A Comparison of Machine Learning Methods for Survival Analysis of High-dimensional Clinical Data for Dementia Prediction

Overview

Journal Sci Rep

Specialty Science

Date 2020 Nov 24

PMID 33230128

Citations 85

Authors

Annette Spooner

Emily Chen

Arcot Sowmya

Perminder Sachdev

Nicole A Kochan

Julian Trollor

Henry Brodaty

Affiliations

Soon will be listed here.

Abstract

Data collected from clinical trials and cohort studies, such as dementia studies, are often high-dimensional, censored, heterogeneous and contain missing information, presenting challenges to traditional statistical analysis. There is an urgent need for methods that can overcome these challenges to model this complex data. At present there is no cure for dementia and no treatment that can successfully change the course of the disease. Machine learning models that can predict the time until a patient develops dementia are important tools in helping understand dementia risks and can give more accurate results than traditional statistical methods when modelling high-dimensional, heterogeneous, clinical data. This work compares the performance and stability of ten machine learning algorithms, combined with eight feature selection methods, capable of performing survival analysis of high-dimensional, heterogeneous, clinical data. We developed models that predict survival to dementia using baseline data from two different studies. The Sydney Memory and Ageing Study (MAS) is a longitudinal cohort study of 1037 participants, aged 70-90 years, that aims to determine the effects of ageing on cognition. The Alzheimer's Disease Neuroimaging Initiative (ADNI) is a longitudinal study aimed at identifying biomarkers for the early detection and tracking of Alzheimer's disease. Using the concordance index as a measure of performance, our models achieve maximum performance values of 0.82 for MAS and 0.93 For ADNI.

Citing Articles

How predictive medicine leads to solidarity gaps in health.

Braun M NPJ Digit Med. 2025; 8(1):111.

PMID: 39966662 PMC: 11836224. DOI: 10.1038/s41746-025-01497-2.

Developing practical machine learning survival models to identify high-risk patients for in-hospital mortality following traumatic brain injury.

Andishgar A, Rismani M, Bazmi S, Mohammadi Z, Hooshmandi S, Kian B Sci Rep. 2025; 15(1):5913.

PMID: 39966448 PMC: 11836111. DOI: 10.1038/s41598-025-89574-0.

Identifying risk factors for Alzheimer's disease from multivariate longitudinal clinical data using temporal pattern mining.

Spooner A, Mohammadi G, Sachdev P, Brodaty H, Sowmya A BMC Bioinformatics. 2025; 26(1):56.

PMID: 39962385 PMC: 11834509. DOI: 10.1186/s12859-024-06018-8.

Comparison of time-to-event machine learning models in predicting biliary complication and mortality rate in liver transplant patients.

Andishgar A, Bazmi S, Lankarani K, Taghavi S, Imanieh M, Sivandzadeh G Sci Rep. 2025; 15(1):4768.

PMID: 39922959 PMC: 11807176. DOI: 10.1038/s41598-025-89570-4.

Survival machine learning methods for mortality prediction after heart transplantation in the contemporary era.

Liou L, Mostofsky E, Lehman L, Salia S, Barrera F, Wei Y PLoS One. 2025; 20(1):e0313600.

PMID: 39775253 PMC: 11706460. DOI: 10.1371/journal.pone.0313600.

References

Sachdev P, Brodaty H, Reppermund S, Kochan N, Trollor J, Draper B . The Sydney Memory and Ageing Study (MAS): methodology and baseline medical and neuropsychiatric characteristics of an elderly epidemiological non-demented cohort of Australians aged 70-90 years. Int Psychogeriatr. 2010; 22(8):1248-64. DOI: 10.1017/S1041610210001067. View

McLaughlin N, Westervelt H . Odor identification deficits in frontotemporal dementia: a preliminary study. Arch Clin Neuropsychol. 2007; 23(1):119-23. DOI: 10.1016/j.acn.2007.07.008. View

Harrell Jr F, Califf R, Pryor D, Lee K, Rosati R . Evaluating the yield of medical tests. JAMA. 1982; 247(18):2543-6. View

Nakagawa T, Ishida M, Naito J, Nagai A, Yamaguchi S, Onoda K . Prediction of conversion to Alzheimer's disease using deep survival analysis of MRI images. Brain Commun. 2020; 2(1):fcaa057. PMC: 7425528. DOI: 10.1093/braincomms/fcaa057. View

