Bayesian scalar-on-tensor regression using the Tucker decomposition for sparse spatial modeling

Modeling with multidimensional arrays, or tensors, often presents a problem due to high dimensionality. In addition, these structures typically exhibit inherent sparsity, requiring the use of regularization methods to properly characterize an association between a tensor covariate and a scalar response. We propose a Bayesian method to efficiently model a scalar response with a tensor covariate using the Tucker tensor decomposition in order to retain the spatial relationship within a tensor coefficient, while reducing the number of parameters varying within the model and applying regularization methods. Simulated data are analyzed to compare the model to recently proposed methods. A neuroimaging analysis using data from the Alzheimer's Data Neuroimaging Initiative shows improved inferential performance compared with other tensor regression methods. Bayesian analysis; tensor decomposition; image analysis; spatial statistics; statistical modeling.

Authors

Institutions

Publication Details

Journal
Biostatistics
Published
2024-12-31
DOI
https://doi.org/10.1093/biostatistics/kxaf029
Primary Topic
Tensor decomposition and applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Bayesian scalar-on-tensor regression using the Tucker decomposition for sparse spatial modeling

J H Karlawish, Rajarshi Guhaniyogi, Aaron Jacobson, Choe Ma et al.
Biostatistics
Tensor decomposition and applications
article

Bayesian scalar-on-tensor regression using the Tucker decomposition for sparse spatial modeling

