Classification of structural brain networks based on information divergence of graph spectra

Dodonova, Yulia; Korolev, S.I.; Tkachev, Anna; Zhukov, Leonid; Belyaev, Mikhail

doi:10.1109/mlsp.2016.7738852

Cited by 7 publications

(10 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The possible explanation is in much higher dimensionality combined with greater noise in the data. The latter statement is partially supported by the fact that typical quality of classification for both datasets [21,9] is much lower than for ADNI2 database considered in this work.…”

Section: Datasupporting

confidence: 60%

“…The problem of brain network classification has been paid much attention recently [17,9,21,19,18]. This problem is non-trivial as most modern classification algorithms can work only with vectorial data while in our case each object in the dataset is represented by graph.…”

Section: Existing Approachesmentioning

confidence: 99%

“…The second important question is what to do with the matrix in order to use it as an input to a classification algorithm. Several approaches have been considered in the literature, including vectorization of the whole matrix, considering the vector of node degrees of a graph [21] or the vector of eigenvalues [9], and different variants of constructing the kernel between graphs [17,9]. We note, that in all these approaches features are generated for each connectome independently.…”

Section: Existing Approachesmentioning

confidence: 99%

See 2 more Smart Citations

Simultaneous Matrix Diagonalization for Structural Brain Networks Classification

Mokrov¹,

Panov²,

Gutman³

et al. 2017

Complex Networks &Amp; Their Applications VI

View full text Add to dashboard Cite

This paper considers the problem of brain disease classification based on connectome data. A connectome is a network representation of a human brain. The typical connectome classification problem is very challenging because of the small sample size and high dimensionality of the data. We propose to use simultaneous approximate diagonalization of adjacency matrices in order to compute their eigenstructures in more stable way. The obtained approximate eigenvalues are further used as features for classification. The proposed approach is demonstrated to be efficient for detection of Alzheimer's disease, outperforming simple baselines and competing with state-of-the-art approaches to brain disease classification.

show abstract

Section: Datasupporting

confidence: 60%

Section: Existing Approachesmentioning

confidence: 99%

Section: Existing Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Simultaneous Matrix Diagonalization for Structural Brain Networks Classification

Mokrov¹,

Panov²,

Gutman³

et al. 2017

Complex Networks &Amp; Their Applications VI

View full text Add to dashboard Cite

show abstract

“…For instance, based on a large body of research and predictable dimensionality reduction [34], [35], it is common to use for classification of functional connectivity matrices [36]- [38] representing correlations in time-series between pre-defined regions derived from blood oxygenation leveldependent (BOLD) sensitive fMRI. Likewise, to classify diffusion weighted images (DWI) it is common to use structural connectivity matrices representing the number of white matter tracts traversing the brain between specific regions [39]- [41].…”

Section: A Machine Learningmentioning

confidence: 99%

Single-participant structural connectivity matrices lead to greater accuracy in classification of participants than function in autism in MRI

Leming,

Baron-Cohen,

Suckling

2020

Preprint

View full text Add to dashboard Cite

“…This kernel improved the classification accuracy of SVM to 71% on a public dataset. This work had been done in 2016 [8].Petrov D et all in 2016, a data pre-processing method that utilized geometric and topological connectome normalization was put forth. The work had a claim that an improvement had been done in SVM classification of ASD and TD [9].…”

Section: Related Workmentioning

confidence: 99%

Altered Structural Connectivity in Autism Spectrum Disorder

Ramani,

Jabarani

2019

IJEAT

View full text Add to dashboard Cite

Research in Neurological field has been in great trend in recent days, since the need of detection and treatment of various neuropsychological disorders are in increasing order. Automated approaches for the detection are possible by various technological methods. Autism Spectrum Disorder (ASD) is a one such serious disorder which can be diagnosed in early ages of children. The Emerging technology had contributed the neuro imaging techniques to understand the various basic features and characteristics that cause the disorder. This neuro imaging had lead to a better perspective called connectome analysis which deals network structures (connectome) derived from the neuro images and are used in detection and treatment of the disorder. For these analysis functional and structural connectomes / network of brain are utilized. In this work structural connectomes derived from the Diffusion Tensor Imaging of Typically Developing and Autism Spectrum Disordered had been considered . This connectome / network consists of 264 regions (based on PowerNeuron_264 atlas) and thus 69696 connectivity features (connection between regions). Using the structural connectomes, average connectome analysis had been done and 91 connections had been identified as altered in ASD. There are 112 distinct regions involved in these altered connections and are having varied number of altered connections from one to six. 15 regions among them found to have much alteration since more number of (More than 2) altered connectivity are involved with these regions. To prove the finding , Data mining technique, Support Vector Machine was applied over 42 connectivity features (0.06% of original) out of 91 and are involved with the 15 regions filtered and the classification is done (with 82% accuracy) . Classifier rules are utilized in the diagnosis of ASD . The 15 regions extracted through thisprocess are found to be altered in ASD. These altered regions are related to sensory(touch and taste), memory, movements control, Lexical processing, Consciousness and sleep. This proposed system surely have effective use in the process of high dimensional and complex brain data and the identification of typically developed and autism spectrum disordered brain .This methodology can also be used in detection of other diseases, Role of various Regions, influential regions, etc.,

show abstract

Classification of structural brain networks based on information divergence of graph spectra

Cited by 7 publications

References 6 publications

Simultaneous Matrix Diagonalization for Structural Brain Networks Classification

Simultaneous Matrix Diagonalization for Structural Brain Networks Classification

Single-participant structural connectivity matrices lead to greater accuracy in classification of participants than function in autism in MRI

Altered Structural Connectivity in Autism Spectrum Disorder

Contact Info

Product

Resources

About