Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis

Biao Jie, Mingxia Liu, Chunfeng Lian, Feng Shi, Dinggang Shen

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

Abstract

Functional magnetic resonance imaging (fMRI) has been widely applied to analysis and diagnosis of brain diseases, including Alzheimer’s disease (AD) and its prodrome, i.e., mild cognitive impairment (MCI). Traditional methods usually construct connectivity networks (CNs) by simply calculating Pearson correlation coefficients (PCCs) between time series of brain regions, and then extract low-level network measures as features to train the learning model. However, the valuable observation information in network construction (e.g., specific contributions of different time points) and high-level (i.e., high-order) network properties are neglected in these methods. In this paper, we first define a novel weighted correlation kernel (called wc-kernel) to measure the correlation of brain regions, by which weighting factors are determined in a data-driven manner to characterize the contribution of each time point, thus conveying the richer interaction information of brain regions compared with the PCC method. Furthermore, we propose a wc-kernel based convolutional neural network (CNN) (called wck-CNN) framework for extracting the hierarchical (i.e., from low-order to high-order) functional connectivities for disease diagnosis, by using fMRI data. Specifically, we first define a layer to build dynamic CNs (DCNs) using the defined wc-kernels. Then, we define three layers to extract local (region specific), global (network specific) and temporal high-order properties from the constructed low-order functional connectivities as features for classification. Results on 174 subjects (a total of 563 scans) with rs-fMRI data from ADNI suggest that the our method can not only improve the performance compared with state-of-the-art methods, but also provide novel insights into the interaction patterns of brain activities and their changes in diseases.

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
EditorsMingxia Liu, Heung-Il Suk, Yinghuan Shi
PublisherSpringer Verlag
Pages1-9
Number of pages9
ISBN (Print)9783030009182
DOIs
Publication statusPublished - 2018 Jan 1
Externally publishedYes
Event9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 2018 Sep 162018 Sep 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11046 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period18/9/1618/9/16

Fingerprint

Functional Magnetic Resonance Imaging
Brain
Connectivity
Neural Networks
kernel
Neural networks
Pearson Correlation
Network Connectivity
Higher Order
Correlation coefficient
Alzheimer's Disease
Conveying
Interaction
Data-driven
Weighting
Time series
Magnetic Resonance Imaging

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Jie, B., Liu, M., Lian, C., Shi, F., & Shen, D. (2018). Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis. In M. Liu, H-I. Suk, & Y. Shi (Eds.), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings (pp. 1-9). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11046 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-00919-9_1

Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis. / Jie, Biao; Liu, Mingxia; Lian, Chunfeng; Shi, Feng; Shen, Dinggang.

Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. ed. / Mingxia Liu; Heung-Il Suk; Yinghuan Shi. Springer Verlag, 2018. p. 1-9 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11046 LNCS).

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

Jie, B, Liu, M, Lian, C, Shi, F & Shen, D 2018, Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis. in M Liu, H-I Suk & Y Shi (eds), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11046 LNCS, Springer Verlag, pp. 1-9, 9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018, Granada, Spain, 18/9/16. https://doi.org/10.1007/978-3-030-00919-9_1
Jie B, Liu M, Lian C, Shi F, Shen D. Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis. In Liu M, Suk H-I, Shi Y, editors, Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Springer Verlag. 2018. p. 1-9. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-00919-9_1
Jie, Biao ; Liu, Mingxia ; Lian, Chunfeng ; Shi, Feng ; Shen, Dinggang. / Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis. Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. editor / Mingxia Liu ; Heung-Il Suk ; Yinghuan Shi. Springer Verlag, 2018. pp. 1-9 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b03451605bfc4c45baa04184feff2783,
title = "Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis",
abstract = "Functional magnetic resonance imaging (fMRI) has been widely applied to analysis and diagnosis of brain diseases, including Alzheimer’s disease (AD) and its prodrome, i.e., mild cognitive impairment (MCI). Traditional methods usually construct connectivity networks (CNs) by simply calculating Pearson correlation coefficients (PCCs) between time series of brain regions, and then extract low-level network measures as features to train the learning model. However, the valuable observation information in network construction (e.g., specific contributions of different time points) and high-level (i.e., high-order) network properties are neglected in these methods. In this paper, we first define a novel weighted correlation kernel (called wc-kernel) to measure the correlation of brain regions, by which weighting factors are determined in a data-driven manner to characterize the contribution of each time point, thus conveying the richer interaction information of brain regions compared with the PCC method. Furthermore, we propose a wc-kernel based convolutional neural network (CNN) (called wck-CNN) framework for extracting the hierarchical (i.e., from low-order to high-order) functional connectivities for disease diagnosis, by using fMRI data. Specifically, we first define a layer to build dynamic CNs (DCNs) using the defined wc-kernels. Then, we define three layers to extract local (region specific), global (network specific) and temporal high-order properties from the constructed low-order functional connectivities as features for classification. Results on 174 subjects (a total of 563 scans) with rs-fMRI data from ADNI suggest that the our method can not only improve the performance compared with state-of-the-art methods, but also provide novel insights into the interaction patterns of brain activities and their changes in diseases.",
author = "Biao Jie and Mingxia Liu and Chunfeng Lian and Feng Shi and Dinggang Shen",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-00919-9_1",
language = "English",
isbn = "9783030009182",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "1--9",
editor = "Mingxia Liu and Heung-Il Suk and Yinghuan Shi",
booktitle = "Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings",