J H Karlawish, Rajarshi Guhaniyogi, Aaron Jacobson, Choe Ma, Ruchi Aggarwal, D. T. Sacrey, Michaël Schumacher, Daniel Spencer, Judith Neugroschl, H. Klemp, C. W. McCurry, Karen Dagerman, Y Castillo Avila, Marissa Natelson-Love, Toga Aw, Elizabeth Trejo, Bryan L. Smelser, Payam H. Mahboubi, Silverberg Nb, Cecily Jenkins, K A Nudelman, M Ashford, Lesley K. Rao, Kristin Harkins, D Flenniken, J. Horenkamp, McKinley Ed, Jagust Wj, Khachaturian Zs, Haley Bernhardt, K Lacy, L. P. Teodoro, Conant Ca, Sheppard Fab, Alexandra Groome, Joshua D. Grill, Cyrus A. Raji, Hergesheimer Lh, Cynthia Stover, Amber Watkins, I Hoang, Lora Gallagher, D. Tosun-Turgut, J. Lagarde, Molchan Se, Connie Mayo, Veronika Solomon, M Nguyen Thao, Melissa Yu, Swarm Rl, S MCKINZIE, Sandra Talavera, Matoush Lo, Hegedus Sa, Lisa C. Silbert, Steven MacDonald, Claire M. Erickson, Katie Vo, Joanne Mun-Yee Lim, Lisa Laverne Barnes, Connie Kreuger, Jonathan Cordona, Zach Bonacorsi, Xiomara Mendoza Apuango, Stacey Winterton, Haley Hendrix, John Moore-Weiss, Renee LaBarge, Brandi Barger, REMA RAMAN, Serena Cepeda, VJ Lowe, Jacob Faircloth, Nathiel Pacini, Kimberly Jackson, Akiva Mintz, Socorro Trujillo, Tascha Helland, Kelly Larkin, Jasmine Stewart, R. L. Nosheny, Jessica Magnuson, Abbey Arnold, Nicole Gagliano, Leon J Thal, Zoe Potter, Raquel Prado, Susan Landau, Edward B Lee, Maria Chaudhary, Carole Ho Denali, Nick C Fox, Agathe Vrillon, Michael Rafii, Joel Felmlee Jr., Sylvia White (Salazar, Zoe Lin, James O'Connor, Naomi Saito, Rene Gutierrez, Mount Sinai, Jessica Nath, Kerry Crawley, Charna Bates, Stephanie Smith, Shaniya Parkins Mt Sinai, Paul Aisen, James Brewer, Joe Strong, Justin Robison, Alex Groves, Adam Diaz, Aidyn Rose, Maison Abu Raya, Kelley Faber, Yanan Shang, Taeho Jo, Chengshi Jin, Y. Stern, Li Shen, David Jones, Cynthia Munoz, Nichol Ferng, Paul Thompson, Sonia Pawluczyk, Charles Windon, Jennifer Frascino, Mary Sano, Hugo Botha, John Morris, Hector González, Clifford R Jack, Adeyinka Ajayi, Michael Holt, Emily Little, Kejal Kantarci, Leslie M Shaw, Sarah Munie, Olusegun Adegoke, Lisa Taylor-Reinwald, Ian Malone, Tera Clay, Dariya Malyarenko, Jennifer Ryan, Andrea Fidell, Diana Ferman, Fabiola Magana, Scott Neu, Nigel J Cairns, David Knopman, Julia Boudreau, J Q Trojanowki, Matthew Senjem, Hillel Grossman, Neda Jahanshad, Madiha Fairooz, Sonia Garrison, Gustavo Jimenez, Katherin Martin, Mauricio Becerra, Anna Middleton, Jeffrey Bishop, Barbara Hackenmiller, Daniel Schwartz, Loren Brown Ashley, Adrienne Kormos, Russell T Shinohara, Kaori Kubo Germano, Omobolanle Ayo, Donna Masterman, Winnie Kwang, Harold Matthew Cooper, Christopher Schwarz, David Geldmacher, Lisa Schoolcraft, Nicole Robert, Taylor Clanton, Kirsten Evans, Lawrence Honig, Madison DeCapo, Richard Frank, Jonathan Graff-Radford, Elizabeth Shaffer, Gil Rabinovici, Andrew J Saykin, Lon Schneider, Lisa Thomas, Kwangsik Nho, Laurie Ryan, Kristin Oleary, Stephanie Rossi Chen, Cat Conti, Shannon L Risacher, Lindsey Booth Robbins, Allison Ardolino, Jennifer Mahaffey, Marie Bernard, Robert Reid, Chi Kim, Fatima S Amtashar, Princess Carter, Eliezer Masliah, Yuliana Cabrera, Tanya Harte, Rima Kaddurah-Daouk, Jeffrey Kaye, William Potter, Magdalena Korecka, Gina Garcia-Camilo, John K Hsiao, Steven Smith, Robert C Green, Maria Carrillo, Judith Heidebrink, Prashanthi Vemuri, Emily Largent, Justin Long, Caitlin Webb, Gillian Devereux, Joseph Di Benedetto, Karen Crawford, Duygu Tosun, Renaud La Joie, Melanie J Miller, Laili Soleimani, Kedir Adem Hussen, Blake Vogler, Randall Barker, Monica Rivera-Mindt, Madeline Paczynski, Ronald Petersen, Sarah Walter, Kara Wever, Janet "Janae" Taylor, Aimee Pierce, Nasrin Ghesani, Vanessa Guzman, Danielle Harvey, Charles DeCarli, Maria Kataki, Talia M Nir, Tatiana M Foroud, Liana G Apostolova, Catherine (Cat) Conti, James Ruszkiewicz, Vijay Ramanan, Jennifer Salazar, Rachel Henson, Virginia Boatwright, Richard J Perrin, Genesis Burgos-Rivera, Sheila Ogwang, Gregory Ikner, Michael W Weiner, Heather Nicole Elmore, Tom Montine, Victor Villemagne, Trinity Weisensel, Jaimie Ziolkowski, Jonathan Jackson, Mike Nguyen, Matt Glittenberg, Hannatu Amaza, Peter J Snyder, Regan Brooks, Hannah Craft, Alzheimer’s Disease Neuroimaging Initiative, Denise Reyes, Paul A Yushkevich, Caileigh Zimmerman, Robert A Koeppe, Jeff Gunter, Colleen Albers, Michelle Hernandez, Zaven Kachaturian, Sandhitsu Das, Virginia Rogers, Antonio Gama, Ozioma Okonkwo, Evan Fletcher, Bradley Boeve, Yang Wan, Sophia I Thomopoulos
article en

Abstract

Modeling with multidimensional arrays, or tensors, often presents a problem due to high dimensionality. In addition, these structures typically exhibit inherent sparsity, requiring the use of regularization methods to properly characterize an association between a tensor covariate and a scalar response. We propose a Bayesian method to efficiently model a scalar response with a tensor covariate using the Tucker tensor decomposition in order to retain the spatial relationship within a tensor coefficient, while reducing the number of parameters varying within the model and applying regularization methods. Simulated data are analyzed to compare the model to recently proposed methods. A neuroimaging analysis using data from the Alzheimer's Data Neuroimaging Initiative shows improved inferential performance compared with other tensor regression methods. Bayesian analysis; tensor decomposition; image analysis; spatial statistics; statistical modeling.

BiostatisticsVol. 26(1)
The University of Texas at El Paso (US), University of California, Santa Cruz (US), Winchester Hospital (US), University of Pennsylvania (US), Texas A&M University (US)
Openalex Percentile: Top 61%
Tensor decomposition and applications
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.