Probability Analysis of Hypertension-Related Symptoms Based on XGBoost and Clustering Algorithm

Abstract

In this paper, cluster analysis and the XGBoost method are used to analyze the related symptoms of various types of young hypertensive patients, and finally guide patients to target treatment. Hypertension is a chronic disease that is common worldwide. The incidence of it is increasing, and the age level of patients is decreasing year by year. Effective treatment of youth hypertension has become a problem in the world. In this paper, young hypertension patients are classified into two groups by cluster analysis; the proportion of different hypertension related symptoms in each group of patients is then counted; and after verifying the prediction accuracy of the XGBoost model with 10-fold cross-validation, the accuracy of clustering is calculated by the XGBoost method. The final result shows that there are significant differences in symptomatic entropy between patients with type II hypertension and those with type I hypertension. Patients with type II hypertension are more likely to have symptoms of ventricular hypertrophy and microalbuminuria. Through this analysis, patients can have preventive treatment according to their own situation, and this can reduce the burden of medical expenses and prevent major diseases. Applying the data analysis into the medical field has great practical significance.

Early Stage Prediction of US County Vulnerability to the COVID-19 Pandemic

“…Additionally, the underlying relationship between variables was assumed to be non-linear. For such cases the literature supports [40][41][42][43][44][45][46][47] using gradient tree boosting and deep learning methods for better prediction results.…”

Exploratory Study of Some Machine Learning Techniques to Classify the Patient Treatment

“…Since its inception [29], the Extreme Gradient Boost (XGBoost) has been the favourite technique used to address the challenge of classification prediction in the medical area. By utilizing XGBoost, many researchers addressed various subtopic in the health/medical field such as diagnoses [5], [8], [30]- [34], related diseases [4] [6], [9], [34], [35], medical treatment [36], [37], patient status [12], [14], [16], [38], or event genomic [39].…”

“…A research conducted an experimental study using XGBoost Classifiers with some scenarios such as transformation, resampling, clustering, and ensemble learning to predict the diagnosis of second primary cancers (SPCs) [30] The resampling and clustering strategies were used to determine the best method used to identify some important risk factors associated with SPCs in patients with breast cancer. The combination of the XGBoost and Clustering analysis approach was also proposed by [8] to predict the hypertensionrelated symptoms from 531 hypertensive patients data in a hospital in Beijing. These combination techniques showed that there are significant differences in symptomatic entropy between patients with type II and type I hypertension.…”

“…The system is available as an open-source package and widely recognized in many challenges of machine learning and data mining. Based on a study conducted by [8], some of the advantages of XGBoost are as follows, supports linear classifier, regularization to control the model complexity, capability in using the second order of Taylor expansion, and some more.…”

Data Analytics in Smart Healthcare: The Recent Developments and Beyond

“…W. Change et al in their manuscript entitled "Probability analysis of hypertension-related symptoms based on XGBoost and clustering algorithm" [16] proposed a clustering based XGBoost algorithm to classify type I and type II hypertension. Moreover, studies showed that symptoms of ventricular hypertrophy, arteriosclerosis and microalbuminuria are more prone to occur for type II hypertension sufferers.…”