June 13, 2018Scientific ReportsOpen Access

Multiclass Classifier based Cardiovascular Condition Detection Using Smartphone Mechanocardiography

Q: What is the clinical evidence from this study?

Study design: Cross-Sectional. Population: Cardiovascular diseases (Atrial Fibrillation, Coronary Artery Disease, STEMI) (n=105). Intervention: Smartphone mechanocardiography (SCG and GCG) vs. Clinical diagnosis and 12-lead ECG (reference standard). Primary outcome: Overall diagnostic accuracy of the 4-class machine learning classifier (Normal, AFib, Pre-PCI, STEMI) with majority voting.

View Full Paper

Ask AI

Bookmark

Key result

A smartphone mechanocardiography-based multiclass classifier using built-in inertial sensors discriminated between normal sinus rhythm, atrial fibrillation, coronary artery disease, and STEMI with an overall accuracy of 75.2%.

Why the study?

Does smartphone mechanocardiography accurately detect and classify cardiovascular conditions such as AFib, CAD, and STEMI?

Study Design

Type

Cross-Sectional (n=105)

Multicenter

Structured PICO

Does smartphone mechanocardiography accurately detect and classify cardiovascular conditions such as AFib, CAD, and STEMI?

Population

105 adults, including healthy controls and patients with atrial fibrillation, coronary artery disease, or myocardial infarction, evaluated using smartphone mechanocardiography.

Exposure

Smartphone mechanocardiography (MCG) using built-in accelerometer and gyroscope sensors (seismocardiography and gyrocardiography) analyzed with machine learning classifiers (KSVM, RF).

Comparator

Healthy controls (normal sinus rhythm) or other disease classes.

Outcome

Classification accuracy, sensitivity, and specificity for detecting cardiovascular conditions (AFib, CAD, STEMI).surrogate

Smartphone mechanocardiography using built-in inertial sensors combined with machine learning can accurately detect and classify cardiovascular conditions like AFib.

Limitations

Control group demographics were significantly different from the patient groups.
Patients in the Pre-PCI and STEMI groups may have suffered from other concurrent diseases such as heart failure.
The study did not consider the detection of other arrhythmias, such as atrial flutter.
The underlying physiological mechanism for the separation between Pre-PCI and STEMI using these features remains unclear.
Small sample size requiring further validation in larger cohorts.
Small or limited amount of samples
AFib patients may have suffered from other conditions such as heart failure

Implication

Should not yet change practice; leaves open validation of smartphone mechanocardiography in larger prospective studies.

Abstract

Cardiac translational and rotational vibrations induced by left ventricular motions are measurable using joint seismocardiography (SCG) and gyrocardiography (GCG) techniques. Multi-dimensional non-invasive monitoring of the heart reveals relative information of cardiac wall motion. A single inertial measurement unit (IMU) allows capturing cardiac vibrations in sufficient details and enables us to perform patient screening for various heart conditions. We envision smartphone mechanocardiography (MCG) for the use of e-health or telemonitoring, which uses a multi-class classifier to detect various types of cardiovascular diseases (CVD) using only smartphone's built-in internal sensors data. Such smartphone App/solution could be used by either a healthcare professional and/or the patient him/herself to take recordings from their heart. We suggest that smartphone could be used to separate heart conditions such as normal sinus rhythm (SR), atrial fibrillation (AFib), coronary artery disease (CAD), and possibly ST-segment elevated myocardial infarction (STEMI) in multiclass settings. An application could run the disease screening and immediately inform the user about the results. Widespread availability of IMUs within smartphones could enable the screening of patients globally in the future, however, we also discuss the possible challenges raised by the utilization of such self-monitoring systems.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Iftikhar et al. (2018) conducted a cross-sectional in Cardiovascular diseases (Atrial Fibrillation, Coronary Artery Disease, STEMI) (n=105). Smartphone mechanocardiography (SCG and GCG) vs. Clinical diagnosis and 12-lead ECG (reference standard) was evaluated on Overall diagnostic accuracy of the 4-class machine learning classifier (Normal, AFib, Pre-PCI, STEMI) with majority voting. A smartphone mechanocardiography-based multiclass classifier using built-in inertial sensors discriminated between normal sinus rhythm, atrial fibrillation, coronary artery disease, and STEMI with an overall accuracy of 75.2%.

synapsesocial.com/papers/6a70738cf44fa9f079de1994 https://doi.org/10.1038/s41598-018-27683-9