Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states

Nicolas Georges, Islem Rekik (Lead / Corresponding author),

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

21 Downloads (Pure)

Abstract

Machine learning methods present unprecedented opportunities to advance our understanding of the connectomics of brain disorders. With the proliferation of extremely high-dimensional connectomic data drawn from multiple neuroimaging sources (e.g., functional and structural MRIs), effective feature selection (FS) methods have become indispensable components for (i) disentangling brain states (e.g., early vs late mild cognitive impairment) and (ii) identifying connectional features that might serve as biomarkers for treatment. Strangely, despite the extensive work on identifying stable discriminative features using a particular FS method, the challenge of choosing the best one from a large pool of existing FS techniques for optimally achieving (i) and (ii) using a dataset of interest remains unexplored. In essence, the question that we aim to address in this work is: “Given a set of feature selection methods {FS1,&FSk}, and a dataset of interest, which FS method might produce the most reproducible and ‘trustworthy’ connectomic features that accurately differentiate between two brain states?” This paper is an attempt to address this question by evaluating the performance of a particular feature selection for a specific data type in fulfilling criteria (i) and (ii). To this aim, we propose to model the relationships between a set of FS methods using a multi-graph architecture, where each graph quantifies the feature reproducibility power between graph nodes at a fixed number of top ranked features. Next, we integrate the reproducibility graphs with a discrepancy graph which captures the difference in classification performance between FS methods. This allows to identify, for a dataset of interest, the ‘central’ node with the highest degree, which reveals the most reliable and reproducible FS method for the target brain state classification task along with the most discriminative features fingerprinting these brain states. We evaluated our method on multi-view brain connectomic data for late mild cognitive impairment vs Alzheimer’s disease classification. Our experiments give insights into reproducible connectional features fingerprinting late dementia brain states.

Original languageEnglish
Title of host publicationConnectomics in NeuroImaging
Subtitle of host publicationSecond International Workshop, CNI 2018, Held in Conjunction with MICCAI 2018, Proceedings
EditorsGuorong Wu, Markus D. Schirmer, Ai Wern Chung, Islem Rekik, Brent Munsell
Place of PublicationSwitzerland
PublisherSpringer Verlag
Pages99-106
Number of pages8
Volume11083
ISBN (Electronic)9783030007553
ISBN (Print)9783030007546
DOIs
Publication statusPublished - 2018
Event2nd International Workshop on Connectomics in NeuroImaging, CNI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 20 Sep 201820 Sep 2018

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume11083
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd International Workshop on Connectomics in NeuroImaging, CNI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period20/09/1820/09/18

Fingerprint

Fingerprinting
Feature Selection
Feature extraction
Brain
Reproducibility
Graph in graph theory
Neuroimaging
Dementia
Alzheimer's Disease
Multigraph
Biomarkers
High-dimensional Data
Vertex of a graph
Proliferation
Differentiate
Magnetic resonance imaging
Discrepancy
Learning systems
Disorder
Machine Learning

