TY - GEN
T1 - Bayesian covariate selection in mixed-effects models for longitudinal shape analysis
AU - Muralidharan, Prasanna
AU - Fishbaugh, James
AU - Kim, Eun Young
AU - Johnson, Hans J.
AU - Paulsen, Jane S.
AU - Gerig, Guido
AU - Fletcher, P. Thomas
N1 - Funding Information:
This research was supported by NIH grants U01 NS082086, NS40068, NS050568 (PREDICTHD), U54 EB005149 (NA-MIC), S10 RR023392 (NCCR Shared Instrumentation Grant), NSF CAREER grant 1054057, NIH (NINDS; 5RO1NS040068, 5RO1NS054893) and the CHDI Foundation to Jane S Paulsen. We thank the PREDICT-HD sites, the study participants the National Research Roster for HD patients and Families, the Huntington Disease Society of America and the Huntington Study Group
Publisher Copyright:
© 2016 IEEE.
PY - 2016/6/15
Y1 - 2016/6/15
N2 - The goal of longitudinal shape analysis is to understand how anatomical shape changes over time, in response to biological processes, including growth, aging, or disease. In many imaging studies, it is also critical to understand how these shape changes are affected by other factors, such as sex, disease diagnosis, IQ, etc. Current approaches to longitudinal shape analysis have focused on modeling age-related shape changes, but have not included the ability to handle covariates. In this paper, we present a novel Bayesian mixed-effects shape model that incorporates simultaneous relationships between longitudinal shape data and multiple predictors or covariates to the model. Moreover, we place an Automatic Relevance Determination (ARD) prior on the parameters, that lets us automatically select which covariates are most relevant to the model based on observed data. We evaluate our proposed model and inference procedure on a longitudinal study of Huntington's disease from PREDICT-HD. We first show the utility of the ARD prior for model selection in a univariate modeling of striatal volume, and next we apply the full high-dimensional longitudinal shape model to putamen shapes.
AB - The goal of longitudinal shape analysis is to understand how anatomical shape changes over time, in response to biological processes, including growth, aging, or disease. In many imaging studies, it is also critical to understand how these shape changes are affected by other factors, such as sex, disease diagnosis, IQ, etc. Current approaches to longitudinal shape analysis have focused on modeling age-related shape changes, but have not included the ability to handle covariates. In this paper, we present a novel Bayesian mixed-effects shape model that incorporates simultaneous relationships between longitudinal shape data and multiple predictors or covariates to the model. Moreover, we place an Automatic Relevance Determination (ARD) prior on the parameters, that lets us automatically select which covariates are most relevant to the model based on observed data. We evaluate our proposed model and inference procedure on a longitudinal study of Huntington's disease from PREDICT-HD. We first show the utility of the ARD prior for model selection in a univariate modeling of striatal volume, and next we apply the full high-dimensional longitudinal shape model to putamen shapes.
KW - Bayesian analysis
KW - Huntington's disease
KW - Longitudinal shape analysis
KW - model selection
UR - http://www.scopus.com/inward/record.url?scp=84978388458&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84978388458&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2016.7493352
DO - 10.1109/ISBI.2016.7493352
M3 - Conference contribution
AN - SCOPUS:84978388458
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 656
EP - 659
BT - 2016 IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
T2 - 2016 IEEE 13th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2016
Y2 - 13 April 2016 through 16 April 2016
ER -