A Extreme Gradient Boosting Classifier for Predicting Chronic Kidney Disease Stages

Scoralick, João P.; Iwashima, Gabriele C.; Colugnati, Fernando Antônio Basile; Goliatt, Leonardo; Capriles, Priscila V. S. Z.

doi:10.1007/978-3-030-71187-0_83

Cited by 4 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The problem of predicting the CVD risk (i.e., categorizing individuals as low and high risk of CVD) is modeled as a supervised classification problem. Recent research has demonstrated the capability of decision trees (DT), random forest (RF), extreme gradient boosting (XGBoost), and deep neural networks (DNN) to accurately predict binary variables in the healthcare domain, such as disease risk 47 , heart disease 48 , and mortality risk 49 . Therefore, we employed and evaluated these four ML models for CVD risk classification.…”

Section: Methodsmentioning

confidence: 99%

Explainable machine learning framework for predicting long-term cardiovascular disease risk among adolescents

Salah

Srinivas

2022

Sci Rep

View full text Add to dashboard Cite

Although cardiovascular disease (CVD) is the leading cause of death worldwide, over 80% of it is preventable through early intervention and lifestyle changes. Most cases of CVD are detected in adulthood, but the risk factors leading to CVD begin at a younger age. This research is the first to develop an explainable machine learning (ML)-based framework for long-term CVD risk prediction (low vs. high) among adolescents. This study uses longitudinal data from a nationally representative sample of individuals who participated in the Add Health study. A total of 14,083 participants who completed relevant survey questionnaires and health tests from adolescence to young adulthood were chosen. Four ML classifiers [decision tree (DT), random forest (RF), extreme gradient boosting (XGBoost), and deep neural networks (DNN)] and 36 adolescent predictors are used to predict adulthood CVD risk. While all ML models demonstrated good prediction capability, XGBoost achieved the best performance (AUC-ROC: 84.5% and AUC-PR: 96.9% on testing data). Besides, critical predictors of long-term CVD risk and its impact on risk prediction are obtained using an explainable technique for interpreting ML predictions. The results suggest that ML can be employed to detect adulthood CVD very early in life, and such an approach may facilitate primordial prevention and personalized intervention.

show abstract

Section: Methodsmentioning

confidence: 99%

Explainable machine learning framework for predicting long-term cardiovascular disease risk among adolescents

Salah

Srinivas

2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The problem of predicting the CVD risk (i.e., categorizing individuals as low and high risk of CVD) is modeled as a supervised classification problem. Recent research has demonstrated the capability of decision trees (DT), random forest (RF), extreme gradient boosting (XGBoost), and deep neural networks (DNN) to accurately predict binary variables in the healthcare domain, such as disease risk [177][178][179], heart disease [180][181][182][183], and mortality risk [184][185][186]. Therefore, we employed and evaluated these four ML algorithms for CVD risk classification.…”

Section: Algorithms and Evaluationmentioning

confidence: 99%

Improving cardiac care delivery using predictive and prescriptive analytics

Salah¹

View full text Add to dashboard Cite

[EMBARGOED UNTIL 6/1/2023] Endured by the growing prevalence of cardiovascular diseases (CVDs), the demand for cardiac care services has increased. On the other hand, the supply of cardiologists is expected to be insufficient to meet this growing demand. Considering this imbalance between demand and capacity, cardiology clinics strive to improve their services. Thus, this dissertation proposes two approaches for improving cardiac care services through the application of predictive and prescriptive analytics. The first approach develops an efficient appointment system (AS) that allocates patients' demand during the clinic session effectively to improve resource utilization and patient satisfaction. The proposed AS also addresses the problem of clinical uncertainties, such as patient no-shows and service-time variability, which adversely impact AS efficiency by developing a predict-then-schedule framework. In the predict step, patient-specific no-show risk and service duration are estimated using machine learning (ML) models. The schedule step determines the appointment time and interval for each patient using a sequential AS that leverages the ML predictions. In addition, four new ML-enabled sequencing rules are proposed. The proposed approach and sequencing rules are validated using real clinical data. Besides, the effectiveness of integrating ML-based uncertainty predictions into the AS design is also evaluated for 32 different clinic environments. Results indicate that an AS design adopting the predict-then-schedule approach always dominates the conventional system and could improve the efficiency by 60 percent. The new sequencing rules can improve the AS performance by up to 40 percent when compared to the existing policies. Finally, several managerial insights on sequencing and overbooking are also provided. On the other hand, the second approach develops an ML-based model to predict the long-term CVD risk that can aid in the early detection of CVD. Unlike the existing tools for CVD risk assessment which are only applicable to adults and use cross-sectional data. This research provides the first long-term ML-based CVD risk prediction model among adolescents based on a longitudinal dataset. Our results indicated the capability of ML models to accurately predict the long-term risk of CVD among adolescents. In addition, the most significant factors for predicting CVD risk among adolescents are identified. Furthermore, the proposed model can be used to identify individuals who are at high risk of developing CVD early in life and provide them with the necessary guidance and preventive treatment, which improves the quality of life, lower healthcare costs, and reduces the demand for cardiac care services later.

show abstract

Unsupervised Analysis of Clinical and Laboratory Parameters of Chronic Kidney Disease

Goliatt,

Capriles,

Iwashima

et al. 2024

Intelligent Systems Design and Applications

View full text Add to dashboard Cite

A Extreme Gradient Boosting Classifier for Predicting Chronic Kidney Disease Stages

Cited by 4 publications

References 22 publications

Explainable machine learning framework for predicting long-term cardiovascular disease risk among adolescents

Explainable machine learning framework for predicting long-term cardiovascular disease risk among adolescents

Improving cardiac care delivery using predictive and prescriptive analytics

Unsupervised Analysis of Clinical and Laboratory Parameters of Chronic Kidney Disease

Contact Info

Product

Resources

About