Predicting incident dementia in cerebral small vessel disease: comparison of machine learning and traditional statistical models

被引:3
作者
Li, Rui [1 ]
Harshfield, Eric L. [1 ,2 ]
Bell, Steven [1 ,2 ,3 ]
Burkhart, Michael [4 ]
Tuladhar, Anil M. [5 ]
Hilal, Saima [6 ,7 ,8 ]
Tozer, Daniel J. [1 ]
Chappell, Francesca M. [9 ]
Makin, Stephen D. J. [10 ]
Lo, Jessica W. [11 ]
Wardlaw, Joanna M. [9 ]
de Leeuw, Frank-Erik [5 ]
Chen, Christopher [6 ]
Kourtzi, Zoe [4 ]
Markus, Hugh S. [1 ,2 ]
机构
[1] Univ Cambridge, Dept Clin Neurosci, Stroke Res Grp, Cambridge, England
[2] Univ Cambridge, Heart & Lung Res Inst, Cambridge, England
[3] Univ Cambridge, Precis Breast Canc Inst, Dept Oncol, Cambridge, England
[4] Univ Cambridge, Dept Psychol, Adapt Brain Lab, Cambridge, England
[5] Radboud Univ Nijmegen Med Ctr, Donders Ctr Med Neurosci, Dept Neurol, Nijmegen, Netherlands
[6] Natl Univ Singapore, Memory Aging & Cognit Ctr, Yong Loo Lin Sch Med, Dept Pharmacol, Singapore, Singapore
[7] Natl Univ Singapore, Saw Swee Hock Sch Publ Hlth, Singapore, Singapore
[8] Natl Univ Hlth Syst, Singapore, Singapore
[9] Univ Edinburgh, Ctr Clin Brain Sci, Edinburgh, Scotland
[10] Univ Aberdeen, Inst Appl Hlth Sci, Ctr Rural Hlth, Aberdeen, Scotland
[11] Univ New South Wales, Ctr Hlth Brain Ageing, Sydney, Australia
来源
CEREBRAL CIRCULATION - COGNITION AND BEHAVIOR | 2023年 / 5卷
基金
英国医学研究理事会; 英国工程与自然科学研究理事会; 英国惠康基金;
关键词
Cerebral small vessel disease; Dementia; Prediction; Machine learning; COGNITIVE IMPAIRMENT; RISK; BIAS;
D O I
10.1016/j.cccb.2023.100179
中图分类号
R74 [神经病学与精神病学];
学科分类号
摘要
Background: Cerebral small vessel disease (SVD) contributes to 45% of dementia cases worldwide, yet we lack a reliable model for predicting dementia in SVD. Past attempts largely relied on traditional statistical approaches. Here, we investigated whether machine learning (ML) methods improved prediction of incident dementia in SVD from baseline SVD-related features over traditional statistical methods.Methods: We included three cohorts with varying SVD severity (RUN DMC, n = 503; SCANS, n = 121; HAR-MONISATION, n = 265). Baseline demographics, vascular risk factors, cognitive scores, and magnetic resonance imaging (MRI) features of SVD were used for prediction. We conducted both survival analysis and classification analysis predicting 3-year dementia risk. For each analysis, several ML methods were evaluated against standard Cox or logistic regression. Finally, we compared the feature importance ranked by different models.Results: We included 789 participants without missing data in the survival analysis, amongst whom 108 (13.7%) developed dementia during a median follow-up of 5.4 years. Excluding those censored before three years, we included 750 participants in the classification analysis, amongst whom 48 (6.4%) developed dementia by year 3. Comparing statistical and ML models, only regularised Cox/logistic regression outperformed their statistical counterparts overall, but not significantly so in survival analysis. Baseline cognition was highly predictive, and global cognition was the most important feature. Conclusions: When using baseline SVD-related features to predict dementia in SVD, the ML survival or classifi-cation models we evaluated brought little improvement over traditional statistical approaches. The benefits of ML should be evaluated with caution, especially given limited sample size and features.
引用
收藏
页数:8
相关论文
共 41 条
[31]   A comparison of machine learning methods for survival analysis of high-dimensional clinical data for dementia prediction [J].
Spooner, Annette ;
Chen, Emily ;
Sowmya, Arcot ;
Sachdev, Perminder ;
Kochan, Nicole A. ;
Trollor, Julian ;
Brodaty, Henry .
SCIENTIFIC REPORTS, 2020, 10 (01)
[32]   Stroke subtype, vascular risk factors, and total MRI brain small-vessel disease burden [J].
