Back to Search Start Over

Survival Analysis in Cognitively Normal Subjects and in Patients with Mild Cognitive Impairment Using a Proportional Hazards Model with Extreme Gradient Boosting Regression.

Authors :
Khajehpiri B
Moghaddam HA
Forouzanfar M
Lashgari R
Ramos-Cejudo J
Osorio RS
Ardekani BA
Source :
Journal of Alzheimer's disease : JAD [J Alzheimers Dis] 2022; Vol. 85 (2), pp. 837-850.
Publication Year :
2022

Abstract

Background: Evaluating the risk of Alzheimer's disease (AD) in cognitively normal (CN) and patients with mild cognitive impairment (MCI) is extremely important. While MCI-to-AD progression risk has been studied extensively, few studies estimate CN-to-MCI conversion risk. The Cox proportional hazards (PH), a widely used survival analysis model, assumes a linear predictor-risk relationship. Generalizing the PH model to more complex predictor-risk relationships may increase risk estimation accuracy.<br />Objective: The aim of this study was to develop a PH model using an Xgboost regressor, based on demographic, genetic, neuropsychiatric, and neuroimaging predictors to estimate risk of AD in patients with MCI, and the risk of MCI in CN subjects.<br />Methods: We replaced the Cox PH linear model with an Xgboost regressor to capture complex interactions between predictors, and non-linear predictor-risk associations. We endeavored to limit model inputs to noninvasive and more widely available predictors in order to facilitate future applicability in a wider setting.<br />Results: In MCI-to-AD (nā€Š=ā€Š882), the Xgboost model achieved a concordance index (C-index) of 84.5%. When the model was used for MCI risk prediction in CN (nā€Š=ā€Š100) individuals, the C-index was 73.3%. In both applications, the C-index was statistically significantly higher in the Xgboost in comparison to the Cox PH model.<br />Conclusion: Using non-linear regressors such as Xgboost improves AD dementia risk assessment in CN and MCI. It is possible to achieve reasonable risk stratification using predictors that are relatively low-cost in terms of time, invasiveness, and availability. Future strategies for improving AD dementia risk estimation are discussed.

Details

Language :
English
ISSN :
1875-8908
Volume :
85
Issue :
2
Database :
MEDLINE
Journal :
Journal of Alzheimer's disease : JAD
Publication Type :
Academic Journal
Accession number :
34864679
Full Text :
https://doi.org/10.3233/JAD-215266