Multi-View Multi-Graph Embedding for Brain Network Clustering Analysis
Authors: Ye Liu, Lifang He, Bokai Cao, Philip Yu, Ann Ragin, Alex Leow
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on real HIV and bipolar disorder brain network datasets demonstrate the superior performance of M2E on clustering brain networks by leveraging the multi-view multigraph interactions. and Experiments and Evaluation In order to empirically evaluate the performance of the proposed M2E approach for multi-view multi-graph brain network clustering analysis, we test our model on two real datasets, HIV and Bipolar disorder with f MRI brain networks and DTI brain networks, and compare with several state-of-the-art multi-view clustering methods. |
| Researcher Affiliation | Academia | Ye Liu,1 Lifang He,2 Bokai Cao,1 Philip S. Yu,1,3 Ann B. Ragin,4 Alex D. Leow5 1Department of Computer Science, 5Department of Bioengineering, University of Illinois at Chicago, IL, USA 2Department of Healthcare Policy and Research, Cornell University, NY, USA 3Institute for Data Science, Tsinghua University, Beijing, China 4Department of Radiology, Northwestern University, IL, USA |
| Pseudocode | Yes | Algorithm 1 M2E Input: Partically-symmetric tensor X (v), weight parameters α1 and α2, and embedding dimension R 1: Initialize H(v), F(v) N(0, 1), U = 0, μ = 10 2: repeat 3: Update H(v) and P(v) by Eq. (14) and Eq. (16) 4: Update U(v) by Eq. (17) 5: Update F(v) by Eq. (19) 6: Update F by Eq. (20) 7: until convergence Output: Consensus embedding matrix F |
| Open Source Code | No | The paper refers to external tools and their websites (e.g., 'http://rfmri.org/DPARSF', 'http://www.l.ion.ucl.ac.uk/spm/software/spm8', 'https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/', 'http://www.nitrc.org/projects/conn'), but it does not state that the source code for the proposed M2E method is openly available, nor does it provide a link to such code. |
| Open Datasets | Yes | Human Immunodeficiency Virus Infection (HIV): The original dataset is unbalanced, we randomly sampled 35 patients and 35 controls from the dataset for performance evaluation. A detailed description about data acquisition is available in (Cao et al. 2015). |
| Dataset Splits | No | The paper states 'For evaluation, since there are two possible label values, normal and control, for each brain network sample on both HIV and BP datasets, we set the number of clusters K to be 2 and test how well our method can group the brain networks of patients and normal controls into two different clusters.' It also mentions 'We repeat this K-means clustering procedure 20 times with random initialization' and parameter tuning via grid search. However, it does not specify explicit train/validation/test dataset splits with percentages or sample counts for the overall datasets. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory, or processing power) used for running the experiments. |
| Software Dependencies | No | The paper mentions several software tools such as 'DPARSF', 'SPM', 'FSL toolbox', 'CONN', and 'Litekmeans function in Matlab', but it does not specify any version numbers for these software components. |
| Experiment Setup | Yes | There are three main parameters in our model, namely λ1, λ2 and R, where λ1 is the parameter of DTI view, λ2 is the parameter of f MRI view and R is the embedded dimension. We apply the grid search to determine the optimal values of these three parameters. In particular, we empirically select λ1 and λ2 from {10−4, 10−2, ..., 104}, and R is selected from {1, 2, ..., 20}. and For evaluation, since there are two possible label values, normal and control, for each brain network sample on both HIV and BP datasets, we set the number of clusters K to be 2 and test how well our method can group the brain networks of patients and normal controls into two different clusters. In order to make a fair comparison, we apply the Litekmeans function in Matlab (Cai 2011) for all the compared methods during their K-means clustering step. We repeat this K-means clustering procedure 20 times with random initialization, as Litekmeans greatly depends on initialization. |