Projects

Project: Brain Network Analysis

Supported by: “Learning Network Structures from Neuroimages for Diagnosing Brain Diseases”, ARC DECRA (Discovery Early Career Researcher Award) DE160100241, Australian Research Council

Many mental disorders, such as Alzheimer’s disease (AD), are associated with subtle abnormalities distributed over the entire brain, rather than an individual brain region. The “distributive” nature of mental disorders suggests the alteration of interactions between brain regions and thus the necessity of studying the brain as a complex network. This is known as brain network analysis, where a brain is usually abstracted mathematically as a probabilistic graph model governed by imaging-based features. My research in brain network analysis covers three types of brain connectivity: structural connectivity studying the physical links (such as white matter fibers) of brain regions, functional connectivity measuring the synchronization of imaging signals of brain regions, and effective connectivity investigating the influence one brain region exerts upon another.

Brain Effective Connectivity (Group Level)

L. Zhou, L. Wang, L. Liu, P. Ogunbona, and D. Shen, “Learning Discriminative Bayesian Networks from High-dimensional Continuous Neuroimaging Data”, IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2016.

In this paper, we propose to improve the discriminative power of Bayesian Network (BN) models for continuous variables from two different perspectives. This brings two general discriminative learning frameworks for Gaussian Bayesian networks (GBN). In the first framework, we employ Fisher kernel to bridge the generative models of GBN and the discriminative classifiers of SVMs, and convert the GBN parameter learning to Fisher kernel learning via minimizing a generalization error bound of SVMs. In the second framework, we employ the max-margin criterion and build it directly upon GBN models to explicitly optimize the classification performance of the GBNs. The advantages and disadvantages of the two frameworks are discussed and experimentally compared. Both of them demonstrate strong power in learning discriminative parameters of GBNs for neuroimaging based brain network analysis, as well as maintaining reasonable representation capacity. The contributions of this paper also include a new Directed Acyclic Graph (DAG) constraint with theoretical guarantee to ensure the graph validity of GBN.

Brain Functional Connectivity (Group Level)

L. Zhou, L. Wang, and P. Ogunbona, “Discriminative Sparse Inverse Covariance Matrix: Application in Brain Functional Network Classification”, IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), Ohio, USA, 2014

Recent studies show that mental disorders change the functional organization of the brain, which could be investigated via various imaging techniques. Analyzing such changes is becoming critical as it could provide new biomarkers for diagnosing and monitoring the progression of the diseases. Functional connectivity analysis studies the covary activity of neuronal populations in different brain regions. The sparse inverse covariance estimation (SICE), also known as graphical LASSO, is one of the most important tools for functional connectivity analysis, which estimates the interregional partial correlations of the brain. Although being increasingly used for predicting mental disorders, SICE is basically a generative method that may not necessarily perform well on classifying neuroimaging data. In this paper, we propose a learning framework to effectively improve the discriminative power of SICEs by taking advantage of the samples in the opposite class. We formulate our objective as convex optimization problems for both one-class and two-class classifications. By analyzing these optimization problems, we not only solve them efficiently in their dual form, but also gain insights into this new learning framework. The proposed framework is applied to analyzing the brain metabolic covariant networks built upon FDG-PET images for the prediction of the Alzheimer's disease, and shows significant improvement of classification performance for both one-class and two-class scenarios. Moreover, as SICE is a general method for learning undirected Gaussian graphical models, this paper has broader meanings beyond the scope of brain research.

Brain Functional Connectivity (Individual Level)

L. Zhou, L. Wang, J. Zhang, Y. Shi and Y. Gao, “Revisiting Distance Metric Learning for SPD Matrix based Visual Representation”, CVPR, Hawaii, USA, 2017

Symmetric Positive Definite (SPD) matrix is also used to represent individual brain functional connectivity extracted from fMRI. However, distance metric learning on SPD matrices has not been sufficiently researched. A few existing works approached this by learning either d^2xp or dxk transformation matrix for dxd SPD matrices. Different from these methods, this paper proposes a new member to the family of distance metric learning for SPD matrices. It learns only d parameters to adjust the eigenvalues of the SPD matrices through an efficient optimisation scheme. Also, it is shown that the proposed method can be interpreted as learning a sample-specific transformation matrix, instead of the fixed transformation matrix learned for all the samples in the existing works. The optimised d parameters can be used to "massage" the SPD matrices for better discrimination while still keeping them in the original space. From this perspective, the proposed method complements, rather than competes with, the existing linear-transformation-based methods, as the latter can always be applied to the output of the former to perform distance metric learning in further.

