Clinical Value of Machine Learning Model Based on Non-Invasive Myocardial Work Parameters and Glycolipid Metabolism Indicators for the Early Diagnosis of Coronary Heart Disease
DOI:
https://doi.org/10.54097/kbd16e33Keywords:
Coronary Heart Disease, Non-Invasive Myocardial Work, Glycolipid Metabolism Indicators, Machine LearningAbstract
Objective This study aims to explore the clinical value of machine learning models constructed by combining non-invasive Myocardial Work (MW) parameters and clinical glycolipid metabolism indicators for the early diagnosis of coronary heart disease (CHD). Methods A total of 183 suspected CHD patients scheduled for coronary angiography (CAG) admitted to the Affiliated Hospital of Youjiang Medical University for Nationalities from October 2024 to June 2025 were enrolled in this study. Subjects were randomly divided into a training set (n=129) and a validation set (n=54) at a ratio of 7:3. General clinical data, laboratory test results and echocardiographic parameters of all enrolled subjects were collected. Univariate Logistic regression analysis, multicollinearity test and multivariate Logistic regression analysis were performed on the training set to screen independent predictive factors for CHD, based on which six distinct machine learning models were established. The optimal machine learning model was selected comprehensively according to the area under the receiver operating characteristic curve (AUC), accuracy (ACC), sensitivity, specificity, F1-score and Youden’s index. Receiver operating characteristic (ROC) curves, calibration curves and clinical decision curves were adopted to evaluate the diagnostic performance and clinical benefit of the model. Finally, SHapley Additive exPlanations (SHAP) algorithm was applied to conduct interpretability analysis of the final model. Results Multivariate Logistic regression analysis revealed that Metabolic Syndrome(MetS) (OR=3.070, P=0.004), Triglyceride-Glucose Index(TyG index) (OR=24.710, P<0.001), High Density Lipoprotein Cholesterol(HDL) (OR=0.224, P=0.012), Global Constructive Work(GCW) (OR=0.999, P=0.004) and Global Wasted Work(GWW) (OR=1.005, P=0.027) were independent predictive factors for CHD. The Multilayer Perceptron (MLP) model achieved the optimal performance, with an AUC of 0.799, an ACC of 0.796, an F1-score of 0.820, a sensitivity of 0.833, a specificity of 0.750 and a Youden’s index of 0.583 in the validation set. Calibration curve and clinical decision curve analyses demonstrated that the predicted probabilities of the MLP model were in good agreement with the observed actual probabilities, and the model yielded considerable clinical net benefit across a wide range of threshold probabilities. The SHAP algorithm was adopted to visualize variable importance within the MLP model. The results indicated that GCW exerted the greatest predictive contribution, followed by MetS and GWW with moderate contributions, HDL with moderate-to-low contribution, and the TyG index with the lowest contribution. Conclusions The MLP machine learning model constructed based on GCW, GWW, MetS, HDL and TyG index exhibits favorable diagnostic performance for the early identification of CHD, which can provide reference for clinical decision-making.
Downloads
References
[1] Mensah, G. A., Fuster, V., Murray, C., & et al. (2023). Global Burden of Cardiovascular Diseases and Risks, 1990 2022. Journal of the American College of Cardiology, 82(25), 2350 2473. https://doi.org/10.1016/j.jacc.2023.09.031. DOI: https://doi.org/10.1016/j.jacc.2023.09.031
[2] Global Burden of Cardiovascular Diseases and Risks 2023 Collaborators. (2025). Global, regional, and national burden of cardiovascular diseases and risk factors in 204 countries and territories, 1990 2023. Journal of the American College of Cardiology, 86(22), 2167 2243. https://doi.org/10. 1016/j. jacc. 2025. 08.021. DOI: https://doi.org/10.1016/j.jacc.2025.08.015
[3] National Center for Cardiovascular Diseases, & The Writing Committee of the Report on Cardiovascular Health and Diseases in China. (2025). Report on cardiovascular health and diseases in China 2024: an updated summary. Chinese Circulation Journal, 40(6), 521 559.
[4] Ahmadi, A., Leipsic, J., Feuchtner, G., & et al. (2015). Is metabolic syndrome predictive of prevalence, extent, and risk of coronary artery disease beyond its components? Results from the multinational coronary CT angiography evaluation for clinical outcome: an international multicenter registry (CONFIRM). PLOS ONE, 10(3), e118998. https://doi.org/10. 1371/ journal.pone.0118998. DOI: https://doi.org/10.1371/journal.pone.0118998
[5] He, S. L., Hou, X. T., Si, X. F., & et al. (2022). Clinical application of non invasive ultrasonic myocardial work technique in evaluating left ventricular systolic function before and after percutaneous coronary intervention in patients with coronary heart disease. Journal of China Clinic Medical Imaging, 33(6), 396 400.
