In the study of Alzheimer’s disease, researchers often collect repeated measurements of clinical variables, event history, and functional data. If the health measurements deteriorate rapidly, patients may reach a level of cognitive impairment and are diagnosed as having dementia. An accurate prediction of the time to dementia based on the information collected is helpful for physicians to monitor patients’ disease progression and to make early informed medical decisions. In this article, we first propose a functional joint model to account for functional predictors in both longitudinal and survival submodels in the joint modeling framework. We then develop a Bayesian approach for parameter estimation and a dynamic prediction framework for predicting the subjects’ future outcome trajectories and risk of dementia, based on their scalar and functional measurements. The proposed Bayesian functional joint model provides a flexible framework to incorporate many features both in joint modeling of longitudinal and survival data and in functional data analysis. Our proposed model is evaluated by a simulation study and is applied to the motivating Alzheimer’s Disease Neuroimaging Initiative study.
FaucettCLThomasDC. Simultaneously modelling censored survival data and repeatedly measured covariates: a Gibbs sampling approach. Stat Med1996; 15: 1663–1685.
2.
WulfsohnMSTsiatisAA. A joint model for survival and longitudinal data measured with error. Biometrics1997; 53: 330–339.
3.
RizopoulosD. Dynamic predictions and prospective accuracy in joint models for longitudinal and time-to-event data. Biometrics2011; 67: 819–829.
4.
TaylorJMParkYAnkerstDPet al.Real-time individual predictions of prostate cancer recurrence using joint models. Biometrics2013; 69: 206–213.
5.
RamsayJODalzellCJ. Some tools for functional data analysis. J Roy Stat Soc Ser B (Methodol)1991; 53: 539–572.
6.
WangJLChiouJMMüllerHG. Functional data analysis. Annu Rev Stat Appl2016; 3: 257–295.
7.
MorrisJS. Functional regression. Annu Rev Stat Appl2015; 2: 321–359.
8.
CardotHFerratyFSardaP. Spline estimators for the functional linear model. Stat Sin2003; 13: 571–591.
9.
JamesGM. Generalized linear models with functional predictors. J Roy Stat Soc: Ser B (Stat Methodol)2002; 64: 411–432.
10.
MüllerHGStadtmüllerU. Generalized functional linear models. Ann Stat2005; 33: 774–805.
11.
ReissPTOgdenRT. Functional principal component regression and functional partial least squares. J Am Stat Assoc2007; 102: 984–996.
12.
JamesGMWangJZhuJ. Functional linear regression that’s interpretable. Ann Stat2009; 37: 2083–2108.
GoldsmithJCrainiceanuCMCaffoBet al.Longitudinal penalized functional regression for cognitive outcomes on neuronal tract measurements. J Roy Stat Soc: Ser C (Appl Stat)2012; 61: 453–469.
15.
GertheissJGoldsmithJCrainiceanuCet al.Longitudinal scalar-on-functions regression with application to tractography data. Biostatistics2013; 14: 447–461.
16.
GellarJEColantuoniENeedhamDMet al.Cox regression models with functional covariates for survival data. Stat Model2015; 15: 256–278.
17.
LeeEZhuHKongDet al.BFLCRM: A Bayesian functional linear Cox regression model for predicting time to conversion to Alzheimer’s disease. Ann Appl Stat2015; 9: 2153–2178.
18.
WeinerMWVeitchDPAisenPSet al.2014 Update of the Alzheimer’s Disease Neuroimaging Initiative: A review of papers published since its inception. Alzheimers Dement2015; 11: e1–e120.
19.
LiKChanWDoodyRSet al.Prediction of conversion to Alzheimers disease with longitudinal measures and time-to-event data. J Alzheimers Dis2017; 58: 361–371.
20.
ClevelandWS. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc1979; 74: 829–836.
21.
DigglePKenwardMG. Informative drop-out in longitudinal data analysis. J Roy Stat Soc Ser C (Appl Stat)1994; 43: 49–93.
22.
LoRYHubbardAEShawLMet al.Longitudinal change of biomarkers in cognitive decline. Archiv Neurol2011; 68: 1257–1266.
23.
ZhangDShenDInitiativeADNet al.Predicting future clinical changes of MCI patients using longitudinal and multimodal biomarkers. PLOS One2012; 7: e33182–e33182.
24.