Heffernan M, Mather K, Xu J, Assareh A, Kochan N, Reppermund S . Alcohol Consumption and Incident Dementia: Evidence from the Sydney Memory and Ageing Study. J Alzheimers Dis. 2016; 52(2):529-38. DOI: 10.3233/JAD-150537. View

Weiner M, Aisen P, Jack Jr C, Jagust W, Trojanowski J, Shaw L . The Alzheimer's disease neuroimaging initiative: progress report and future plans. Alzheimers Dement. 2010; 6(3):202-11.e7. PMC: 2927112. DOI: 10.1016/j.jalz.2010.03.007. View

Steele A, Denaxas S, Shah A, Hemingway H, Luscombe N . Machine learning models in electronic health records can outperform conventional survival models for predicting patient mortality in coronary artery disease. PLoS One. 2018; 13(8):e0202344. PMC: 6118376. DOI: 10.1371/journal.pone.0202344. View

Polsterl S, Conjeti S, Navab N, Katouzian A . Survival analysis for high-dimensional, heterogeneous medical data: Exploring feature extraction as an alternative to feature selection. Artif Intell Med. 2016; 72:1-11. DOI: 10.1016/j.artmed.2016.07.004. View

Dietterich . Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms. Neural Comput. 1998; 10(7):1895-1923. DOI: 10.1162/089976698300017197. View

10.

Cui Y, Sachdev P, Lipnicki D, Jin J, Luo S, Zhu W . Predicting the development of mild cognitive impairment: a new use of pattern recognition. Neuroimage. 2012; 60(2):894-901. DOI: 10.1016/j.neuroimage.2012.01.084. View

11.

Leger S, Zwanenburg A, Pilz K, Lohaus F, Linge A, Zophel K . A comparative study of machine learning methods for time-to-event survival data for radiomics risk modelling. Sci Rep. 2017; 7(1):13206. PMC: 5643429. DOI: 10.1038/s41598-017-13448-3. View

12.

Rathore S, Habes M, Iftikhar M, Shacklett A, Davatzikos C . A review on neuroimaging-based classification studies and associated feature extraction methods for Alzheimer's disease and its prodromal stages. Neuroimage. 2017; 155():530-548. PMC: 5511557. DOI: 10.1016/j.neuroimage.2017.03.057. View

13.

De Bin R, Sauerbrei W, Boulesteix A . Investigating the prediction ability of survival models based on both clinical and omics data: two case studies. Stat Med. 2014; 33(30):5310-29. DOI: 10.1002/sim.6246. View

14.

Connors M, Sachdev P, Kochan N, Xu J, Draper B, Brodaty H . Cognition and mortality in older people: the Sydney Memory and Ageing Study. Age Ageing. 2015; 44(6):1049-54. DOI: 10.1093/ageing/afv139. View

15.

Bovelstad H, Nygard S, Storvold H, Aldrin M, Borgan O, Frigessi A . Predicting survival from microarray data--a comparative study. Bioinformatics. 2007; 23(16):2080-7. DOI: 10.1093/bioinformatics/btm305. View

16.

Buuren S, Boshuizen H, Knook D . Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med. 1999; 18(6):681-94. DOI: 10.1002/(sici)1097-0258(19990330)18:6<681::aid-sim71>3.0.co;2-r. View

17.

Maroco J, Silva D, Rodrigues A, Guerreiro M, Santana I, de Mendonca A . Data mining methods in the prediction of Dementia: A real-data comparison of the accuracy, sensitivity and specificity of linear discriminant analysis, logistic regression, neural networks, support vector machines, classification trees and random.... BMC Res Notes. 2011; 4:299. PMC: 3180705. DOI: 10.1186/1756-0500-4-299. View

18.

Schemper M, Kaider A, Wakounig S, Heinze G . Estimating the correlation of bivariate failure times under censoring. Stat Med. 2013; 32(27):4781-90. DOI: 10.1002/sim.5874. View

19.

Selkoe D . Preventing Alzheimer's disease. Science. 2012; 337(6101):1488-92. DOI: 10.1126/science.1228541. View

20.

Witten D, Tibshirani R . Survival analysis with high-dimensional covariates. Stat Methods Med Res. 2009; 19(1):29-51. PMC: 4806549. DOI: 10.1177/0962280209105024. View