Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis

Weizheng Yan, Han Zhang, Jing Sui, Dinggang Shen

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

2 Citations (Scopus)

Abstract

Brain functional connectivity (FC) extracted from resting-state fMRI (RS-fMRI) has become a popular approach for disease diagnosis, where discriminating subjects with mild cognitive impairment (MCI) from normal controls (NC) is still one of the most challenging problems. Dynamic functional connectivity (dFC), consisting of time-varying spatiotemporal dynamics, may characterize “chronnectome” diagnostic information for improving MCI classification. However, most of the current dFC studies are based on detecting discrete major “brain status” via spatial clustering, which ignores rich spatiotemporal dynamics contained in such chronnectome. We propose Deep Chronnectome Learning for exhaustively mining the comprehensive information, especially the hidden higher-level features, i.e., the dFC time series that may add critical diagnostic power for MCI classification. To this end, we devise a new Fully-connected bidirectional Long Short-Term Memory (LSTM) network (Full-BiLSTM) to effectively learn the periodic brain status changes using both past and future information for each brief time segment and then fuse them to form the final output. We have applied our method to a rigorously built large-scale multi-site database (i.e., with 164 data from NCs and 330 from MCIs, which can be further augmented by 25 folds). Our method outperforms other state-of-the-art approaches with an accuracy of 73.6% under solid cross-validations. We also made extensive comparisons among multiple variants of LSTM models. The results suggest high feasibility of our method with promising value also for other brain disorder diagnoses.

Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings
EditorsAlejandro F. Frangi, Christos Davatzikos, Gabor Fichtinger, Carlos Alberola-López, Julia A. Schnabel
PublisherSpringer Verlag
Pages249-257
Number of pages9
ISBN (Print)9783030009304
DOIs
Publication statusPublished - 2018 Jan 1
Externally publishedYes
Event21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 2018 Sep 162018 Sep 20

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11072 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period18/9/1618/9/20

Fingerprint

Memory Term
Brain
Connectivity
Diagnostics
Spatial Clustering
Multiple Comparisons
Functional Magnetic Resonance Imaging
Memory Model
Electric fuses
Cross-validation
Disorder
Time series
Mining
Time-varying
Fold
Learning
Long short-term memory
Deep learning
Output

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Yan, W., Zhang, H., Sui, J., & Shen, D. (2018). Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. In A. F. Frangi, C. Davatzikos, G. Fichtinger, C. Alberola-López, & J. A. Schnabel (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings (pp. 249-257). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11072 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-00931-1_29

Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. / Yan, Weizheng; Zhang, Han; Sui, Jing; Shen, Dinggang.

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. ed. / Alejandro F. Frangi; Christos Davatzikos; Gabor Fichtinger; Carlos Alberola-López; Julia A. Schnabel. Springer Verlag, 2018. p. 249-257 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11072 LNCS).

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

Yan, W, Zhang, H, Sui, J & Shen, D 2018, Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. in AF Frangi, C Davatzikos, G Fichtinger, C Alberola-López & JA Schnabel (eds), Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11072 LNCS, Springer Verlag, pp. 249-257, 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018, Granada, Spain, 18/9/16. https://doi.org/10.1007/978-3-030-00931-1_29
Yan W, Zhang H, Sui J, Shen D. Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. In Frangi AF, Davatzikos C, Fichtinger G, Alberola-López C, Schnabel JA, editors, Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Springer Verlag. 2018. p. 249-257. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-00931-1_29
Yan, Weizheng ; Zhang, Han ; Sui, Jing ; Shen, Dinggang. / Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. editor / Alejandro F. Frangi ; Christos Davatzikos ; Gabor Fichtinger ; Carlos Alberola-López ; Julia A. Schnabel. Springer Verlag, 2018. pp. 249-257 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{1ac5a764cf524bfea2ab5b4d9abb38c1,
title = "Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis",
abstract = "Brain functional connectivity (FC) extracted from resting-state fMRI (RS-fMRI) has become a popular approach for disease diagnosis, where discriminating subjects with mild cognitive impairment (MCI) from normal controls (NC) is still one of the most challenging problems. Dynamic functional connectivity (dFC), consisting of time-varying spatiotemporal dynamics, may characterize “chronnectome” diagnostic information for improving MCI classification. However, most of the current dFC studies are based on detecting discrete major “brain status” via spatial clustering, which ignores rich spatiotemporal dynamics contained in such chronnectome. We propose Deep Chronnectome Learning for exhaustively mining the comprehensive information, especially the hidden higher-level features, i.e., the dFC time series that may add critical diagnostic power for MCI classification. To this end, we devise a new Fully-connected bidirectional Long Short-Term Memory (LSTM) network (Full-BiLSTM) to effectively learn the periodic brain status changes using both past and future information for each brief time segment and then fuse them to form the final output. We have applied our method to a rigorously built large-scale multi-site database (i.e., with 164 data from NCs and 330 from MCIs, which can be further augmented by 25 folds). Our method outperforms other state-of-the-art approaches with an accuracy of 73.6{\%} under solid cross-validations. We also made extensive comparisons among multiple variants of LSTM models. The results suggest high feasibility of our method with promising value also for other brain disorder diagnoses.",
author = "Weizheng Yan and Han Zhang and Jing Sui and Dinggang Shen",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-00931-1_29",
language = "English",
isbn = "9783030009304",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "249--257",
editor = "Frangi, {Alejandro F.} and Christos Davatzikos and Gabor Fichtinger and Carlos Alberola-L{\'o}pez and Schnabel, {Julia A.}",
booktitle = "Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings",

