1
|
Esmaeili P, Roshanravan N, Mousavi S, Ghaffari S, Mesri Alamdari N, Asghari-Jafarabadi M. Machine learning framework for atherosclerotic cardiovascular disease risk assessment. J Diabetes Metab Disord 2023; 22:423-430. [PMID: 37255822 PMCID: PMC10225383 DOI: 10.1007/s40200-022-01160-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/15/2022] [Accepted: 11/20/2022] [Indexed: 06/01/2023]
Abstract
Introduction Atherosclerotic cardiovascular disease (ASCVD) is the first leading cause of mortality globally. To identify the individual risk factors of ASCVD utilizing the machine learning (ML) approaches. Materials & methods This cohort-based cross-sectional study was conducted on data of 500 participants with ASCVD among Tabriz University Medical Sciences employees, during 2020. The data with ML methods were developed and validated to predict ASCVD risk with naive Bayes (NB), spurt vesture machines (SVM), regression tree (RT), k-nearest neighbors (KNN), artificial neural networks (ANN), generalized additive models (GAM), and logistic regression (LR). Results Accuracy of the models ranged from 95.7 to 98.1%, with a sensitivity of 50.0 to 97.3%, specificity of 74.3 to 99.1%, positive predictive value (PPV) of 0.0 to 98.0%, negative predictive value (NPV) of 68.4 to 100.0%, positive likelihood ratio (LR +) of 13.8 to 96.4%, negative likelihood ratio (LR-) of 3.6 to 51.9%, and area under ROC curve (AUC) of 62.5 to 99.4%. The ANN fit the data best with an accuracy of 98.1% (95% CI: 96.5-99.1), a specificity of 99.1% (95% CI: 97.7-99.9), a LR + of 96.4% (95% CI: 36.2-258.8), and AUC of 99.4% (95% CI: 85.2-97.0). Based on the optimal model, sex (females), age, smoking, and metabolic syndrome were shown to be the most important risk factors of ASCVD. Conclusion Sex (females), age, smoking, and metabolic syndrome were predictors obtained by ANN. Considering the ANN as the optimal model identified, more accurate prevention planning may be designed.
Collapse
Affiliation(s)
- Parya Esmaeili
- Liver and Gastrointestinal Diseases Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
- Department of Epidemiology and Biostatistics, Faculty of Health, Tabriz University of Medical Sciences, Tabriz, Iran
| | - Neda Roshanravan
- Cardiovascular Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
| | - Saeid Mousavi
- Department of Epidemiology and Biostatistics, Faculty of Health, Tabriz University of Medical Sciences, Tabriz, Iran
| | - Samad Ghaffari
- Cardiovascular Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
| | | | - Mohammad Asghari-Jafarabadi
- Department of Epidemiology and Biostatistics, Faculty of Health, Tabriz University of Medical Sciences, Tabriz, Iran
- Road Traffic Injury Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
- Cabrini Research, Cabrini Health, 154 Wattletree Rd, Malvern, VIC 3144 Australia
- School of Public Health and Preventative Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Clayton, VIC 3800 Australia
| |
Collapse
|
2
|
Amiri M, Ahmadi N, Hadaegh F, Mousavi M, Azizi F, Ramezani Tehrani F. Does the addition of serum antimüllerian hormone concentrations to the Framingham Risk Score and Pooled Cohort Equations improve the prediction of cardiovascular disease? Menopause 2023; 30:406-413. [PMID: 36720078 DOI: 10.1097/gme.0000000000002145] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/02/2023]
Abstract
Abstract
The present study revealed that the addition of serum antimüllerian hormone concentrations to Framingham Risk Score and Pooled Cohort Equations could potentially improve the risk prediction of cardiovascular disease.
Objective
The current study aimed to examine the added value of serum antimüllerian hormone (AMH) concentration to the Framingham Risk Score (FRS) and Pooled Cohort Equations (PCE) in predicting the risk of cardiovascular disease (CVD) in women of reproductive age.
Methods
Women 30 years and older were considered eligible for this population-based prospective study. The univariate and multivariate Cox proportional hazard models were used to evaluate the association between the serum concentrations of AMH and the risk of CVD.