}

TY - GEN

T1 - Developing novel weighted correlation kernels for convolutional neural networks to extract hierarchical functional connectivities from fMRI for disease diagnosis

AU - Jie, Biao

AU - Liu, Mingxia

AU - Lian, Chunfeng

AU - Shi, Feng

AU - Shen, Dinggang

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Functional magnetic resonance imaging (fMRI) has been widely applied to analysis and diagnosis of brain diseases, including Alzheimer’s disease (AD) and its prodrome, i.e., mild cognitive impairment (MCI). Traditional methods usually construct connectivity networks (CNs) by simply calculating Pearson correlation coefficients (PCCs) between time series of brain regions, and then extract low-level network measures as features to train the learning model. However, the valuable observation information in network construction (e.g., specific contributions of different time points) and high-level (i.e., high-order) network properties are neglected in these methods. In this paper, we first define a novel weighted correlation kernel (called wc-kernel) to measure the correlation of brain regions, by which weighting factors are determined in a data-driven manner to characterize the contribution of each time point, thus conveying the richer interaction information of brain regions compared with the PCC method. Furthermore, we propose a wc-kernel based convolutional neural network (CNN) (called wck-CNN) framework for extracting the hierarchical (i.e., from low-order to high-order) functional connectivities for disease diagnosis, by using fMRI data. Specifically, we first define a layer to build dynamic CNs (DCNs) using the defined wc-kernels. Then, we define three layers to extract local (region specific), global (network specific) and temporal high-order properties from the constructed low-order functional connectivities as features for classification. Results on 174 subjects (a total of 563 scans) with rs-fMRI data from ADNI suggest that the our method can not only improve the performance compared with state-of-the-art methods, but also provide novel insights into the interaction patterns of brain activities and their changes in diseases.

AB - Functional magnetic resonance imaging (fMRI) has been widely applied to analysis and diagnosis of brain diseases, including Alzheimer’s disease (AD) and its prodrome, i.e., mild cognitive impairment (MCI). Traditional methods usually construct connectivity networks (CNs) by simply calculating Pearson correlation coefficients (PCCs) between time series of brain regions, and then extract low-level network measures as features to train the learning model. However, the valuable observation information in network construction (e.g., specific contributions of different time points) and high-level (i.e., high-order) network properties are neglected in these methods. In this paper, we first define a novel weighted correlation kernel (called wc-kernel) to measure the correlation of brain regions, by which weighting factors are determined in a data-driven manner to characterize the contribution of each time point, thus conveying the richer interaction information of brain regions compared with the PCC method. Furthermore, we propose a wc-kernel based convolutional neural network (CNN) (called wck-CNN) framework for extracting the hierarchical (i.e., from low-order to high-order) functional connectivities for disease diagnosis, by using fMRI data. Specifically, we first define a layer to build dynamic CNs (DCNs) using the defined wc-kernels. Then, we define three layers to extract local (region specific), global (network specific) and temporal high-order properties from the constructed low-order functional connectivities as features for classification. Results on 174 subjects (a total of 563 scans) with rs-fMRI data from ADNI suggest that the our method can not only improve the performance compared with state-of-the-art methods, but also provide novel insights into the interaction patterns of brain activities and their changes in diseases.

UR - http://www.scopus.com/inward/record.url?scp=85054552914&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054552914&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00919-9_1

DO - 10.1007/978-3-030-00919-9_1

M3 - Conference contribution

AN - SCOPUS:85054552914

SN - 9783030009182

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

SP - 1

EP - 9

BT - Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings

A2 - Liu, Mingxia

A2 - Suk, Heung-Il

A2 - Shi, Yinghuan

PB - Springer Verlag

ER -