What is the clinical evidence from this study?

Study design: Observational. Population: Coronary artery disease (n=933). Intervention: Machine learning-enhanced echocardiography (Gradient boosting classifier) vs. Coronary angiography. Primary outcome: Area under the receiver operating characteristic curve (AUC) for CAD diagnosis in the test group (AUC 0.852).

May 11, 2023Open Access

Machine learning-enhanced echocardiography for screening coronary artery disease

YGYing GuoKunming University of Science and Technology CXChenxi XiaGuangdong University of Foreign Studies YZYou ZhongPurdue University West Lafayette

Key Result

A gradient-boosting machine learning model utilizing myocardial work and left atrial strain features detected coronary artery disease with an AUC of 0.852, sensitivity of 0.952, and specificity of 0.691.

Study Design

Type

Observational (n=933)

Multicenter

Structured PICO

Does a machine learning-enhanced echocardiography model improve the detection of coronary artery disease in patients with suspected CAD?

Population

933 patients with clinically suspected coronary artery disease (CAD) undergoing coronary angiography (818 in training/testing cohort, 115 in validation cohort), mean age 64.1 years, 37% female, in Beijing, China. Key inclusion: typical myocardial ischemia-related symptoms or positive examinations, ≥18 years, sinus rhythm.

Intervention

Machine learning-enhanced echocardiography model (gradient-boosting classifier using 59 features including myocardial work and left atrial strain)

Comparator

Coronary angiography (reference standard for CAD diagnosis, defined as ≥50% stenosis in one or more coronary arteries)

Outcome

Diagnostic accuracy (Area Under the Curve [AUC], sensitivity, and specificity) for detecting CADsurrogate

A machine learning model incorporating myocardial work and left atrial strain from echocardiography provides high sensitivity for non-invasive screening of coronary artery disease.

Main Result

Effect estimate: AUC 0.852

Limitations

Low specificity of the model may increase the false-positive rate.
Potential confounding by medication differences between patients with and without CAD.
Regional wall motion abnormality was not adopted in the final model due to high collinearity with myocardial work and left atrial strain parameters.
Single-center study design.

Abstract

BACKGROUND: Since myocardial work (MW) and left atrial strain are valuable for screening coronary artery disease (CAD), this study aimed to develop a novel CAD screening approach based on machine learning-enhanced echocardiography. METHODS: This prospective study used data from patients undergoing coronary angiography, in which the novel echocardiography features were extracted by a machine learning algorithm. A total of 818 patients were enrolled and randomly divided into training (80%) and testing (20%) groups. An additional 115 patients were also enrolled in the validation group. RESULTS: The superior diagnosis model of CAD was optimized using 59 echocardiographic features in a gradient-boosting classifier. This model showed that the value of the receiver operating characteristic area under the curve (AUC) was 0.852 in the test group and 0.834 in the validation group, with high sensitivity (0.952) and low specificity (0.691), suggesting that this model is very sensitive for detecting CAD, but its low specificity may increase the high false-positive rate. We also determined that the false-positive cases were more susceptible to suffering cardiac events than the true-negative cases. CONCLUSIONS: Machine learning-enhanced echocardiography can improve CAD detection based on the MW and left atrial strain features. Our developed model is valuable for estimating the pre-test probability of CAD and screening CAD patients in clinical practice. TRIAL REGISTRATION: Registered as NCT03905200 at ClinicalTrials.gov. Registered on 5 April 2019.

KI fragen

Bookmark

View Full Paper