ECG Models
It is shown how to rigorously factor a DNN into a hierarchical equation consisting of black box variables, and it is argued that these component parts should also be "black boxes" --additional targets to interpret heuristically with clear functional connection to the original. Expand
Machine Learning for Biomedical Time Series Classification: From Shapelets to Deep Learning
A real-world use case, the early recognition of sepsis, demonstrates the applicability of the methods discussed, as a gentle introduction to time series classification. Expand
Hybrid modelling for stroke care: Review and suggestions of new approaches for risk assessment and simulation of scenarios
A new hybrid scheme for calculation of stroke risk calculation and simulation of care scenarios is proposed and proposed, which makes use of machine learning, large-scale network, and mechanistic models' respective strengths. Expand


Machine Learning Improves Risk Stratification After Acute Coronary Syndrome
An ANN is constructed that identifies patients at high risk of cardiovascular death (CVD) and captures small changes in the ST segment over time that cannot be detected by visual inspection, highlighting the important role that ANNs can play in risk stratification. Expand
ECG Morphological Variability in Beat Space for Risk Stratification After Acute Coronary Syndrome
ECG morphological variability in beat space contains prognostic information complementary to the clinical variables, LVEF and BNP, in patients with low‐to‐moderate TRS, which could help to risk stratify patients who might not otherwise be considered at high risk of CVD post‐ACS. Expand
ECG Analysis Using Multiple Instance Learning for Myocardial Infarction Detection
The rationale for applying multiple instance learning (MIL) to automated ECG classification is discussed and a new MIL strategy called latent topic MIL is proposed, by which ECGs are mapped into a topic space defined by a number of topics identified over all the unlabeled training heartbeats and support vector machine is directly applied to the ECG-level topic vectors. Expand
Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification
A cardiovascular-patient classifier to identify HCM patients using standard 10-second, 12-lead ECG signals, and results show that a relatively small subset of 264 highly informative features can achieve performance measures comparable to those achieved by using the complete set of features. Expand
Integrated Machine Learning Approaches for Predicting Ischemic Stroke and Thromboembolism in Atrial Fibrillation
In this study, integrated machine learning and data mining approaches are used to build 2-year TE prediction models for AF from Chinese Atrial Fibrillation Registry data to achieve higher prediction performance and identify new potential risk factors as well. Expand
Relation of death within 90 days of non-ST-elevation acute coronary syndromes to variability in electrocardiographic morphology.
It is suggested that increased variation in the entire heart beat morphology is associated with a considerably elevated risk of death and may provide information complementary to the analysis of heart rate. Expand
Classification of cardiac patient states using artificial neural networks.
Three neural network techniques, namely, back propagation, self-organizing maps and radial basis functions, were used for classification of the patient states and the ANN classifier was observed to be correct in approximately 99% of the test cases. Expand
Global Registry of Acute Coronary Events (GRACE) hospital discharge risk score accurately predicts long-term mortality post acute coronary syndrome.
The GRACE postdischarge risk score contains relevant prognostic factors and accurately discriminate survivors from nonsurvivors over the longer term (up to 4 years) in all subsets of acute coronary syndrome patients. Expand
Validation of the atherosclerotic cardiovascular disease Pooled Cohort risk equations.
In this cohort of US adults for whom statin initiation is considered based on the ACC/AHA Pooled Cohort risk equations, observed and predicted 5-year atherosclerotic CVD risks were similar, indicating that these risk equations were well calibrated in the population for which they were designed to be used, and demonstrated moderate to good discrimination. Expand
The TIMI risk score for unstable angina/non-ST elevation MI: A method for prognostication and therapeutic decision making.
In patients with UA/NSTEMI, the TIMI risk score is a simple prognostication scheme that categorizes a patient's risk of death and ischemic events and provides a basis for therapeutic decision making. Expand