Cite this

Georges, N., Rekik, I. (2018). Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states. In G. Wu, M. D. Schirmer, A. W. Chung, I. Rekik, & B. Munsell (Eds.), Connectomics in NeuroImaging: Second International Workshop, CNI 2018, Held in Conjunction with MICCAI 2018, Proceedings (Vol. 11083, pp. 99-106). (Lecture Notes in Computer Science; Vol. 11083). Switzerland: Springer Verlag. https://doi.org/10.1007/978-3-030-00755-3_11
Georges, Nicolas ; Rekik, Islem. / Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states. Connectomics in NeuroImaging: Second International Workshop, CNI 2018, Held in Conjunction with MICCAI 2018, Proceedings. editor / Guorong Wu ; Markus D. Schirmer ; Ai Wern Chung ; Islem Rekik ; Brent Munsell. Vol. 11083 Switzerland : Springer Verlag, 2018. pp. 99-106 (Lecture Notes in Computer Science).
@inproceedings{5686d9c58a474b2a917e7d1fb96fc208,
title = "Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states",
abstract = "Machine learning methods present unprecedented opportunities to advance our understanding of the connectomics of brain disorders. With the proliferation of extremely high-dimensional connectomic data drawn from multiple neuroimaging sources (e.g., functional and structural MRIs), effective feature selection (FS) methods have become indispensable components for (i) disentangling brain states (e.g., early vs late mild cognitive impairment) and (ii) identifying connectional features that might serve as biomarkers for treatment. Strangely, despite the extensive work on identifying stable discriminative features using a particular FS method, the challenge of choosing the best one from a large pool of existing FS techniques for optimally achieving (i) and (ii) using a dataset of interest remains unexplored. In essence, the question that we aim to address in this work is: “Given a set of feature selection methods {FS1,&FSk}, and a dataset of interest, which FS method might produce the most reproducible and ‘trustworthy’ connectomic features that accurately differentiate between two brain states?” This paper is an attempt to address this question by evaluating the performance of a particular feature selection for a specific data type in fulfilling criteria (i) and (ii). To this aim, we propose to model the relationships between a set of FS methods using a multi-graph architecture, where each graph quantifies the feature reproducibility power between graph nodes at a fixed number of top ranked features. Next, we integrate the reproducibility graphs with a discrepancy graph which captures the difference in classification performance between FS methods. This allows to identify, for a dataset of interest, the ‘central’ node with the highest degree, which reveals the most reliable and reproducible FS method for the target brain state classification task along with the most discriminative features fingerprinting these brain states. We evaluated our method on multi-view brain connectomic data for late mild cognitive impairment vs Alzheimer’s disease classification. Our experiments give insights into reproducible connectional features fingerprinting late dementia brain states.",
author = "Nicolas Georges and Islem Rekik",
year = "2018",
doi = "10.1007/978-3-030-00755-3_11",
language = "English",
isbn = "9783030007546",
volume = "11083",
series = "Lecture Notes in Computer Science",
publisher = "Springer Verlag",
pages = "99--106",
editor = "Guorong Wu and Schirmer, {Markus D.} and Chung, {Ai Wern} and Islem Rekik and Brent Munsell",
booktitle = "Connectomics in NeuroImaging",
address = "Germany",

}

Georges, N, Rekik, I 2018, Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states. in G Wu, MD Schirmer, AW Chung, I Rekik & B Munsell (eds), Connectomics in NeuroImaging: Second International Workshop, CNI 2018, Held in Conjunction with MICCAI 2018, Proceedings. vol. 11083, Lecture Notes in Computer Science, vol. 11083, Springer Verlag, Switzerland, pp. 99-106, 2nd International Workshop on Connectomics in NeuroImaging, CNI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018, Granada, Spain, 20/09/18. https://doi.org/10.1007/978-3-030-00755-3_11

Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states. / Georges, Nicolas; Rekik, Islem (Lead / Corresponding author).

Connectomics in NeuroImaging: Second International Workshop, CNI 2018, Held in Conjunction with MICCAI 2018, Proceedings. ed. / Guorong Wu; Markus D. Schirmer; Ai Wern Chung; Islem Rekik; Brent Munsell. Vol. 11083 Switzerland : Springer Verlag, 2018. p. 99-106 (Lecture Notes in Computer Science; Vol. 11083).