Results
In the enhanced model, which integrated AMH into FRS and PCE and was adjusted for family history of premature CVD, AMH showed a significant association with the risk of CVD during a 19-year follow-up of 800 women (hazard ratio, 0.77 [95% CI, 0.60-0.99] and hazard ratio, 0.64 [95% CI, 0.48-0.84], respectively). According to the likelihood-ratio test, the addition of AMH measurements to FRS and PCE could significantly improve the risk prediction of CVD (P = 0.02 and P < 0.001, respectively); however, the integration of this biomarker did not improve the classification of risk categories.
Conclusions
The present findings revealed that the addition of serum AMH concentrations to FRS and PCE could potentially improve the risk prediction of CVD.
Collapse
Affiliation(s)
- Mina Amiri
- From the Reproductive Endocrinology Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Narjes Ahmadi
- Department of internal Medicine, School of Medicine, Reproductive Endocrinology Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Farzad Hadaegh
- Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Maryam Mousavi
- From the Reproductive Endocrinology Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Fereidoun Azizi
- Endocrine Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Fahimeh Ramezani Tehrani
- From the Reproductive Endocrinology Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| |
Collapse
|
3
|
Tschiderer L, Seekircher L, Willeit P, Peters SAE. Assessment of Cardiovascular Risk in Women: Progress so Far and Progress to Come. Int J Womens Health 2023; 15:191-212. [PMID: 36798791 PMCID: PMC9926980 DOI: 10.2147/ijwh.s364012] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2022] [Accepted: 02/03/2023] [Indexed: 02/11/2023] Open
Abstract
Cardiovascular disease is the leading cause of death in women worldwide. Nonetheless, there exist several uncertainties in the prediction, diagnosis, and treatment of cardiovascular disease in women. A cornerstone in the prediction of cardiovascular disease is the implementation of risk scores. A variety of pregnancy- and reproductive-factors have been associated with lower or higher risk of cardiovascular disease. Consequently, the question has been raised, whether these female-specific factors also provide added value to cardiovascular risk prediction. In this review, we provide an overview of the existing literature on sex differences in the association of established cardiovascular risk factors with cardiovascular disease and the relation between female-specific factors and cardiovascular risk. Furthermore, we systematically reviewed the literature for studies that assessed the added value of female-specific factors beyond already established cardiovascular risk factors. Adding female-specific factors to models containing established cardiovascular risk factors has led to little or no significant improvement in the prediction of cardiovascular events. However, analyses primarily relied on data from women aged ≥40 years. Future investigations are needed to quantify whether pregnancy-related factors improve cardiovascular risk prediction in young women in order to support adequate treatment of risk factors and enhance prevention of cardiovascular disease in women.
Collapse
Affiliation(s)
- Lena Tschiderer
- Institute of Health Economics, Medical University of Innsbruck, Innsbruck, Austria,Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, the Netherlands,Correspondence: Lena Tschiderer, Institute of Health Economics, Medical University of Innsbruck, Innsbruck, Austria, Tel +43 50 504 26272, Email
| | - Lisa Seekircher
- Institute of Health Economics, Medical University of Innsbruck, Innsbruck, Austria
| | - Peter Willeit
- Institute of Health Economics, Medical University of Innsbruck, Innsbruck, Austria,Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK
| | - Sanne A E Peters
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, the Netherlands,The George Institute for Global Health, School of Public Health, Imperial College London, London, UK,The George Institute for Global Health, University of New South Wales, Sydney, New South Wales, Australia
| |
Collapse
|
4
|
Tohidi M, Asgari S, Chary A, Azizi F, Hadaegh F. The association between low-density and non-high-density lipoprotein cholesterol with incident cardiovascular disease among low-risk Iranians during 2 decades follow-up. Clin Biochem 2022; 109-110:28-36. [PMID: 35970222 DOI: 10.1016/j.clinbiochem.2022.08.007] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/22/2022] [Revised: 07/31/2022] [Accepted: 08/10/2022] [Indexed: 11/24/2022]