[6] Koolaji, S., Azadnajafabad, S., Yoosefi, M., & et al. (2025). Subgroups with metabolic syndrome show different association with ischemic heart disease and stroke; findings of Iran STEPS Survey 2021. Journal of the American Heart Association, 14(19), e039240. https://doi.org/10. 1161/ JAHA. 125. 039240.
[7] Mirjalili, S. R., Soltani, S., Heidari, M. Z., & et al. (2023). An innovative model for predicting coronary heart disease using triglyceride glucose index: a machine learning based cohort study. Cardiovascular Diabetology, 22(1), 200. https://doi.org/ 10. 1186/s12933 023 01937 2. DOI: https://doi.org/10.1186/s12933-023-01939-9
[8] Wu, X., Qiu, W., Yang, H., & et al. (2024). Associations of the triglyceride glucose index and atherogenic index of plasma with the severity of new onset coronary artery disease in different glucose metabolic states. Cardiovascular Diabetology, 23(1), 76. https://doi.org/10.1186/s12933 024 02351 8. DOI: https://doi.org/10.1186/s12933-024-02163-9
[9] Xia, X., Chen, S., Tian, X., & et al. (2024). Association of triglyceride glucose index and its related parameters with atherosclerotic cardiovascular disease: evidence from a 15 year follow up of Kailuan cohort. Cardiovascular Diabetology, 23(1), 208. https://doi.org/10.1186/s12933 024 02418 6. DOI: https://doi.org/10.1186/s12933-024-02290-3
[10] Zhao, J., Fan, H., Wang, T., & et al. (2022). TyG index is positively associated with risk of CHD and coronary atherosclerosis severity among NAFLD patients. Cardiovascular Diabetology, 21(1), 123. https://doi.org/10. 1186/ s12933 022 01557 8. DOI: https://doi.org/10.1186/s12933-022-01548-y
[11] Li, C., Zhao, S., Li, Y., & et al. (2025). Triglyceride glucose index predicts adverse cardiovascular events in patients with H type hypertension combined with coronary heart disease: a retrospective cohort study. Cardiovascular Diabetology, 24(1), 45. https://doi.org/10.1186/s12933 025 02431 6. DOI: https://doi.org/10.1186/s12933-024-02555-x
[12] Saffi, H., Winsløw, U., Sakthivel, T., & et al. (2023). Global constructive work is associated with ventricular arrhythmias after cardiac resynchronization therapy. European Heart Journal Cardiovascular Imaging, 25(1), 29 36. https://doi.org/ 10. 1093/ehjci/jead248. DOI: https://doi.org/10.1093/ehjci/jead180
[13] Lin, M., Qin, Y., Ding, X., & et al. (2023). Association between left ventricular geometry and global myocardial work in patients with heart failure with preserved ejection fraction: assessment using strain pressure loop. International Journal of Cardiovascular Imaging, 39(2), 319 329. https://doi.org/10. 1007/ s10554 022 02732 6. DOI: https://doi.org/10.1007/s10554-022-02731-8
[14] Li, X., Chen, H., Han, M., & et al. (2022). Quantitative assessment of left ventricular systolic function in patients with systemic lupus erythematosus: a non invasive pressure strain loop technique. Quantitative Imaging in Medicine and Surgery, 12(6), 3170 3183. https://doi.org/10.21037/qims 22 312. DOI: https://doi.org/10.21037/qims-21-951
[15] Smiseth, O. A., Donal, E., Penicka, M., & et al. (2021). How to measure left ventricular myocardial work by pressure strain loops. European Heart Journal Cardiovascular Imaging, 22(3), 259 261. https://doi.org/10.1093/ehjci/jeaa331. DOI: https://doi.org/10.1093/ehjci/jeaa301
[16] Edwards, N., Scalia, G. M., Shiino, K., & et al. (2019). Global Myocardial Work Is Superior to Global Longitudinal Strain to Predict Significant Coronary Artery Disease in Patients With Normal Left Ventricular Function and Wall Motion. Journal of the American Society of Echocardiography, 32(8), 947 957. https://doi.org/10.1016/j.echo.2019.04.013. DOI: https://doi.org/10.1016/j.echo.2019.02.014
[17] Diamond, G. A., Forrester, J. S. (1979). Analysis of probability as an aid in the clinical diagnosis of coronary artery disease. The New England Journal of Medicine, 300(24), 1350 1358. https://doi.org/10.1056/NEJM197906143002402. DOI: https://doi.org/10.1056/NEJM197906143002402
[18] Luis, S. A., Yamada, A., Khandheria, B. K., & et al. (2014). Use of three dimensional speckle tracking echocardiography for quantitative assessment of global left ventricular function: a comparative study to three dimensional echocardiography. Journal of the American Society of Echocardiography, 27(3), 285 291. https://doi.org/10.1016/j.echo.2013.11.008. DOI: https://doi.org/10.1016/j.echo.2013.11.002
[19] Mao, S., Zhao, L. P., Zhao, X. L., & et al. (2022). Quantitative study on myocardial work evaluated by left ventricular pressure strain loop in patients with different degrees of coronary artery stenosis. Journal of Chinese Physicians, 24(10), 1515 1520.
