A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort

Archana Venkataraman, Nicholas Wymbs, Mary Beth Nebel, Stewart H Mostofsky

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

Abstract

We present a generative Bayesian framework that automatically extracts the hubs of altered functional connectivity between a neurotypical and a patient group, while simultaneously incorporating an observed clinical severity measure for each patient. The key to our framework is the latent or hidden organization in the brain that we cannot directly access. Instead, we observe noisy measurements of the latent structure through functional connectivity data. We derive a variational EM algorithm to infer both the latent network topology and the unknown model parameters. We demonstrate the robustness and clinical relevance of our model on a population study of autism acquired at the Kennedy Krieger Institute in Baltimore, MD. Our model results implicate a more diverse pattern of functional differences than two baseline techniques, which do not incorporate patient heterogeneity.

Original languageEnglish (US)
Title of host publicationConnectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings
PublisherSpringer Verlag
Pages60-69
Number of pages10
Volume10511 LNCS
ISBN (Print)9783319671581
DOIs
StatePublished - 2017
Event1st International Workshop on Connectomics in NeuroImaging, CNI 2017 held in conjunction with the 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada
Duration: Sep 14 2017Sep 14 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10511 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other1st International Workshop on Connectomics in NeuroImaging, CNI 2017 held in conjunction with the 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
CountryCanada
CityQuebec City
Period9/14/179/14/17

Fingerprint

Bayesian Approach
Connectivity
EM Algorithm
Network Topology
Baseline
Brain
Topology
Model
Robustness
Unknown
Demonstrate
Framework

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Venkataraman, A., Wymbs, N., Nebel, M. B., & Mostofsky, S. H. (2017). A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort. In Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings (Vol. 10511 LNCS, pp. 60-69). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10511 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-67159-8_8

A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort. / Venkataraman, Archana; Wymbs, Nicholas; Nebel, Mary Beth; Mostofsky, Stewart H.

Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10511 LNCS Springer Verlag, 2017. p. 60-69 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10511 LNCS).

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

Venkataraman, A, Wymbs, N, Nebel, MB & Mostofsky, SH 2017, A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort. in Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings. vol. 10511 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10511 LNCS, Springer Verlag, pp. 60-69, 1st International Workshop on Connectomics in NeuroImaging, CNI 2017 held in conjunction with the 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, Quebec City, Canada, 9/14/17. https://doi.org/10.1007/978-3-319-67159-8_8
Venkataraman A, Wymbs N, Nebel MB, Mostofsky SH. A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort. In Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10511 LNCS. Springer Verlag. 2017. p. 60-69. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-67159-8_8
Venkataraman, Archana ; Wymbs, Nicholas ; Nebel, Mary Beth ; Mostofsky, Stewart H. / A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort. Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10511 LNCS Springer Verlag, 2017. pp. 60-69 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b068e3c15bc34d6d824c386b75b825ed,
title = "A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort",
abstract = "We present a generative Bayesian framework that automatically extracts the hubs of altered functional connectivity between a neurotypical and a patient group, while simultaneously incorporating an observed clinical severity measure for each patient. The key to our framework is the latent or hidden organization in the brain that we cannot directly access. Instead, we observe noisy measurements of the latent structure through functional connectivity data. We derive a variational EM algorithm to infer both the latent network topology and the unknown model parameters. We demonstrate the robustness and clinical relevance of our model on a population study of autism acquired at the Kennedy Krieger Institute in Baltimore, MD. Our model results implicate a more diverse pattern of functional differences than two baseline techniques, which do not incorporate patient heterogeneity.",
author = "Archana Venkataraman and Nicholas Wymbs and Nebel, {Mary Beth} and Mostofsky, {Stewart H}",
year = "2017",
doi = "10.1007/978-3-319-67159-8_8",
language = "English (US)",
isbn = "9783319671581",
volume = "10511 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "60--69",
booktitle = "Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings",

}

TY - GEN

T1 - A unified Bayesian approach to extract network-based functional differences from a heterogeneous patient cohort

AU - Venkataraman, Archana

AU - Wymbs, Nicholas

AU - Nebel, Mary Beth

AU - Mostofsky, Stewart H

PY - 2017

Y1 - 2017

N2 - We present a generative Bayesian framework that automatically extracts the hubs of altered functional connectivity between a neurotypical and a patient group, while simultaneously incorporating an observed clinical severity measure for each patient. The key to our framework is the latent or hidden organization in the brain that we cannot directly access. Instead, we observe noisy measurements of the latent structure through functional connectivity data. We derive a variational EM algorithm to infer both the latent network topology and the unknown model parameters. We demonstrate the robustness and clinical relevance of our model on a population study of autism acquired at the Kennedy Krieger Institute in Baltimore, MD. Our model results implicate a more diverse pattern of functional differences than two baseline techniques, which do not incorporate patient heterogeneity.

AB - We present a generative Bayesian framework that automatically extracts the hubs of altered functional connectivity between a neurotypical and a patient group, while simultaneously incorporating an observed clinical severity measure for each patient. The key to our framework is the latent or hidden organization in the brain that we cannot directly access. Instead, we observe noisy measurements of the latent structure through functional connectivity data. We derive a variational EM algorithm to infer both the latent network topology and the unknown model parameters. We demonstrate the robustness and clinical relevance of our model on a population study of autism acquired at the Kennedy Krieger Institute in Baltimore, MD. Our model results implicate a more diverse pattern of functional differences than two baseline techniques, which do not incorporate patient heterogeneity.

UR - http://www.scopus.com/inward/record.url?scp=85029457348&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029457348&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-67159-8_8

DO - 10.1007/978-3-319-67159-8_8

M3 - Conference contribution

SN - 9783319671581

VL - 10511 LNCS

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

SP - 60

EP - 69

BT - Connectomics in NeuroImaging - 1st International Workshop, CNI 2017 Held in Conjunction with MICCAI 2017, Proceedings

PB - Springer Verlag

ER -