Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models

Ayagoz Mussabayeva, Maxim Pisov, Anvar Kurmukov, Alexey Kroshnin, Yulia Denisova, Li Shen, Shan Cong, Lei Wang, Boris Gutman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a method for metric optimization and template construction in the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. The construction treats the Riemannian metric on the space of diffeomorphisms as a data-embedding kernel in the context of predictive modeling, here Kernel Logistic Regression (KLR). The task is then to optimize kernel parameters, including the LDDMM metric parameters as well as the registration template, resulting in a parameterized argminimum optimization. In practice, this leads to a group-wise registration problem with the goal of improving predictive performance, for example by focusing the metric and template on discriminating patient and control populations. We validate our algorithm using two discriminative problems on a synthetic data set as well as 3D subcortical shapes from the SchizConnect cohort. Though secondary to the template and kernel optimization, accuracy of schizophrenia classification is improved by LDDMM-KLR compared to linear and RBF-KLR.

Original languageEnglish (US)
Title of host publicationMultimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings
EditorsDajiang Zhu, Jingwen Yan, Heng Huang, Li Shen, Paul M. Thompson, Carl-Fredrik Westin, Xavier Pennec, Sarang Joshi, Mads Nielsen, Stefan Sommer, Tom Fletcher, Stanley Durrleman
PublisherSpringer
Pages151-161
Number of pages11
ISBN (Print)9783030332259
DOIs
StatePublished - Jan 1 2019
Externally publishedYes
Event4th International Workshop on Multimodal Brain Image Analysis, MBAI 2019, and the 7th International Workshop on Mathematical Foundations of Computational Anatomy, MFCA 2019, held in conjunction with the 22nd International Conference on Medical Imaging and Computer Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: Oct 17 2019Oct 17 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11846 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference4th International Workshop on Multimodal Brain Image Analysis, MBAI 2019, and the 7th International Workshop on Mathematical Foundations of Computational Anatomy, MFCA 2019, held in conjunction with the 22nd International Conference on Medical Imaging and Computer Assisted Intervention, MICCAI 2019
CountryChina
CityShenzhen
Period10/17/1910/17/19

Fingerprint

Predictive Model
Registration
Logistics
Template
Metric
Kernel Regression
Optimization
Large Deformation
Logistic Regression
kernel
Predictive Modeling
3D shape
Riemannian Metric
Synthetic Data
Diffeomorphisms
Learning
Optimise

Keywords

  • Image registration
  • LDDMM
  • Machine learning
  • Metric learning
  • Subcortical shape

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Mussabayeva, A., Pisov, M., Kurmukov, A., Kroshnin, A., Denisova, Y., Shen, L., ... Gutman, B. (2019). Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models. In D. Zhu, J. Yan, H. Huang, L. Shen, P. M. Thompson, C-F. Westin, X. Pennec, S. Joshi, M. Nielsen, S. Sommer, T. Fletcher, ... S. Durrleman (Eds.), Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings (pp. 151-161). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11846 LNCS). Springer. https://doi.org/10.1007/978-3-030-33226-6_17

Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models. / Mussabayeva, Ayagoz; Pisov, Maxim; Kurmukov, Anvar; Kroshnin, Alexey; Denisova, Yulia; Shen, Li; Cong, Shan; Wang, Lei; Gutman, Boris.

Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings. ed. / Dajiang Zhu; Jingwen Yan; Heng Huang; Li Shen; Paul M. Thompson; Carl-Fredrik Westin; Xavier Pennec; Sarang Joshi; Mads Nielsen; Stefan Sommer; Tom Fletcher; Stanley Durrleman. Springer, 2019. p. 151-161 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11846 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mussabayeva, A, Pisov, M, Kurmukov, A, Kroshnin, A, Denisova, Y, Shen, L, Cong, S, Wang, L & Gutman, B 2019, Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models. in D Zhu, J Yan, H Huang, L Shen, PM Thompson, C-F Westin, X Pennec, S Joshi, M Nielsen, S Sommer, T Fletcher & S Durrleman (eds), Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11846 LNCS, Springer, pp. 151-161, 4th International Workshop on Multimodal Brain Image Analysis, MBAI 2019, and the 7th International Workshop on Mathematical Foundations of Computational Anatomy, MFCA 2019, held in conjunction with the 22nd International Conference on Medical Imaging and Computer Assisted Intervention, MICCAI 2019, Shenzhen, China, 10/17/19. https://doi.org/10.1007/978-3-030-33226-6_17
Mussabayeva A, Pisov M, Kurmukov A, Kroshnin A, Denisova Y, Shen L et al. Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models. In Zhu D, Yan J, Huang H, Shen L, Thompson PM, Westin C-F, Pennec X, Joshi S, Nielsen M, Sommer S, Fletcher T, Durrleman S, editors, Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings. Springer. 2019. p. 151-161. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-33226-6_17
Mussabayeva, Ayagoz ; Pisov, Maxim ; Kurmukov, Anvar ; Kroshnin, Alexey ; Denisova, Yulia ; Shen, Li ; Cong, Shan ; Wang, Lei ; Gutman, Boris. / Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models. Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings. editor / Dajiang Zhu ; Jingwen Yan ; Heng Huang ; Li Shen ; Paul M. Thompson ; Carl-Fredrik Westin ; Xavier Pennec ; Sarang Joshi ; Mads Nielsen ; Stefan Sommer ; Tom Fletcher ; Stanley Durrleman. Springer, 2019. pp. 151-161 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{042326b3ea954f29bfd13dcc5fccd184,
title = "Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models",
abstract = "We present a method for metric optimization and template construction in the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. The construction treats the Riemannian metric on the space of diffeomorphisms as a data-embedding kernel in the context of predictive modeling, here Kernel Logistic Regression (KLR). The task is then to optimize kernel parameters, including the LDDMM metric parameters as well as the registration template, resulting in a parameterized argminimum optimization. In practice, this leads to a group-wise registration problem with the goal of improving predictive performance, for example by focusing the metric and template on discriminating patient and control populations. We validate our algorithm using two discriminative problems on a synthetic data set as well as 3D subcortical shapes from the SchizConnect cohort. Though secondary to the template and kernel optimization, accuracy of schizophrenia classification is improved by LDDMM-KLR compared to linear and RBF-KLR.",
keywords = "Image registration, LDDMM, Machine learning, Metric learning, Subcortical shape",
author = "Ayagoz Mussabayeva and Maxim Pisov and Anvar Kurmukov and Alexey Kroshnin and Yulia Denisova and Li Shen and Shan Cong and Lei Wang and Boris Gutman",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-33226-6_17",
language = "English (US)",
isbn = "9783030332259",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "151--161",
editor = "Dajiang Zhu and Jingwen Yan and Heng Huang and Li Shen and Thompson, {Paul M.} and Carl-Fredrik Westin and Xavier Pennec and Sarang Joshi and Mads Nielsen and Stefan Sommer and Tom Fletcher and Stanley Durrleman",
booktitle = "Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings",

}

TY - GEN

T1 - Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models

AU - Mussabayeva, Ayagoz

AU - Pisov, Maxim

AU - Kurmukov, Anvar

AU - Kroshnin, Alexey

AU - Denisova, Yulia

AU - Shen, Li

AU - Cong, Shan

AU - Wang, Lei

AU - Gutman, Boris

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We present a method for metric optimization and template construction in the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. The construction treats the Riemannian metric on the space of diffeomorphisms as a data-embedding kernel in the context of predictive modeling, here Kernel Logistic Regression (KLR). The task is then to optimize kernel parameters, including the LDDMM metric parameters as well as the registration template, resulting in a parameterized argminimum optimization. In practice, this leads to a group-wise registration problem with the goal of improving predictive performance, for example by focusing the metric and template on discriminating patient and control populations. We validate our algorithm using two discriminative problems on a synthetic data set as well as 3D subcortical shapes from the SchizConnect cohort. Though secondary to the template and kernel optimization, accuracy of schizophrenia classification is improved by LDDMM-KLR compared to linear and RBF-KLR.

AB - We present a method for metric optimization and template construction in the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. The construction treats the Riemannian metric on the space of diffeomorphisms as a data-embedding kernel in the context of predictive modeling, here Kernel Logistic Regression (KLR). The task is then to optimize kernel parameters, including the LDDMM metric parameters as well as the registration template, resulting in a parameterized argminimum optimization. In practice, this leads to a group-wise registration problem with the goal of improving predictive performance, for example by focusing the metric and template on discriminating patient and control populations. We validate our algorithm using two discriminative problems on a synthetic data set as well as 3D subcortical shapes from the SchizConnect cohort. Though secondary to the template and kernel optimization, accuracy of schizophrenia classification is improved by LDDMM-KLR compared to linear and RBF-KLR.

KW - Image registration

KW - LDDMM

KW - Machine learning

KW - Metric learning

KW - Subcortical shape

UR - http://www.scopus.com/inward/record.url?scp=85075568847&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075568847&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-33226-6_17

DO - 10.1007/978-3-030-33226-6_17

M3 - Conference contribution

AN - SCOPUS:85075568847

SN - 9783030332259

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

SP - 151

EP - 161

BT - Multimodal Brain Image Analysis and Mathematical Foundations of Computational Anatomy - 4th International Workshop, MBIA 2019, and 7th International Workshop, MFCA 2019, Held in Conjunction with MICCAI 2019, Proceedings

A2 - Zhu, Dajiang

A2 - Yan, Jingwen

A2 - Huang, Heng

A2 - Shen, Li

A2 - Thompson, Paul M.

A2 - Westin, Carl-Fredrik

A2 - Pennec, Xavier

A2 - Joshi, Sarang

A2 - Nielsen, Mads

A2 - Sommer, Stefan

A2 - Fletcher, Tom

A2 - Durrleman, Stanley

PB - Springer

ER -