Machine Learning Approaches for Stroke Risk Prediction: Findings from the Suita Study

被引:5
作者
Vu, Thien [1 ,2 ,3 ]
Kokubo, Yoshihiro [2 ]
Inoue, Mai [1 ,2 ]
Yamamoto, Masaki [1 ,2 ]
Mohsen, Attayeb [1 ]
Martin-Morales, Agustin [1 ,2 ]
Inoue, Takao [4 ]
Dawadi, Research [1 ,2 ]
Araki, Michihiro [1 ,2 ,5 ,6 ]
机构
[1] Natl Inst Biomed Innovat Hlth & Nutr, Artificial Intelligence Ctr Hlth & Biomed Res, 3-17 Senrioka shinmachi, Settsu 5660002, Japan
[2] Natl Cerebral & Cardiovasc Ctr, 6-1 Kishibe Shinmachi, Suita, Osaka 5648565, Japan
[3] Cho Ray Hosp, Cardiovasc Ctr, Dept Vasc Surg, Ho Chi Minh City 72713, Vietnam
[4] Yamato Univ, Fac Informat, 2-5-1 Katayama, Suita 5640082, Japan
[5] Kyoto Univ, Grad Sch Med, Dept Resp Med, 54 Shogoin Kawahara cho,Sakyo ku, Kyoto 6068507, Japan
[6] Kobe Univ, Grad Sch Sci Technol & Innovat, 1-1 Rokkodai Cho,Nada Ku, Kobe 6578501, Japan
基金
日本科学技术振兴机构;
关键词
stroke; supervised machine learning; unsupervised machine learning; logistic regression; random forest; support vector machine (SVM); extreme gradient boost (XGBoost); light gradient boosted machine (LightGBM); k-prototype clustering; Shapley Additive Explanations (SHAP); JAPANESE URBAN COHORT; CARDIOVASCULAR-DISEASE; HEMOGLOBIN CONCENTRATION; ATRIAL-FIBRILLATION; GLYCATED ALBUMIN; ISCHEMIC-STROKE; BLOOD-PRESSURE; ASSOCIATION; INCIDENT; FRUCTOSAMINE;
D O I
10.3390/jcdd11070207
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Stroke constitutes a significant public health concern due to its impact on mortality and morbidity. This study investigates the utility of machine learning algorithms in predicting stroke and identifying key risk factors using data from the Suita study, comprising 7389 participants and 53 variables. Initially, unsupervised k-prototype clustering categorized participants into risk clusters, while five supervised models including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosted Machine (LightGBM) were employed to predict stroke outcomes. Stroke incidence disparities among identified risk clusters using the unsupervised k-prototype clustering method are substantial, according to the findings. Supervised learning, particularly RF, was a preferable option because of the higher levels of performance metrics. The Shapley Additive Explanations (SHAP) method identified age, systolic blood pressure, hypertension, estimated glomerular filtration rate, metabolic syndrome, and blood glucose level as key predictors of stroke, aligning with findings from the unsupervised clustering approach in high-risk groups. Additionally, previously unidentified risk factors such as elbow joint thickness, fructosamine, hemoglobin, and calcium level demonstrate potential for stroke prediction. In conclusion, machine learning facilitated accurate stroke risk predictions and highlighted potential biomarkers, offering a data-driven framework for risk assessment and biomarker discovery.
引用
收藏
页数:13
相关论文
共 44 条
[1]  
Akiba T, 2019, Arxiv, DOI [arXiv:1907.10902, 10.48550/arXiv.1907.10902]
[2]   Cardiovascular Event Prediction by Machine Learning The Multi-Ethnic Study of Atherosclerosis [J].
Ambale-Venkatesh, Bharath ;
Yang, Xiaoying ;
Wu, Colin O. ;
Liu, Kiang ;
Hundley, W. Gregory ;
McClelland, Robyn ;
Gomes, Antoinette S. ;
Folsom, Aaron R. ;
Shea, Steven ;
Guallar, Eliseo ;
Bluemke, David A. ;
Lima, Joao A. C. .
CIRCULATION RESEARCH, 2017, 121 (09) :1092-+
[3]  
[Anonymous], 2018, TOP 10 CAUS DEATH
[4]   Developing a Stroke Risk Prediction Model Using Cardiovascular Risk Factors: The Suita Study [J].
Arafa, Ahmed ;
Kokubo, Yoshihiro ;
Sheerah, Haytham A. ;
Sakai, Yukie ;
Watanabe, Emi ;
Li, Jiaqi ;
Honda-Kohmo, Kyoko ;
Teramoto, Masayuki ;
Kashima, Rena ;
Nakao, Yoko M. ;
Koga, Masatoshi .
CEREBROVASCULAR DISEASES, 2022, 51 (03) :323-330
[5]   Association of Prediabetes and Diabetes With Stroke Symptoms [J].
Carson, April P. ;
Muntner, Paul ;
Kissela, Brett M. ;
Kleindorfer, Dawn O. ;
Howard, Virginia J. ;
Meschia, James F. ;
Williams, Linda S. ;
Prineas, Ronald J. ;
Howard, George ;
Safford, Monika M. .
DIABETES CARE, 2012, 35 (09) :1845-1852
[6]   Association between estimated glomerular filtration rate and clinical outcomes in ischemic stroke patients with high-grade carotid artery stenosis [J].
Chao, Chung-Hao ;
Wu, Chia-Lun ;
Huang, Wen-Yi .
BMC NEUROLOGY, 2021, 21 (01)
[7]   Calcium Intake and Serum Calcium Level in Relation to the Risk of Ischemic Stroke: Findings from the REGARDS Study [J].
Dibaba, Daniel T. ;
Xun, Pengcheng ;
Fly, Alyce D. ;
Bidulescu, Aurelian ;
Tsinovoi, Cari L. ;
Judd, Suzanne E. ;
McClure, Leslie A. ;
Cushman, Mary ;
Unverzagt, Frederick W. ;
He, Ka .
JOURNAL OF STROKE, 2019, 21 (03) :312-+
[8]   Stroke Risk Prediction with Machine Learning Techniques [J].
Dritsas, Elias ;
Trigka, Maria .
SENSORS, 2022, 22 (13)
[9]   World Stroke Organization (WSO): Global Stroke Fact Sheet 2022 [J].
Feigin, Valery L. ;
Brainin, Michael ;
Norrving, Bo ;
Martins, Sheila ;
Sacco, Ralph L. ;
Hacke, Werner ;
Fisher, Marc ;
Pandian, Jeyaraj ;
Lindsay, Patrice .
INTERNATIONAL JOURNAL OF STROKE, 2022, 17 (01) :18-29
[10]   Random forest-based prediction of stroke outcome [J].
Fernandez-Lozano, Carlos ;
Hervella, Pablo ;
Mato-Abad, Virginia ;
Rodriguez-Yanez, Manuel ;
Suarez-Garaboa, Sonia ;
Lopez-Dequidt, Iria ;
Estany-Gestal, Ana ;
Sobrino, Tomas ;
Campos, Francisco ;
Castillo, Jose ;
Rodriguez-Yanez, Santiago ;
Iglesias-Rey, Ramon .
SCIENTIFIC REPORTS, 2021, 11 (01)