DuAT. Magnetic resonance imaging of the entorhinal cortex and hippocampus in mild cognitive impairment and Alzheimer’s disease. J Neurol Neurosurg Psychiatry2001; 71: 441–447.
25.
CuiYLiuBLuoSet al.Identification of conversion from mild cognitive impairment to Alzheimer’s disease using multivariate predictors. PLOS One2011; 6: e21896–e21896.
26.
LiSOkonkwoOAlbertMet al.Variation in variables that predict progression from MCI to AD dementia over duration of follow-up. Am J Alzheimers Dis2013; 2: 12–28.
27.
ApostolovaLGMosconiLThompsonPMet al.Subregional hippocampal atrophy predicts Alzheimer’s dementia in the cognitively normal. Neurobiol Aging2010; 31: 1077–1088.
28.
QiuAFennema-NotestineCDaleAMet al.Regional shape abnormalities in mild cognitive impairment and Alzheimer’s disease. NeuroImage2009; 45: 656–661.
29.
RuppertD. Selecting the number of knots for penalized splines. J Comput Graph Stat2002; 11: 735–757.
30.
StaniswalisJGLeeJJ. Nonparametric regression analysis of longitudinal data. J Am Stat Assoc1998; 93: 1403–1418.
31.
YaoFMüllerHGCliffordAJet al.Shrinkage estimation for functional principal component scores with application to the population kinetics of plasma folate. Biometrics2003; 59: 676–685.
32.
Stefan LangAB. Bayesian P-splines. J Comput Graph Stat2004; 13: 183–212.
33.
HoffmanMDGelmanA. The No-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo. J Mach Learn Res2014; 15: 1593–1623.
34.
Gelman A, Carlin JB, Stern HS, et al. Bayesian data analysis. Boca Raton, FL: CRC Press, 2013.
35.
GilksWRBestNGTanKKC. Adaptive rejection Metropolis sampling within Gibbs sampling. J Roy Stat Soc Ser C (Appl Stat)1995; 44: 455–472.
36.
Li L, Greene T and Hu B. A simple method to estimate the time-dependent receiver operating characteristic curve and the area under the curve with right censored data. Stat Methods Med Res 2016; 0: 0962280216680239.
37.
HeagertyPJLumleyTPepeMS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics2000; 56: 337–344.
38.
BlanchePProust-LimaCLoubreLet al.Quantifying and comparing dynamic predictive accuracy of joint models for longitudinal marker and time-to-event in presence of censoring and competing risks. Biometrics2015; 71: 102–113.
39.
SneMTaylorJMGDignamJJet al.Individualized dynamic prediction of prostate cancer recurrence with and without the initiation of a second treatment: development and validation. Stat Methods Med Res2016; 25: 2972–2991.
40.
Crainiceanu CM, Reiss PT, Goldsmith J, et al. Refund: Regression with functional data, http://CRAN.R-project.org/package=refund, r package version 0.1-6 (2013).
41.
PengJPaulD. A geometric approach to maximum likelihood estimation of the functional principal components from sparse longitudinal data. J Comput Graph Stat2009; 18: 995–1015.
42.
RisacherSLSaykinAJWesJDet al.Baseline MRI predictors of conversion from MCI to probable AD in the ADNI cohort. Curr Alzheimer Res2009; 6: 347–361.
43.
CorderESaundersAStrittmatterWet al.Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science1993; 261: 921–923.
44.
ApostolovaLGDuttonRADinovIDet al.Conversion of mild cognitive impairment to Alzheimer disease predicted by hippocampal atrophy maps. Archiv Neurol2006; 63: 693–699.
45.
GrevenSCrainiceanuCCaffoBet al.Longitudinal functional principal component analysis. Electron J Stat2010; 4: 1022–1054.
46.
RizopoulosDHatfieldLACarlinBPet al.Combining dynamic predictions from joint models for longitudinal and time-to-event data using Bayesian model averaging. J Am Stat Assoc2014; 109: 1385–1397.
47.
YangLYuMGaoS. Prediction of coronary artery disease risk based on multiple longitudinal biomarkers. Stat Med2016; 35: 1299–1314.
48.
Wang X, Guo F, Heller KA, et al. Parallelizing MCMC with random partition trees. In: Cortes C, Lawrence ND and Lee DD (eds) Advances in neural information processing systems 28. Red Hook, NY: Curran Associates, Inc., 2015, pp.451–459.
49.
OrmerodJTWandMP. Explaining variational approximations. Am Stat2010; 64: 140–153.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
