Coupled Stable Overlapping Replicator Dynamics for Multimodal Brain Subnetwork Identification

Yoldemir, Burak; Ng, Bernard; Abugharbieh, Rafeef

doi:10.1007/978-3-319-19992-4_61

Cited by 7 publications

(11 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most standard measure of correlation is the Pearson's correlation coefficient. However, in the case of N M , Pearson's correlation overestimates the number of correlated pairs of regions [197], especially when the data is noisy as is the case with fMRI [167]. Gellerup et al performed a comparison of twelve different measures of connectivity and found Pearson's correlation with Bonferroni multiple comparison correction, which reduces the number of false positives by imposing a very strict significance threshold, worked best [65].…”

Section: Functional Connectomesmentioning

confidence: 99%

“…Inaccurate placement of ROIs [160] and patient motion during scan, due to the long scan times required for both dMRI and fMRI [147,177,136], can affect connectomes of both modalities. Though, whereas structural connectomes tend to contain more false negative connections, due to tractography algorithms terminating too early in crossing and heavily curved white-matter regions, functional connectomes tend to contain more false positive connections, in part due to patient breathing and pulse (which affect blood oxygen levels across the brain) [197].…”

Section: Signal Noise and Biasmentioning

confidence: 99%

“…Anderson et al extracted the participation coefficient at each node, along with non-connectome features of cortical thickness and non-image-based metadata [3]. Lastly, some fMRI studies simply used the BOLD signal at each node as features [198,197]. In these cases, the connectome nature of the data was not explicitly encoded in the input features, but instead it was encoded into the learning models as prior information.…”

Section: Connectome Featuresmentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning on Human Connectome Data from MRI

Brown,

Hamarneh

2016

Preprint

View full text Add to dashboard Cite

Functional MRI (fMRI) and diffusion MRI (dMRI) are non-invasive imaging modalities that allow in-vivo analysis of a patient's brain network (known as a connectome). Use of these technologies has enabled faster and better diagnoses and treatments of neurological disorders and a deeper understanding of the human brain. Recently, researchers have been exploring the application of machine learning models to connectome data in order to predict clinical outcomes and analyze the importance of subnetworks in the brain. Connectome data has unique properties, which present both special challenges and opportunities when used for machine learning. The purpose of this work is to review the literature on the topic of applying machine learning models to MRI-based connectome data. This field is growing rapidly and now encompasses a large body of research. To summarize the research done to date, we provide a comparative, structured summary of 77 relevant works, tabulated according to different criteria, that represent the majority of the literature on this topic. (We also published a living version of this table online at http://connectomelearning.cs.sfu.ca/ that the community can continue to contribute to.) After giving an overview of how connectomes are constructed from dMRI and fMRI data, we discuss the variety of machine learning tasks that have been explored with connectome data. We then compare the advantages and drawbacks of different machine learning approaches that have been employed, discussing different feature selection and feature extraction schemes, as well as the learning models and regularization penalties themselves. Throughout this discussion, we focus particularly on how the methods are adapted to the unique nature of graphical connectome data. Finally, we conclude by summarizing the current state of the art and by outlining what we believe are strategic directions for future research.

show abstract

Section: Functional Connectomesmentioning

confidence: 99%

Section: Signal Noise and Biasmentioning

confidence: 99%

Section: Connectome Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning on Human Connectome Data from MRI

Brown,

Hamarneh

2016

Preprint

View full text Add to dashboard Cite

show abstract

“…Moreover, the neighborhood model always outperforms randomly generated graph with the same number of edges, indicating that the "expert" knowledge embedded in the graph is indeed a valuable prior to constrain the estimation of the model. Importantly, the flexibility of graphical models can allow us to test a number of prior graphs, including network-based graph generated from diffusion MRI and/or functional MRI network analyses Yoldemir et al (2015); Vergara et al (2016). This is certainly a topic we plan to investigate in future work.…”

Section: Discussionmentioning

confidence: 99%

“…Importantly, the flexibility of graphical models can allow us to test a number of prior graphs, including network-based graph generated from diffusion MRI and/or functional MRI network analyses Yoldemir et al (2015); Vergara et al (2016). This is certainly a topic we plan to investigate in future work.…”

Section: Discussionmentioning

confidence: 99%

Subject-specific abnormal region detection in traumatic brain injury using sparse model selection on high dimensional diffusion data

Shaker

Erdoğmuş

et al. 2017

Medical Image Analysis

View full text Add to dashboard Cite

We present a method to estimate a multivariate Gaussian distribution of diffusion tensor features in a set of brain regions based on a small sample of healthy individuals, and use this distribution to identify imaging abnormalities in subjects with mild traumatic brain injury. The multivariate model receives apriori knowledge in the form of a neighborhood graph imposed on the precision matrix, which models brain region interactions, and an additional L1 sparsity constraint. The model is then estimated using the graphical LASSO algorithm and the Mahalanobis distance of healthy and TBI subjects to the distribution mean is used to evaluate the discriminatory power of the model. Our experiments show that the addition of the apriori neighborhood graph results in significant improvements in classification performance compared to a model which does not take into account the brain region interactions or one which uses a fully connected prior graph. In addition, we describe a method, using our model, to detect the regions that contribute the most to the overall abnormality of the DTI profile of a subject’s brain.

show abstract