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

TY - GEN

T1 - Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states

AU - Georges, Nicolas

AU - Rekik, Islem

PY - 2018

Y1 - 2018

N2 - Machine learning methods present unprecedented opportunities to advance our understanding of the connectomics of brain disorders. With the proliferation of extremely high-dimensional connectomic data drawn from multiple neuroimaging sources (e.g., functional and structural MRIs), effective feature selection (FS) methods have become indispensable components for (i) disentangling brain states (e.g., early vs late mild cognitive impairment) and (ii) identifying connectional features that might serve as biomarkers for treatment. Strangely, despite the extensive work on identifying stable discriminative features using a particular FS method, the challenge of choosing the best one from a large pool of existing FS techniques for optimally achieving (i) and (ii) using a dataset of interest remains unexplored. In essence, the question that we aim to address in this work is: “Given a set of feature selection methods {FS1,&FSk}, and a dataset of interest, which FS method might produce the most reproducible and ‘trustworthy’ connectomic features that accurately differentiate between two brain states?” This paper is an attempt to address this question by evaluating the performance of a particular feature selection for a specific data type in fulfilling criteria (i) and (ii). To this aim, we propose to model the relationships between a set of FS methods using a multi-graph architecture, where each graph quantifies the feature reproducibility power between graph nodes at a fixed number of top ranked features. Next, we integrate the reproducibility graphs with a discrepancy graph which captures the difference in classification performance between FS methods. This allows to identify, for a dataset of interest, the ‘central’ node with the highest degree, which reveals the most reliable and reproducible FS method for the target brain state classification task along with the most discriminative features fingerprinting these brain states. We evaluated our method on multi-view brain connectomic data for late mild cognitive impairment vs Alzheimer’s disease classification. Our experiments give insights into reproducible connectional features fingerprinting late dementia brain states.

AB - Machine learning methods present unprecedented opportunities to advance our understanding of the connectomics of brain disorders. With the proliferation of extremely high-dimensional connectomic data drawn from multiple neuroimaging sources (e.g., functional and structural MRIs), effective feature selection (FS) methods have become indispensable components for (i) disentangling brain states (e.g., early vs late mild cognitive impairment) and (ii) identifying connectional features that might serve as biomarkers for treatment. Strangely, despite the extensive work on identifying stable discriminative features using a particular FS method, the challenge of choosing the best one from a large pool of existing FS techniques for optimally achieving (i) and (ii) using a dataset of interest remains unexplored. In essence, the question that we aim to address in this work is: “Given a set of feature selection methods {FS1,&FSk}, and a dataset of interest, which FS method might produce the most reproducible and ‘trustworthy’ connectomic features that accurately differentiate between two brain states?” This paper is an attempt to address this question by evaluating the performance of a particular feature selection for a specific data type in fulfilling criteria (i) and (ii). To this aim, we propose to model the relationships between a set of FS methods using a multi-graph architecture, where each graph quantifies the feature reproducibility power between graph nodes at a fixed number of top ranked features. Next, we integrate the reproducibility graphs with a discrepancy graph which captures the difference in classification performance between FS methods. This allows to identify, for a dataset of interest, the ‘central’ node with the highest degree, which reveals the most reliable and reproducible FS method for the target brain state classification task along with the most discriminative features fingerprinting these brain states. We evaluated our method on multi-view brain connectomic data for late mild cognitive impairment vs Alzheimer’s disease classification. Our experiments give insights into reproducible connectional features fingerprinting late dementia brain states.

UR - http://www.scopus.com/inward/record.url?scp=85054393367&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00755-3_11

DO - 10.1007/978-3-030-00755-3_11

M3 - Conference contribution

SN - 9783030007546

VL - 11083

T3 - Lecture Notes in Computer Science

SP - 99

EP - 106

BT - Connectomics in NeuroImaging

A2 - Wu, Guorong

A2 - Schirmer, Markus D.

A2 - Chung, Ai Wern

A2 - Rekik, Islem

A2 - Munsell, Brent

PB - Springer Verlag

CY - Switzerland

ER -

Georges N, Rekik I. Data-specific feature selection method identification for most reproducible connectomic feature discovery fingerprinting brain states. In Wu G, Schirmer MD, Chung AW, Rekik I, Munsell B, editors, Connectomics in NeuroImaging: Second International Workshop, CNI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Vol. 11083. Switzerland: Springer Verlag. 2018. p. 99-106. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-030-00755-3_11