Better decisions through science: exercise testing scores

Identification of ischemic heart disease via machine learning analysis on magnetocardiograms

Ischemic heart disease (IHD) is predominantly the leading cause of death worldwide. Early detection of IHD may effectively prevent severity and reduce mortality rate. Recently, magnetocardiography (MCG) has been developed for the detection of heart malfunction. Although MCG is capable of monitoring the abnormal patterns of magnetic field as emitted by physiologically defective heart, data interpretation is time-consuming and requires highly trained professional. Hence, we propose an automatic method for the interpretation of IHD pattern of MCG recordings using machine learning approaches. Two types of machine learning techniques, namely back-propagation neural network (BNN) and direct kernel self-organizing map (DK-SOM), were applied to explore the IHD pattern recorded by MCG. Data sets were obtained by sequential measurement of magnetic field emitted by cardiac muscle of 125 individuals. Data were divided into training set and testing set of 74 cases and 51 cases, respectively. Predictive performance was obtained by both machine learning approaches. The BNN exhibited sensitivity of 89.7%, specificity of 54.5% and accuracy of 74.5%, while the DK-SOM provided relatively higher prediction performance with a sensitivity, specificity and accuracy of 86.2%, 72.7% and 80.4%, respectively. This finding suggests a high potential of applying machine learning approaches for high-throughput detection of IHD from MCG data.

Usefulness of chronotropic incompetence in response to exercise as a predictor of myocardial infarction in middle-aged men without cardiovascular disease

An attenuated heart rate (HR) response to exercise, or chronotropic incompetence, has been shown to predict adverse cardiac events in subjects without known cardiovascular disease (CVD). The aim of the present study was to investigate whether chronotropic incompetence independently predicts acute myocardial infarction (AMI) in middle-aged men. In addition to previously established chronotropic incompetence variables, we focused on a new chronotropic incompetence variable, the HR increase from 40% to 100% of maximal work capacity (HR40-100), as a predictor of AMI. The subjects were a representative sample of 1,176 middle-aged men who did not have CVD and did not use HR-lowering medication at baseline. The association of chronotropic incompetence variables with the risk of AMI was examined by Cox regression models including numerous known risk factors for AMI. During an average follow-up of 11.0 years, there were 106 AMIs (9.0%). In Cox multivariable model, the risk of AMI increased by 33% for each SD decrement of 13 beats/min in HR40-100 (95% confidence interval [CI] 9 to 64). When considered concurrently, HR40-100 was the only chronotropic incompetence variable that improved the predictive value of the model containing other risk factors for AMI. Men with a low HR40-100 (<46 beats/min) and a heightened increase in systolic blood pressure (SBP) (>67 mm Hg) were at particularly high risk, with a 3.1-times higher incidence of AMI than those with a normal HR40-100 and SBP increase (95% CI 1.7 to 5.7). In conclusion, a low HR40-100 predicted AMI in men without previous CVD independent of other exercise test or clinical variables.

Determining the need for coronary revascularization by an exercise test assessment computer program (ETAP)

The rate at which coronary artery revascularization procedures are performed remains inconsistent, and their risks may be greater and long-term benefits less than imagined by the general public and open to considerable inter-individual variation. However, these risks and benefits can be explicitly estimated for an individual patient from a brief medical history and the results of a standard exercise test by a computer program that uses conventional medical decision making techniques. The program first estimates the prior and post-exercise test probability of coronary artery disease and then employs a decision analysis model to define the risks and benefits associated with different treatment options. These results are provided in a printed report that can become part of the clinical record to be reviewed with the patient. In contrast with traditional clinical intuition, the program consistently and explicitly defines the risks and benefits of coronary artery disease treatments. The program forces physicians and their patients to appraise critically the information and beliefs upon which they base their clinical decisions.