Abstract
INTRODUCTION To examine the associations between low-density and non-high-density lipoprotein cholesterol (LDL-C and non-HDL-C, respectively) with incident cardiovascular disease (CVD) in low-risk subjects. MATERIALS AND METHODS From a total of 2476 non-diabetic aged 40-70 years, free of CVD with LDL-C range 1.81≤LDL-C<4.91mmol/L with 10-year atherosclerotic cardiovascular disease (ASCVD) risk <7.5%, the associations of LDL-C and non-HDL-C with incident CVD were assessed using multivariable Cox proportional hazard regression analyses adjusted for age, sex, body mass index, waist circumference, HDL-C, triglycerides, chronic kidney disease, current smoking, hypertension, and family history of CVD. RESULTS During a median follow-up of 18 years, 559 CVD events occurred. Compared to the LDL-C <2.59 mmol/L as reference, the categories of 2.59≤LDL-C<3.36, 3.36≤LDL-C<4.14, and ≥4.14 mmol/L were associated with hazard ratios (95% confidence intervals) of 1.39(0.89-2.18), 1.72(1.11-2.68), and 2.19(1.36-3.51) for incident CVD (P for trend<0.0001), respectively. Compared to the non-HDL-C<3.36 as reference, the categories of 3.36≤non-HDL-C<4.14, 4.14≤non-HDL-C<4.91, and ≥4.91 mmol/L were associated with 1.48(0.96-2.30), 1.37(0.89-2.16), and 2.15(1.36-3.39) higher risk for incident CVD (P for trend <0.0001), respectively. Among those with ASCVD score <5% (n=2070), even the 2.59≤LDL-C<3.36 mmol/L increased the risk for CVD [1.73(1.01-2.97)]. Results for non-HDL-C categories remained unchanged except for the category of 4.14≤non-HDL-C<4.91 mmol/L that was not associated with CVD. CONCLUSIONS Among Iranian individuals with ASCVD risk as little as <5%, LDL-C≥2.59 mmol/L and non-HDL-C≥3.36 mmol/L, independent of traditional risk factors, were associated with a significantly higher risk of incident CVD, individuals that might potentially benefit from pharmacological therapy.
Collapse
Affiliation(s)
- Maryam Tohidi
- Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Samaneh Asgari
- Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Abdolreza Chary
- Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Fereidoun Azizi
- Endocrine Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Farzad Hadaegh
- Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
| |
Collapse
|
5
|
Prevalence and significance of risk enhancing biomarkers in the United States population at intermediate risk for atherosclerotic disease: Risk Enhancing Factors in Intermediate Risk for ASCVD. J Clin Lipidol 2021; 16:66-74. [PMID: 34922882 DOI: 10.1016/j.jacl.2021.11.009] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2021] [Revised: 10/20/2021] [Accepted: 11/23/2021] [Indexed: 11/23/2022]
Abstract
BACKGROUND Pooled cohort equations (PCEs) estimate 10-year risk for atherosclerotic cardiovascular disease (ASCVD) in US adults. One use is to guide statin eligibility. However, PCEs risk estimate is inaccurate in some US subpopulations. OBJECTIVE Recent cholesterol guidelines proposed addition of risk enhancing factors to improve risk assessment for selection of statin therapy. This study examines frequencies of several risk enhancing biomarkers in NHANES subjects at intermediate risk (7.5 -<20% 10-year risk for ASCVD) and considers how they may be used to better assess risk for individuals. METHODS Prevalence of the following biomarkers were determined; elevations in apolipoprotein B-containing lipoproteins, i.e., LDL cholesterol (LDL-C) (160-189 mg/dL), non-HDL-cholesterol (non-HDL-C) (190-219 mg/dL), or total apolipoprotein B (apoB) (≥ 130 mg/dL), serum triglyceride (≥175 mg/dL), hemoglobin A1c (5.7-6.4%), high sensitivity C-reactive protein (2-10 mg/L), and waist circumference ≥ 102 cm, and abnormal estimated glomerular filtration rate (15 - ≤ 60 mg/min/1.73 m2). RESULTS 25% of NHANES population had intermediate risk. In this subpopulation, 85% had ≥ 1 biomarkers-similarly in women and men-with a third having ≥3 abnormal markers. Frequencies were not age-related, except in those 40-49 years, in whom > 40% had ≥3 abnormal biomarkers. It made little difference whether LDL-C, non-HDL-C or apoB was used as the atherogenic lipoprotein. CONCLUSION Three or more enhancing risk factors in intermediate risk subjects can complement PCE-estimated 10-year risk and guide the patient-provider discussion toward use of lipid-lowering medication. Future research is needed to integrate risk estimates by PCE and multiple risk enhancers.
Collapse
|