Enhancing Prediction of Myocardial Recovery After Coronary Revascularization: Integrating Radiomics from Myocardial Contrast Echocardiography with Machine Learning
DOI: https://doi.org/10.2147/ijgm.s465023
IF: 2.145
2024-05-31
International Journal of General Medicine
Abstract:Deyi Huang, 1 Xingan Yang, 2 Hongbiao Ruan, 3 Yushui Zhuo, 1 Kai Yuan, 4 Bowen Ruan, 1 Fang Li 1 1 Department of Ultrasound, The People's Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People's Republic of China; 2 Department of Ultrasound, Taizhou Hospital of Zhejiang Province, Linhai City, Zhejiang Province, People's Republic of China; 3 Department of Cardiology, The People's Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People's Republic of China; 4 Department of Clinical Laboratory, The People's Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People's Republic of China Correspondence: Fang Li, Department of Ultrasound, The People's Hospital of Yuhuan, No. 18, Changle Road, Yuhuan City, Zhejiang Province, 317600, People's Republic of China, Tel +8613967672898, Email Introduction: Chronic coronary artery disease (CAD) management often relies on myocardial contrast echocardiography (MCE), yet its effectiveness is limited by subjective interpretations and difficulty in distinguishing hibernating from necrotic myocardium. This study explores the integration of machine learning (ML) with radiomics to predict functional recovery in dyskinetic myocardial segments in CAD patients undergoing revascularization, aiming to overcome these limitations. Methods: This prospective study enrolled 55 chronic CAD patients, dividing into training (39 patients, 205 segments) and testing sets (16 patients, 68 segments). Dysfunctional myocardial segments were identified by initial wall motion scores (WMS) of ≥ 2 (hypokinesis or higher). Functional recovery was defined as a decrease of ≥ 1 grade in WMS during follow-up echocardiography. Radiomics features were extracted from dyssynergic segments in end-systolic phase MCE images across five cardiac cycles post- "flash" impulse and processed through a five-step feature selection. Four ML classifiers were trained and compared using these features and MCE parameters, to identify the optimal model for myocardial recovery prediction. Results: Functional improvement was noted in 139 out of 273 dyskinetic segments (50.9%) following revascularization. Receiver Operating Characteristic (ROC) analysis determined that myocardial blood flow (MBF) was the most precise clinical predictor of recovery, with an area under the curve (AUC) of 0.770. Approximately 1.34 million radiomics features were extracted, with nine features identified as key predictors of myocardial recovery. The random forest (RF) model, integrating MBF values and radiomics features, demonstrated superior predictive accuracy over other ML classifiers. Validation of the RF model on the testing dataset demonstrated its effectiveness, evidenced by an AUC of 0.821, along with consistent calibration and clinical utility. Conclusion: The integration of ML with radiomics from MCE effectively predicts myocardial recovery in CAD. The RF model, combining radiomics and MBF values, presents a non-invasive, precise approach, significantly enhancing CAD management. Keywords: coronary artery disease, myocardial contrast echocardiography, radiomics, machine learning, myocardial recovery prediction, random forest In patients with chronic coronary artery disease (CAD), not all viable myocardium regains contractile function after revascularization. Myocardium that can partially or fully recover contractile function is termed hibernating myocardium (HM). 1 The presence of HM is a key factor in functional recovery post-revascularization. 2,3 Revascularization in myocardial segments without recovery potential offers neither therapeutic benefit nor clinical improvement, and it increases the risks of procedural complications and healthcare costs. 4,5 Therefore, detailed assessment of myocardial viability is essential to target potentially recoverable myocardium and enhance the efficacy of CAD interventions. It is known that myocardial perfusion imaging is crucial for identifying HM. Positron emission tomography (PET) is considered the gold standard for assessing myocardial viability, as it reflects regional myocardial blood flow. 6 However, the high cost of PET scanners and the use of radioactive tracers, which pose potential risks to patients and require specialized facilities and stringent protocols, limit its widespread application in routine clinical practice. In contrast, myocardial contrast echocardiography (MCE) is increasingly recognized for its ability to identify HM. Recommended by the European Society of Cardiology for initial screening in patients suspected of myocardial ischemia, 7 MCE utilizes safe and non-toxic microbubble contrast agents to provide insights into myocardial pe -Abstract Truncated-
medicine, general & internal