Staals, Julie ;
Makin, Stephen D. J. ;
Doubal, Fergus N. ;
Dennis, Martin S. ;
Wardlaw, Joanna M. .
NEUROLOGY, 2014, 83 (14) :1228-1234
[33]   Survival prediction models: an introduction to discrete-time modeling [J].
Suresh, Krithika ;
Severn, Cameron ;
Ghosh, Debashis .
BMC MEDICAL RESEARCH METHODOLOGY, 2022, 22 (01)
[34]   Nonlinear temporal dynamics of cerebral small vessel disease The RUN DMC study [J].
van Leijsen, Esther M. C. ;
van Uden, Ingeborg W. M. ;
Ghafoorian, Mohsen ;
Bergkamp, Mayra I. ;
Lohner, Valerie ;
Kooijmans, Eline C. M. ;
van der Holst, Helena M. ;
Tuladhar, Anil M. ;
Norris, David G. ;
van Dijk, Ewoud J. ;
Rutten-Jacobs, Loes C. A. ;
Platel, Bram ;
Klijn, Catharina J. M. ;
de Leeuw, Frank-Erik .
NEUROLOGY, 2017, 89 (15) :1569-1577
[35]   Diffusion Tensor Imaging of the Hippocampus Predicts the Risk of Dementia; the RUN DMC Study [J].
van Uden, I. W. M. ;
Tuladhar, A. M. ;
van der Holst, H. M. ;
van Leijsen, E. M. C. ;
van Norden, A. G. W. ;
de Laat, K. F. ;
Rutten-Jacobs, L. C. A. ;
Norris, D. G. ;
Claassen, J. A. H. R. ;
van Dijk, E. J. ;
Kessels, R. P. C. ;
de Leeuw, F-E .
HUMAN BRAIN MAPPING, 2016, 37 (01) :327-337
[36]   Baseline white matter microstructural integrity is not related to cognitive decline after 5 years: The RUN DMC study [J].
van Uden, I. W. M. ;
van der Holst, H. M. ;
Schaapsmeerders, P. ;
Tuladhar, A. M. ;
van Norden, A. G. W. ;
de Laat, K. F. ;
Norris, D. G. ;
Claassen, J. A. H. R. ;
van Dijk, E. J. ;
Richard, E. ;
Kessels, R. P. C. ;
de Leeuw, F-E. .
BBA CLINICAL, 2015, 4 :108-114
[37]   Dementia risk prediction in individuals with mild cognitive impairment: a comparison of Cox regression and machine learning models [J].
Wang, Meng ;
Greenberg, Matthew ;
Forkert, Nils D. ;
Chekouo, Thierry ;
Afriyie, Gabriel ;
Ismail, Zahinoor ;
Smith, Eric E. ;
Sajobi, Tolulope T. .
BMC MEDICAL RESEARCH METHODOLOGY, 2022, 22 (01)
[38]   Small vessel disease: mechanisms and clinical implications [J].
Wardlaw, Joanna M. ;
Smith, Colin ;
Dichgans, Martin .
LANCET NEUROLOGY, 2019, 18 (07) :684-696
[39]   Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration [J].
Wardlaw, Joanna M. ;
Smith, Eric E. ;
Biessels, Geert J. ;
Cordonnier, Charlotte ;
Fazekas, Franz ;
Frayne, Richard ;
Lindley, Richard I. ;
O'Brien, John T. ;
Barkhof, Frederik ;
Benavente, Oscar R. ;
Black, Sandra E. ;
Brayne, Carol ;
Breteler, Monique ;
Chabriat, Hugues ;
DeCarli, Charles ;
de Leeuw, Frank-Erik ;
Doubal, Fergus ;
Duering, Marco ;
Fox, Nick C. ;
Greenberg, Steven ;
Hachinski, Vladimir ;
Kilimann, Ingo ;
Mok, Vincent ;
van Oostenbrugge, Robert ;
Pantoni, Leonardo ;
Speck, Oliver ;
Stephan, Blossom C. M. ;
Teipel, Stefan ;
Viswanathan, Anand ;
Werring, David ;
Chen, Christopher ;
Smith, Colin ;
van Buchem, Mark ;
Norrving, Bo ;
Gorelick, Philip B. ;
Dichgans, Martin .
LANCET NEUROLOGY, 2013, 12 (08) :822-838
[40]   Bias and efficiency of multiple imputation compared with complete-case analysis for missing covariate values [J].
White, Ian R. ;
Carlin, John B. .
STATISTICS IN MEDICINE, 2010, 29 (28) :2920-2931