Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis

Monami Banerjee, Rudrasis Chakraborty, Edward Ofori, David Vaillancourt, Baba C. Vemuri

Research output: Chapter in Book/Report/Conference proceedingChapter

13 Citations (Scopus)

Abstract

Regression in its most common form where independent and dependent variables are in Rn is a ubiquitous tool in Sciences and Engineering. Recent advances in Medical Imaging has lead to a wide spread availability of manifoldvalued data leading to problems where the independent variables are manifoldvalued and dependent are real-valued or vice-versa. The most common method of regression on a manifold is the geodesic regression, which is the counterpart of linear regression in Euclidean space. Often, the relation between the variables is highly complex, and existing most commonly used geodesic regression can prove to be inaccurate. Thus, it is necessary to resort to a non-linear model for regression. In this work we present a novel Kernel based non-linear regression method when the mapping to be estimated is either from M → ℝn or ℝn → M, where M is a Riemannian manifold. A key advantage of this approach is that there is no requirement for the manifold-valued data to necessarily inherit an ordering from the data in ℝn.We present several synthetic and real data experiments along with comparisons to the state-of-the-art geodesic regression method in literature and thus validating the effectiveness of the proposed algorithm.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages719-727
Number of pages9
DOIs
StatePublished - Oct 1 2015
Externally publishedYes

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9349
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Nonlinear Regression
Medical imaging
Linear regression
Image Analysis
Image analysis
Riemannian Manifold
Regression
Availability
Geodesic
Experiments
Dependent
Medical Imaging
Inaccurate
Nonlinear Model
Euclidean space
kernel
Engineering
Necessary
Requirements
Experiment

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Banerjee, M., Chakraborty, R., Ofori, E., Vaillancourt, D., & Vemuri, B. C. (2015). Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 719-727). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9349). Springer Verlag. https://doi.org/10.1007/978-3-319-24553-9_88

Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis. / Banerjee, Monami; Chakraborty, Rudrasis; Ofori, Edward; Vaillancourt, David; Vemuri, Baba C.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Verlag, 2015. p. 719-727 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9349).

Research output: Chapter in Book/Report/Conference proceedingChapter

Banerjee, M, Chakraborty, R, Ofori, E, Vaillancourt, D & Vemuri, BC 2015, Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9349, Springer Verlag, pp. 719-727. https://doi.org/10.1007/978-3-319-24553-9_88
Banerjee M, Chakraborty R, Ofori E, Vaillancourt D, Vemuri BC. Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Verlag. 2015. p. 719-727. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-24553-9_88
Banerjee, Monami ; Chakraborty, Rudrasis ; Ofori, Edward ; Vaillancourt, David ; Vemuri, Baba C. / Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Verlag, 2015. pp. 719-727 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inbook{2f0b975e5ed841ca9fa3678c67f456ca,
title = "Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis",
abstract = "Regression in its most common form where independent and dependent variables are in Rn is a ubiquitous tool in Sciences and Engineering. Recent advances in Medical Imaging has lead to a wide spread availability of manifoldvalued data leading to problems where the independent variables are manifoldvalued and dependent are real-valued or vice-versa. The most common method of regression on a manifold is the geodesic regression, which is the counterpart of linear regression in Euclidean space. Often, the relation between the variables is highly complex, and existing most commonly used geodesic regression can prove to be inaccurate. Thus, it is necessary to resort to a non-linear model for regression. In this work we present a novel Kernel based non-linear regression method when the mapping to be estimated is either from M → ℝn or ℝn → M, where M is a Riemannian manifold. A key advantage of this approach is that there is no requirement for the manifold-valued data to necessarily inherit an ordering from the data in ℝn.We present several synthetic and real data experiments along with comparisons to the state-of-the-art geodesic regression method in literature and thus validating the effectiveness of the proposed algorithm.",
author = "Monami Banerjee and Rudrasis Chakraborty and Edward Ofori and David Vaillancourt and Vemuri, {Baba C.}",
year = "2015",
month = "10",
day = "1",
doi = "10.1007/978-3-319-24553-9_88",
language = "English (US)",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "719--727",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - CHAP

T1 - Nonlinear regression on riemannian manifolds and its applications to neuro-image analysis

AU - Banerjee, Monami

AU - Chakraborty, Rudrasis

AU - Ofori, Edward

AU - Vaillancourt, David

AU - Vemuri, Baba C.

PY - 2015/10/1

Y1 - 2015/10/1

N2 - Regression in its most common form where independent and dependent variables are in Rn is a ubiquitous tool in Sciences and Engineering. Recent advances in Medical Imaging has lead to a wide spread availability of manifoldvalued data leading to problems where the independent variables are manifoldvalued and dependent are real-valued or vice-versa. The most common method of regression on a manifold is the geodesic regression, which is the counterpart of linear regression in Euclidean space. Often, the relation between the variables is highly complex, and existing most commonly used geodesic regression can prove to be inaccurate. Thus, it is necessary to resort to a non-linear model for regression. In this work we present a novel Kernel based non-linear regression method when the mapping to be estimated is either from M → ℝn or ℝn → M, where M is a Riemannian manifold. A key advantage of this approach is that there is no requirement for the manifold-valued data to necessarily inherit an ordering from the data in ℝn.We present several synthetic and real data experiments along with comparisons to the state-of-the-art geodesic regression method in literature and thus validating the effectiveness of the proposed algorithm.

AB - Regression in its most common form where independent and dependent variables are in Rn is a ubiquitous tool in Sciences and Engineering. Recent advances in Medical Imaging has lead to a wide spread availability of manifoldvalued data leading to problems where the independent variables are manifoldvalued and dependent are real-valued or vice-versa. The most common method of regression on a manifold is the geodesic regression, which is the counterpart of linear regression in Euclidean space. Often, the relation between the variables is highly complex, and existing most commonly used geodesic regression can prove to be inaccurate. Thus, it is necessary to resort to a non-linear model for regression. In this work we present a novel Kernel based non-linear regression method when the mapping to be estimated is either from M → ℝn or ℝn → M, where M is a Riemannian manifold. A key advantage of this approach is that there is no requirement for the manifold-valued data to necessarily inherit an ordering from the data in ℝn.We present several synthetic and real data experiments along with comparisons to the state-of-the-art geodesic regression method in literature and thus validating the effectiveness of the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=84947556741&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947556741&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-24553-9_88

DO - 10.1007/978-3-319-24553-9_88

M3 - Chapter

AN - SCOPUS:84947556741

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 719

EP - 727

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -