Modeling task-based fMRI data via deep belief network with neural architecture search

It has been shown that deep neural networks are powerful and flexible models that can be applied on fMRI data with superb representation ability over traditional methods. However, a challenge of neural network architecture design has also attracted attention: due to the high dimension of fMRI volume images, the manual process of network model design is very time-consuming and not optimal. To tackle this problem, we proposed an unsupervised neural architecture search (NAS) framework on a deep belief network (DBN) that models volumetric fMRI data, named NAS-DBN. The NAS-DBN framework is based on Particle Swarm Optimization (PSO) where the swarms of neural architectures can evolve and converge to a feasible optimal solution. The experiments showed that the proposed NAS-DBN framework can quickly find a robust architecture of DBN, yielding a hierarchy organization of functional brain networks (FBNs) and temporal responses. Compared with 3 manually designed DBNs, the proposed NAS-DBN has the lowest testing loss of 0.0197, suggesting an overall performance improvement of up to 47.9 %. For each task, the NAS-DBN identified 260 FBNs, including task-specific FBNs and resting state networks (RSN), which have high overlap rates to general linear model (GLM) derived templates and independent component analysis (ICA) derived RSN templates. The average overlap rate of NAS-DBN to GLM on 20 task-specific FBNs is as high as 0.536, indicating a performance improvement of up to 63.9 % in respect of network modeling. Besides, we showed that the NAS-DBN can simultaneously generate temporal responses that resemble the task designs very well, and it was observed that widespread overlaps between FBNs from different layers of NAS-DBN model form a hierarchical organization of FBNs. Our NAS-DBN framework contributes an effective, unsupervised NAS method for modeling volumetric task fMRI data.

[1]  Stephen M. Smith,et al.  General multilevel linear modeling for group analysis in FMRI , 2003, NeuroImage.

[2]  Yu Zhao,et al.  Modeling task fMRI data via mixture of deep expert networks , 2018, 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018).

[3]  Junwei Han,et al.  Signal sampling for efficient sparse representation of resting state FMRI data , 2015, 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI).

[4]  Elliot Meyerson,et al.  Evolving Deep Neural Networks , 2017, Artificial Intelligence in the Age of Neural Networks and Brain Computing.

[5]  V. Calhoun,et al.  Multisubject Independent Component Analysis of fMRI: A Decade of Intrinsic Networks, Default Mode, and Neurodiagnostic Discovery , 2012, IEEE Reviews in Biomedical Engineering.

[6]  Alok Aggarwal,et al.  Regularized Evolution for Image Classifier Architecture Search , 2018, AAAI.

[7]  Qing Li,et al.  Identify Hierarchical Structures from Task-Based fMRI Data via Hybrid Spatiotemporal Neural Architecture Search Net , 2019, MICCAI.

[8]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[9]  Stephen M. Smith,et al.  Investigations into resting-state connectivity using independent component analysis , 2005, Philosophical Transactions of the Royal Society B: Biological Sciences.

[10]  Tianming Liu,et al.  Large-Scale Circuitry Interactions Upon Earthquake Experiences Revealed by Recurrent Neural Networks , 2018, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[11]  Han Wang,et al.  Recognizing Brain States Using Deep Sparse Recurrent Neural Network , 2019, IEEE Transactions on Medical Imaging.

[12]  J. Pekar,et al.  A method for making group inferences from functional MRI data using independent component analysis , 2001, Human brain mapping.

[13]  Vince D. Calhoun,et al.  A review of group ICA for fMRI data and ICA for joint inference of imaging, genetic, and ERP data , 2009, NeuroImage.

[14]  M. McKeown Detection of Consistently Task-Related Activations in fMRI Data with Hybrid Independent Component Analysis , 2000, NeuroImage.

[15]  Yul-Wan Sung,et al.  Functional magnetic resonance imaging , 2004, Scholarpedia.

[16]  Xiang Li,et al.  Modeling 4D fMRI Data via Spatio-Temporal Convolutional Neural Networks (ST-CNN) , 2018, MICCAI.

[17]  Yu Zhao,et al.  Automatic recognition of holistic functional brain networks using iteratively optimized convolutional neural networks (IO‐CNN) with weak label initialization , 2018, Medical Image Anal..

[18]  Quoc V. Le,et al.  Neural Architecture Search with Reinforcement Learning , 2016, ICLR.

[19]  Yu Zhao,et al.  Modeling Task fMRI Data Via Deep Convolutional Autoencoder , 2018, IEEE Transactions on Medical Imaging.

[20]  Heng Huang,et al.  Sparse representation of whole-brain fMRI signals for identification of functional networks , 2015, Medical Image Anal..

[21]  Vince D. Calhoun,et al.  Restricted Boltzmann machines for neuroimaging: An application in identifying intrinsic networks , 2014, NeuroImage.

[22]  Jieping Ye,et al.  Holistic Atlases of Functional Networks and Interactions Reveal Reciprocal Organizational Architecture of Cortical Function , 2015, IEEE Transactions on Biomedical Engineering.

[23]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[24]  Vince D. Calhoun,et al.  Deep learning for neuroimaging: a validation study , 2013, Front. Neurosci..

[25]  Li Fei-Fei,et al.  Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  Jinglei Lv,et al.  Characterizing and differentiating task-based and resting state fMRI signals via two-stage sparse representations , 2016, Brain Imaging and Behavior.

[27]  Quoc V. Le,et al.  Large-Scale Evolution of Image Classifiers , 2017, ICML.

[28]  Heng Huang,et al.  Deep Neural Networks for In Situ Hybridization Grid Completion and Clustering , 2020, IEEE/ACM Transactions on Computational Biology and Bioinformatics.

[29]  Christian Igel,et al.  An Introduction to Restricted Boltzmann Machines , 2012, CIARP.

[30]  Dinggang Shen,et al.  State-space model with deep learning for functional dynamics estimation in resting-state fMRI , 2016, NeuroImage.

[31]  Alan L. Yuille,et al.  Genetic CNN , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[32]  Bo Peng,et al.  Latent source mining in FMRI via restricted Boltzmann machine , 2018, Human brain mapping.

[33]  Vincent J Schmithorst,et al.  Comparison of three methods for generating group statistical inferences from independent component analysis of functional magnetic resonance imaging data , 2004, Journal of magnetic resonance imaging : JMRI.

[34]  Ramesh Raskar,et al.  Accelerating Neural Architecture Search using Performance Prediction , 2017, ICLR.

[35]  N. Logothetis What we can do and what we cannot do with fMRI , 2008, Nature.

[36]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[37]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[38]  Jinglei Lv,et al.  Sparsity-Constrained fMRI Decoding of Visual Saliency in Naturalistic Video Streams , 2015, IEEE Transactions on Autonomous Mental Development.

[39]  Jinglei Lv,et al.  Modeling Hierarchical Brain Networks via Volumetric Sparse Deep Belief Network , 2020, IEEE Transactions on Biomedical Engineering.

[40]  Jinglei Lv,et al.  Experimental Comparisons of Sparse Dictionary Learning and Independent Component Analysis for Brain Network Inference From fMRI Data , 2019, IEEE Transactions on Biomedical Engineering.

[41]  Stephen M Smith,et al.  Correspondence of the brain's functional architecture during activation and rest , 2009, Proceedings of the National Academy of Sciences.

[42]  J. Lv,et al.  Assessing effects of prenatal alcohol exposure using group-wise sparse representation of fMRI data , 2015, Psychiatry Research: Neuroimaging.

[43]  Oriol Vinyals,et al.  Hierarchical Representations for Efficient Architecture Search , 2017, ICLR.

[44]  Quoc V. Le,et al.  Efficient Neural Architecture Search via Parameter Sharing , 2018, ICML.

[45]  Lei Guo,et al.  Identifying Brain Networks of Multiple Time Scales via Deep Recurrent Neural Network , 2018, MICCAI.

[46]  Michael Eickenberg,et al.  Machine learning for neuroimaging with scikit-learn , 2014, Front. Neuroinform..

[47]  Wei Zhang,et al.  Task fMRI data analysis based on supervised stochastic coordinate coding , 2017, Medical Image Anal..

[48]  Vijay Vasudevan,et al.  Learning Transferable Architectures for Scalable Image Recognition , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[49]  Abraham Z. Snyder,et al.  Function in the human connectome: Task-fMRI and individual differences in behavior , 2013, NeuroImage.