Integrate Brain Functional Connectivity and Structure Connectivity

J. Huang, L. Zhou, L. Wang, and D. Zhang, "Attention-Diffusion-Bilinear Neural Network for Brain Network Analysis", IEEE Transactions on Medical Imaging (IEEE T-MI), 2020

Integrative analysis of multiple types of connectivity, e.g, functional connectivity (FC) and structural connectivity (SC), can take advantage of their complementary information and therefore may help to identify patients. However, traditional brain network methods usually focus on either FC or SC for describing node interactions and only consider the interaction between paired network nodes. To tackle this problem, in this paper, we propose an Attention-Diffusion-Bilinear Neural Network (ADB-NN) framework for brain network analysis. The proposed network seamlessly couples FC and SC to learn wider node interactions and generates a joint representation of FC and SC for diagnosis. Specifically, a brain network (graph) is first defined, where each node corresponding to a brain region is governed by the features of brain activities (i.e., FC) extracted from fMRI, and the presence of edges is determined by neural fiber physical connections (i.e., SC) extracted from DTI. Based on this graph, we train two Attention-Diffusion-Bilinear (ADB) modules jointly. In each module, an attention model is utilized to automatically learn the strength of node interactions. This information further guides a diffusion process that generates new node representations by considering the influence from other nodes as well. After that, the second-order statistics of these node representations are extracted by bilinear pooling to form connectivity-based features for disease prediction. The two ADB modules correspond to the one-step and two-step diffusion, respectively.

Selected Related Publications:

J. Zhang, L. Zhou, L. Wang, M. Liu, and D. Shen, "Diffusion Kernel Attention Network for Brain Disorder Classification", IEEE Transactions on Medical Imaging (IEEE T-MI), 2022
J. Huang, L. Zhou, L. Wang, and D. Zhang, "Attention-Diffusion-Bilinear Neural Network for Brain Network Analysis", IEEE Transactions on Medical Imaging (IEEE T-MI), 2020
J. Huang, L. Zhou, D. Zhang, L. Wang, "Integrating Functional and Structural Connectivities via Diffusion-Convolution-Bilinear Neural Network", International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Shenzhen, China, 2019
J. Huang, Q. Zhu, M. Wang, L. Zhou, Z. Zhang, and D. Zhang, “Coherent Pattern in Multi-layer Brain Networks: Application to Epilepsy Identification”, IEEE Journal of Biomedical and Health Informatics (JBHI), 2019
L. Zhou, L. Wang, J. Zhang, Y. Shi and Y. Gao, “Revisiting Distance Metric Learning for SPD Matrix based Visual Representation”, CVPR, Hawaii, USA, 2017
J. Zhang, L. Zhou, and L. Wang, “Subject-adaptive Integration of Multiple SICE Brain Networks with Different Sparsity”, Pattern Recognition (PR), 2017
H. Ni, J. Qin, L. Zhou, Z. Zhao, J. Wang, and F. Hou, “Network analysis in detection of early-stage mild cognitive impairment”, Physica A: Statistical Mechanics and its Applications, 2017
L. Zhou, L. Wang, L. Liu, P. Ogunbona, and D. Shen, “Learning Discriminative Bayesian Networks from High-dimensional Continuous Neuroimaging Data”, IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2016.
J. Zhang, L. Wang, L. Zhou, and W. Li, “Learning Discriminative Stein Kernel for SPD Matrices and Its Applications”, IEEE Transactions on Neural Networks and Learning Systems (IEEE TNNLS), 2015
J. Zhang, L. Zhou, L. Wang, and W. Li, “Functional Brain Network Classification with Compact Representation of SICE Matrices”, IEEE Transactions on Biomedical Engineering (IEEE TBME), 2015
L. Wang, J. Zhang, L. Zhou, C. Tang, and W. Li, “Beyond Covariance: Feature Representation with Nonlinear Kernel Matrices”, ICCV, Santiago, Chile, 2015
L. Zhou, L. Wang, and P. Ogunbona, “Discriminative Sparse Inverse Covariance Matrix: Application in Brain Functional Network Classification”, CVPR, Ohio, USA, 2014
L. Zhou, L. Wang, L. Liu, P. Ogunbona, and D. Shen, “Max-margin Based Learning for Discriminative Bayesian Network from Neuroimaging Data”, MICCAI, Boston, USA, 2014
J. Zhang, L. Zhou, L. Wang, L. Li, “Exploring Compact Representation of SICE Matrices for Functional Brain Network Classification”, MICCAI Workshop on Machine Learning in Medical Imaging (MLMI), Boston, USA, 2014. (oral presentation)
L. Zhou, L. Wang, L. Liu, P. Ogunbona, and D. Shen, “Discriminative Brain Effective Connectivity Analysis for Alzheimers Disease: A Kernel Learning Approach upon Sparse Gaussian Bayesian Network”, CVPR, Oregon, USA, 2013
L. Zhou, Y. Wang, Y. Li, P.T. Yap, and D. Shen, “Hierarchical Anatomical Brain Networks for MCI Prediction by Partial Least Square Analysis”, CVPR, Colorado Springs, USA, June 2011

Host GLMI (Graph Learning in Medical Imaging) Workshop

Graph learning plays an essential role in brain network analysis. At the same time, it has been widely involved in computer-aided diagnosis, image segmentation, image registration, image fusion, image-guided therapy, image annotation and image database retrieval. To advance the technology in this field, we hosted the international workshop of Graph Learning in Medical Imaging (GLMI) with MICCAI 2019 in Shenzhen. This workshop focused on major trends and challenges in this area, and presented original work aimed to identify new cutting-edge techniques and their applications in medical imaging. The workshop weblink is: http://glmi.web.unc.edu/.

The proceedings of GLMI could be found online here.

Google Sites

Report abuse