Comprehensive Analysis of Factors Influencing Heart Disease Risk DOI Creative Commons
Qi Wu

Highlights in Science Engineering and Technology, Год журнала: 2024, Номер 123, С. 124 - 130

Опубликована: Дек. 24, 2024

Research Background and Significance: Heart disease continues to be a leading cause of mortality globally, posing significant challenges in the realms prevention management. The complexity cardiovascular diseases arises from an interplay genetic, lifestyle, environmental factors, making their study prediction critically important for public health. integration machine learning techniques into medical research has opened new avenues understanding these diseases, significantly advancing capabilities predictive models. This paper leverages comprehensive dataset Kaggle, incorporating diverse range variables such as lifestyle habits physiological markers known influence health, which provides foundation robust analytical exploration. Contributions Paper: makes several contributions field research. Firstly, it employs advanced statistical Principal Component Analysis (PCA) K-means clustering effectively reduce data multicollinearity dimensionality, enhances clarity reliability findings. PCA approach successfully condensed principal components that explain substantial portion variability, while categorized meaningful risk profiles. Secondly, this demonstrates utility factor analysis identifying major factors like smoking, age, gender, furthering roles heart risk. Finally, application various

Язык: Английский

Spline Estimator in Nonparametric Ordinal Logistic Regression Model for Predicting Heart Attack Risk DOI Open Access
Nur Chamidah,

Budi Lestari,

Hendri Susilo

и другие.

Symmetry, Год журнала: 2024, Номер 16(11), С. 1440 - 1440

Опубликована: Окт. 30, 2024

In Indonesia, one of the main causes death for both young and elderly people is heart attacks, cause attacks non-communicable diseases such as hypertension. Deaths due to caused by diseases, namely hypertension, rank first in Indonesia. Therefore, predictions risk having a attack hypertension need serious attention. Further, determining whether patient experiencing attack, an effective method prediction required. One efficient approach use statistical models. This study discusses predicting via modeling classifying based on factors that influence it, namely, age, cholesterol levels, triglyceride levels using spline estimator Nonparametric Ordinal Logistic Regression (NOLR) model. this study, we assume ordinal scale response variable with q categories have asymmetric distribution, multinomial distribution. The data used are secondary from medical records cardiac poly patients at Haji General Hospital Surabaya, results show proposed model has greatest classification accuracy sensitivity values compared NOLR GAM, classical approach, Parametric (POLR) means suitable risks. Also, estimated LS-Spline obtained valid value 85% 100%.

Язык: Английский

Процитировано

1
An Analysis and Prediction of Health Insurance Costs Using Machine Learning-Based Regressor Techniques DOI Open Access

Gagan Kumar Patra,

Chandrababu Kuraku,

Siddharth Konkimalla

и другие.

Journal of Data Analysis and Information Processing, Год журнала: 2024, Номер 12(04), С. 581 - 596

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

1
Comprehensive Analysis of Factors Influencing Heart Disease Risk DOI Creative Commons
Qi Wu

Highlights in Science Engineering and Technology, Год журнала: 2024, Номер 123, С. 124 - 130

Опубликована: Дек. 24, 2024

Research Background and Significance: Heart disease continues to be a leading cause of mortality globally, posing significant challenges in the realms prevention management. The complexity cardiovascular diseases arises from an interplay genetic, lifestyle, environmental factors, making their study prediction critically important for public health. integration machine learning techniques into medical research has opened new avenues understanding these diseases, significantly advancing capabilities predictive models. This paper leverages comprehensive dataset Kaggle, incorporating diverse range variables such as lifestyle habits physiological markers known influence health, which provides foundation robust analytical exploration. Contributions Paper: makes several contributions field research. Firstly, it employs advanced statistical Principal Component Analysis (PCA) K-means clustering effectively reduce data multicollinearity dimensionality, enhances clarity reliability findings. PCA approach successfully condensed principal components that explain substantial portion variability, while categorized meaningful risk profiles. Secondly, this demonstrates utility factor analysis identifying major factors like smoking, age, gender, furthering roles heart risk. Finally, application various

Язык: Английский

Процитировано

0