PMID- 32508974
OWN - NLM
STAT- MEDLINE
DCOM- 20210426
LR  - 20231103
IS  - 1748-6718 (Electronic)
IS  - 1748-670X (Print)
IS  - 1748-670X (Linking)
VI  - 2020
DP  - 2020
TI  - Interpretable Learning Approaches in Resting-State Functional Connectivity 
      Analysis: The Case of Autism Spectrum Disorder.
PG  - 1394830
LID - 10.1155/2020/1394830 [doi]
LID - 1394830
AB  - Deep neural networks have recently been applied to the study of brain disorders 
      such as autism spectrum disorder (ASD) with great success. However, the internal 
      logics of these networks are difficult to interpret, especially with regard to 
      how specific network architecture decisions are made. In this paper, we study an 
      interpretable neural network model as a method to identify ASD participants from 
      functional magnetic resonance imaging (fMRI) data and interpret results of the 
      model in a precise and consistent manner. First, we propose an interpretable 
      fully connected neural network (FCNN) to classify two groups, ASD versus healthy 
      controls (HC), based on input data from resting-state functional connectivity 
      (rsFC) between regions of interests (ROIs). The proposed FCNN model is a 
      piecewise linear neural network (PLNN) which uses piecewise linear function 
      LeakyReLU as its activation function. We experimentally compared the FCNN model 
      against widely used classification models including support vector machine (SVM), 
      random forest, and two new classes of deep neural network models in a large 
      dataset containing 871 subjects from ABIDE I database. The results show the 
      proposed FCNN model achieves the highest classification accuracy. Second, we 
      further propose an interpreting method which could explain the trained model 
      precisely with a precise linear formula for each input sample and decision 
      features which contributed most to the classification of ASD versus HC 
      participants in the model. We also discuss the implications of our proposed 
      approach for fMRI data classification and interpretation.
CI  - Copyright (c) 2020 Jinlong Hu et al.
FAU - Hu, Jinlong
AU  - Hu J
AUID- ORCID: 0000-0003-3602-7603
AD  - School of Computer Science and Engineering, South China University of Technology, 
      Guangzhou, China.
AD  - Communication and Computer Network Laboratory of Guangdong, South China 
      University of Technology, Guangzhou, China.
FAU - Cao, Lijie
AU  - Cao L
AUID- ORCID: 0000-0002-6763-8768
AD  - School of Computer Science and Engineering, South China University of Technology, 
      Guangzhou, China.
AD  - Communication and Computer Network Laboratory of Guangdong, South China 
      University of Technology, Guangzhou, China.
FAU - Li, Tenghui
AU  - Li T
AUID- ORCID: 0000-0002-8810-962X
AD  - School of Computer Science and Engineering, South China University of Technology, 
      Guangzhou, China.
AD  - Communication and Computer Network Laboratory of Guangdong, South China 
      University of Technology, Guangzhou, China.
FAU - Liao, Bin
AU  - Liao B
AUID- ORCID: 0000-0003-3422-9092
AD  - College of Mathematics and Informatics, South China Agricultural University, 
      Guangzhou, China.
FAU - Dong, Shoubin
AU  - Dong S
AUID- ORCID: 0000-0003-0153-850X
AD  - School of Computer Science and Engineering, South China University of Technology, 
      Guangzhou, China.
AD  - Communication and Computer Network Laboratory of Guangdong, South China 
      University of Technology, Guangzhou, China.
FAU - Li, Ping
AU  - Li P
AD  - Faculty of Humanities, The Hong Kong Polytechnic University, Hong Kong, China.
LA  - eng
PT  - Journal Article
DEP - 20200518
PL  - United States
TA  - Comput Math Methods Med
JT  - Computational and mathematical methods in medicine
JID - 101277751
SB  - IM
MH  - Autism Spectrum Disorder/classification/*diagnostic imaging/physiopathology
MH  - Case-Control Studies
MH  - Computational Biology
MH  - Connectome/statistics & numerical data
MH  - Databases, Factual
MH  - *Deep Learning
MH  - Functional Neuroimaging/statistics & numerical data
MH  - Humans
MH  - Linear Models
MH  - Magnetic Resonance Imaging/statistics & numerical data
MH  - Neural Networks, Computer
MH  - Support Vector Machine
PMC - PMC7251440
COIS- The authors have nothing to disclose.
EDAT- 2020/06/09 06:00
MHDA- 2021/04/27 06:00
PMCR- 2020/05/18
CRDT- 2020/06/09 06:00
PHST- 2020/03/26 00:00 [received]
PHST- 2020/05/05 00:00 [accepted]
PHST- 2020/06/09 06:00 [entrez]
PHST- 2020/06/09 06:00 [pubmed]
PHST- 2021/04/27 06:00 [medline]
PHST- 2020/05/18 00:00 [pmc-release]
AID - 10.1155/2020/1394830 [doi]
PST - epublish
SO  - Comput Math Methods Med. 2020 May 18;2020:1394830. doi: 10.1155/2020/1394830. 
      eCollection 2020.