}

TY - GEN

T1 - Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis

AU - Yan, Weizheng

AU - Zhang, Han

AU - Sui, Jing

AU - Shen, Dinggang

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Brain functional connectivity (FC) extracted from resting-state fMRI (RS-fMRI) has become a popular approach for disease diagnosis, where discriminating subjects with mild cognitive impairment (MCI) from normal controls (NC) is still one of the most challenging problems. Dynamic functional connectivity (dFC), consisting of time-varying spatiotemporal dynamics, may characterize “chronnectome” diagnostic information for improving MCI classification. However, most of the current dFC studies are based on detecting discrete major “brain status” via spatial clustering, which ignores rich spatiotemporal dynamics contained in such chronnectome. We propose Deep Chronnectome Learning for exhaustively mining the comprehensive information, especially the hidden higher-level features, i.e., the dFC time series that may add critical diagnostic power for MCI classification. To this end, we devise a new Fully-connected bidirectional Long Short-Term Memory (LSTM) network (Full-BiLSTM) to effectively learn the periodic brain status changes using both past and future information for each brief time segment and then fuse them to form the final output. We have applied our method to a rigorously built large-scale multi-site database (i.e., with 164 data from NCs and 330 from MCIs, which can be further augmented by 25 folds). Our method outperforms other state-of-the-art approaches with an accuracy of 73.6% under solid cross-validations. We also made extensive comparisons among multiple variants of LSTM models. The results suggest high feasibility of our method with promising value also for other brain disorder diagnoses.

AB - Brain functional connectivity (FC) extracted from resting-state fMRI (RS-fMRI) has become a popular approach for disease diagnosis, where discriminating subjects with mild cognitive impairment (MCI) from normal controls (NC) is still one of the most challenging problems. Dynamic functional connectivity (dFC), consisting of time-varying spatiotemporal dynamics, may characterize “chronnectome” diagnostic information for improving MCI classification. However, most of the current dFC studies are based on detecting discrete major “brain status” via spatial clustering, which ignores rich spatiotemporal dynamics contained in such chronnectome. We propose Deep Chronnectome Learning for exhaustively mining the comprehensive information, especially the hidden higher-level features, i.e., the dFC time series that may add critical diagnostic power for MCI classification. To this end, we devise a new Fully-connected bidirectional Long Short-Term Memory (LSTM) network (Full-BiLSTM) to effectively learn the periodic brain status changes using both past and future information for each brief time segment and then fuse them to form the final output. We have applied our method to a rigorously built large-scale multi-site database (i.e., with 164 data from NCs and 330 from MCIs, which can be further augmented by 25 folds). Our method outperforms other state-of-the-art approaches with an accuracy of 73.6% under solid cross-validations. We also made extensive comparisons among multiple variants of LSTM models. The results suggest high feasibility of our method with promising value also for other brain disorder diagnoses.

UR - http://www.scopus.com/inward/record.url?scp=85053926013&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053926013&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00931-1_29

DO - 10.1007/978-3-030-00931-1_29

M3 - Conference contribution

SN - 9783030009304

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

SP - 249

EP - 257

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings

A2 - Frangi, Alejandro F.

A2 - Davatzikos, Christos

A2 - Fichtinger, Gabor

A2 - Alberola-López, Carlos

A2 - Schnabel, Julia A.

PB - Springer Verlag

ER -