[20] Lin, J., Wu, W., Gao, L., & et al. (2022). Global Myocardial Work Combined with Treadmill Exercise Stress to Detect Significant Coronary Artery Disease. Journal of the American Society of Echocardiography, 35(3), 247 257. https://doi.org/ 10. 1016/j.echo.2021.11.012. DOI: https://doi.org/10.1016/j.echo.2021.10.009
[21] Zhao, Y., He, F., Guo, W., & et al. (2024). The clinical value of noninvasive left ventricular myocardial work in the diagnosis of myocardial ischemia in coronary heart disease: a comparative study with coronary flow reserve fraction. International Journal of Cardiovascular Imaging, 40(10), 2167 2179. https://doi.org/10.1007/s10554 024 02881 3. DOI: https://doi.org/10.1007/s10554-024-03208-6
[22] Heusch, G. (2016). Myocardial Ischemia: Lack of Coronary Blood Flow or Myocardial Oxygen Supply/Demand Imbalance? Circulation Research, 119(2), 194 196. https://doi.org/10. 1161/ CIRCRESAHA.116.309271. DOI: https://doi.org/10.1161/CIRCRESAHA.116.308925
[23] van der Velden, J. (2015). Historical perspective on heart function: the Frank Starling Law. Biophysical Reviews, 7(4), 421 447. https://doi.org/10.1007/s12551 015 0174 4. DOI: https://doi.org/10.1007/s12551-015-0184-4
[24] Galassi, A., Reynolds, K., & He, J. (2006). Metabolic syndrome and risk of cardiovascular disease: a meta analysis. The American Journal of Medicine, 119(10), 812 819. https:// doi. org/10.1016/j.amjmed.2006.02.031. DOI: https://doi.org/10.1016/j.amjmed.2006.02.031
[25] Gami, A. S., Witt, B. J., Howard, D. E., & et al. (2007). Metabolic syndrome and risk of incident cardiovascular events and death: a systematic review and meta analysis of longitudinal studies. Journal of the American College of Cardiology, 49(4), 403 414. https://doi.org/10. 1016/j.jacc. 2006. 08.060. DOI: https://doi.org/10.1016/j.jacc.2006.09.032
[26] Ferrannini, E. (2007). Metabolic syndrome: a solution in search of a problem. The Journal of Clinical Endocrinology & Metabolism, 92(2), 396 398. https://doi.org/10. 1210/jc. 2006 2151. DOI: https://doi.org/10.1210/jc.2006-0944
[27] Wang, Y., Li, W., Ji, C. Y., & et al. (2024). A multi parameter nomogram for predicting the risk of concurrent coronary heart disease in patients with metabolic syndrome. Anhui Medical and Pharmaceutical Journal, 28(12), 2492 2496.
[28] AV, K., M, C., de la Llera Moya, M., & et al. (2011). Cholesterol efflux capacity, high density lipoprotein function, and atherosclerosis. The New England Journal of Medicine, 364(2), 127 135. https://doi.org/10.1056/NEJMoa1001889. DOI: https://doi.org/10.1056/NEJMoa1001689
[29] C, L., D, D., Z, A., & et al. (2022). Association Between High Density Lipoprotein Cholesterol Levels and Adverse Cardiovascular Outcomes in High risk Populations [Z]. 7, 672 680. DOI: https://doi.org/10.1001/jamacardio.2022.0912
[30] Sajdeya, O., Beran, A., Mhanna, M., & et al. (2022). Triglyceride Glucose Index for the Prediction of Subclinical Atherosclerosis and Arterial Stiffness: A Meta analysis of 37,780 Individuals. Current Problems in Cardiology, 47(12), 101390. https://doi.org/10.1016/j.cpcardiol.2022.101390. DOI: https://doi.org/10.1016/j.cpcardiol.2022.101390
[31] Beck, M. W. (2018). NeuralNetTools: Visualization and Analysis Tools for Neural Networks. Journal of Statistical Software, 85(11), 1 20. https://doi.org/10.18637/jss.v085.i11. DOI: https://doi.org/10.18637/jss.v085.i11
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
