Tensors for neuroimaging

Erol, Aybüke; Hunyadi, Borbála

doi:10.1016/b978-0-12-824447-0.00018-2

Cited by 7 publications

(3 citation statements)

References 148 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ultimately, the complete decomposition of the swept-3D fUS data (of size 71280 × 21402) requires approximately half an hour when executed using MATLAB 2021a on a high-performance computing system running Linux, equipped with two AMD EPYC 32-Core Processors and 528 GB of memory. To enhance the algorithm's efficiency, additional constraints can be imposed, such as assuming a low-rank structure in space (Chatzichristos et al, 2019) or time (Erol and Hunyadi, 2022).…”

Section: Discussionmentioning

confidence: 99%

Evoked Component Analysis (ECA): Decomposing the Functional Ultrasound Signal with GLM-Regularization

Erol,

Generowicz,

Kruizinga

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Evoked Component Analysis (ECA): Decomposing the Functional Ultrasound Signal with GLM-Regularization

Erol,

Generowicz,

Kruizinga

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In recent years, special attention has been given to neuroimaging techniques for identifying locations of brain regions and their specific activation. Most commonly used are functional magnetic resonance imaging (fMRI), computerized tomography (CT), Positron emission tomography (PET) [ 34 , 35 ], and similar techniques with excellent spatial resolution, but they are very expensive and unavailable without experts in this field [ 36 ]. Electroencephalography (EEG), due to the development of new methods for EEG signal analysis, has begun to provide new insights into the field of cognitive neuroscience.…”

Section: Introductionmentioning

confidence: 99%

EEG Correlates of Cognitive Functions in a Child with ASD and White Matter Signal Abnormalities: A Case Report with Two-and-a-Half-Year Follow-Up

Ćirović,

Jeličić,

Maksimović

et al. 2023

Diagnostics

View full text Add to dashboard Cite

This research aimed to examine the EEG correlates of different stimuli processing instances in a child with ASD and white matter signal abnormalities and to investigate their relationship to the results of behavioral tests. The prospective case study reports two and a half years of follow-up data from a child aged 38 to 66 months. Cognitive, speech–language, sensory, and EEG correlates of auditory–verbal and auditory–visual–verbal information processing were recorded during five test periods, and their mutual interrelation was analyzed. EEG findings revealed no functional theta frequency range redistribution in the frontal regions favoring the left hemisphere during speech processing. The results pointed to a positive linear trend in the relative theta frequency range and a negative linear trend in the relative alpha frequency range when listening to and watching the cartoon. There was a statistically significant correlation between EEG signals and behavioral test results. Based on the obtained results, it may be concluded that EEG signals and their association with the results of behavioral tests should be evaluated with certain restraints considering the characteristics of the stimuli during EEG recording.

show abstract

“…As data is often represented in the form of tensors, tensor completion is an essential topic to many fields. Examples of its applications include visual data recovery [30], [52], super-resolution [35], big data analytics [36], anomaly detection [43], traffic data analysis [7], [40], [45], neuroimaging [12], or even recommender systems and knowledge graph completion [3]. Therefore, many academic topics are closely connected to tensor completion, and its improvement could lead to myriad potential advances in other fields.…”

Section: Introductionmentioning

confidence: 99%

Low-Rank Tensor Completion With Generalized CP Decomposition and Nonnegative Integer Tensor Completion

Yuan¹

2023

Preprint

View full text Add to dashboard Cite

The problem of tensor completion is important to many areas such as computer vision, data analysis, signal processing, etc. Previously, a category of methods known as low-rank tensor completion has been proposed and developed, involving the enforcement of low-rank structures on completed tensors. While such methods have been constantly improved, none have previously considered exploiting the numerical properties of tensor elements. This work attempts to construct a new methodological framework called GCDTC (Generalized CP Decomposition Tensor Completion) based on these properties. In this newly introduced framework, the CP Decomposition is reformulated as a Maximum Likelihood Estimate (MLE) problem, and generalized via the introduction of differing loss functions. The generalized decomposition is subsequently applied to low-rank tensor completion. Such loss functions can also be easily adjusted to consider additional factors in completion, such as smoothness, standardization, etc. An example of nonnegative integer tensor decomposition via the Poisson CP Decomposition is given to demonstrate the new methodology's potentials. Through experimentation with real-life data, it is confirmed that this method could produce results superior to current state-of-the-art methodologies. It is expected that the proposed notion would inspire a new set of tensor completion methods based on the generalization of decompositions, thus contributing to related fields.

show abstract

Tensors for neuroimaging

Cited by 7 publications

References 148 publications

Evoked Component Analysis (ECA): Decomposing the Functional Ultrasound Signal with GLM-Regularization

Evoked Component Analysis (ECA): Decomposing the Functional Ultrasound Signal with GLM-Regularization

EEG Correlates of Cognitive Functions in a Child with ASD and White Matter Signal Abnormalities: A Case Report with Two-and-a-Half-Year Follow-Up

Low-Rank Tensor Completion With Generalized CP Decomposition and Nonnegative Integer Tensor Completion

Contact Info

Product

Resources

About