Enhancing Cardiovascular Risk Prediction: Development of an Advanced Xgboost Model with Hospital-Level Random Effects

Data de publicação:

Autores da FMUP

  • José Alberto Da Silva Freitas

    Autor

Participantes de fora da FMUP

  • Dong, Tim
  • Oronti, Iyabosola Busola
  • Sinha, Shubhra
  • Zhai, Bing
  • Chan, Jeremy
  • Fudulu, Daniel P.
  • Caputo, Massimo
  • Angelini, Gianni D.

Unidades de investigação

Abstract

Background: Ensemble tree-based models such as Xgboost are highly prognostic in cardiovascular medicine, as measured by the Clinical Effectiveness Metric (CEM). However, their ability to handle correlated data, such as hospital-level effects, is limited. Objectives: The aim of this work is to develop a binary-outcome mixed-effects Xgboost (BME) model that integrates random effects at the hospital level. To ascertain how well the model handles correlated data in cardiovascular outcomes, we aim to assess its performance and compare it to fixed-effects Xgboost and traditional logistic regression models. Methods: A total of 227,087 patients over 17 years of age, undergoing cardiac surgery from 42 UK hospitals between 1 January 2012 and 31 March 2019, were included. The dataset was split into two cohorts: training/validation (n = 157,196; 2012-2016) and holdout (n = 69,891; 2017-2019). The outcome variable was 30-day mortality with hospitals considered as the clustering variable. The logistic regression, mixed-effects logistic regression, Xgboost and binary-outcome mixed-effects Xgboost (BME) were fitted to both standardized and unstandardized datasets across a range of sample sizes and the estimated prediction power metrics were compared to identify the best approach. Results: The exploratory study found high variability in hospital-related mortality across datasets, which supported the adoption of the mixed-effects models. Unstandardized Xgboost BME demonstrated marked improvements in prediction power over the Xgboost model at small sample size ranges, but performance differences decreased as dataset sizes increased. Generalized linear models (glms) and generalized linear mixed-effects models (glmers) followed similar results, with the Xgboost models also excelling at greater sample sizes. Conclusions: These findings suggest that integrating mixed effects into machine learning models can enhance their performance on datasets where the sample size is small.

Dados da publicação

ISSN/ISSNe:
2306-5354, 2306-5354

BIOENGINEERING-BASEL  MDPI AG

Tipo:
Article
Páginas:
-
Link para outro recurso:
www.scopus.com

Documentos

  • Não há documentos

Métricas

Filiações mostrar / ocultar

Keywords

  • machine learning; AI; random effects; cardiovascular medicine; risk prediction; expectation-maximization; xgboost

Proyectos asociados

Stimulate continous monitoring in personal and physical health.

Investigador Principal: José Alberto da Silva Freitas

Estudo Observacional Académico (INNO4HEALTH) . FCT . 2021

Estudos de avaliação de exequibilidade, usabilidade e utilização de uma app para telemóvel para gestão da diabetes tipo 2.

Investigador Principal: José Alberto da Silva Freitas

Estudo Observacional Académico (FoodFriend) . FCT . 2022

Portuguese Public Hospitals Financial Performance between 2014-2020

Investigador Principal: José Alberto da Silva Freitas

Estudo Clínico Académico (Financial Performance) . 2023

Tendências nas Hospitalizações por Insuficiência Cardíaca durante um Período de Dezasseis Anos: Dados de Abrangência Nacional para Portugal

Investigador Principal: José Alberto da Silva Freitas

Estudo Clínico Académico (Hospitalizações IC) . 2022

The use of secondary data in Mental Health research

Investigador Principal: José Alberto da Silva Freitas

Estudo Clínico Académico . 2023

Health priorities in the European Union - a novel framework

Investigador Principal: José Alberto da Silva Freitas

Estudo Clínico Académico . 2023

Healthcare Human Resources and Quality Indicators: Approaches to Strengthening Primary Care.

Investigador Principal: José Alberto da Silva Freitas

Estudo Clínico Académico . 2022

A machine learning-based approach to support the assessment of clinical coded data quality in the context of Diagnosis-Related Groups classification systems

Investigador Principal: José Alberto da Silva Freitas

Estudo Clínico Académico . 2020

Citar a publicação

Partilhar a publicação