Topological seizure origin detection in electroencephalographic signals

Wang, Yuan; Ombao, Hernando; Chung, Moo K.

doi:10.1109/isbi.2015.7163885

Cited by 9 publications

(7 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A measure of the importance of a node, which we dub nodal strength, is then obtained by summing the weights of the edges stemming from that node in the scaffold. This method has been applied to resting state fMRI data and has revealed topological correlates of altered states of consciousness (Petri et al, 2014) and epileptic seizures (Wang, Ombao, & Chung, 2015), as well as pointing to specific topological structures in resting state (Lord et al, 2016) and during attention modulation (Yoo, Kim, Ahn, & Ye, 2016).…”

Section: Topological Data Analysismentioning

confidence: 99%

Spectral and topological analyses of the cortical representation of the head position: Does hypnotizability matter?

et al. 2019

View full text Add to dashboard Cite

Introduction The aim of this exploratory study was to assess the EEG correlates of head positions (which have never been studied in humans) in participants with different psychophysiological characteristics, as encoded by their hypnotizability scores. This choice is motivated by earlier studies suggesting different processing of vestibular/neck proprioceptive information in subjects with high (highs) and low (lows) hypnotizability scores maintaining their head rotated toward one side ( RH ). Methods We analyzed EEG signals recorded in 20 highs and 19 lows in basal conditions (head forward) and during RH using spectral analysis, which captures changes localized to specific recording sites, and topological data analysis ( TDA ), which instead describes large‐scale differences in processing and representing sensorimotor information. Results Spectral analysis revealed significant differences related to head position for alpha 1, beta 2, beta 3, and gamma bands, but not to hypnotizability. TDA instead revealed global hypnotizability‐related differences in the strengths of the correlations among recording sites during RH . Significant changes were observed in lows on the left parieto‐occipital side and in highs in right frontoparietal region. Significant differences between the two groups were found in the occipital region, where changes were larger in lows than in highs . Conclusions This study reports finding of the EEG correlates of changes in the head posture for the first time, indicating that hypnotizability is related to the head posture representation/processing on large‐scale networks and that spectral and topological data analyses provide complementary results.

show abstract

Section: Topological Data Analysismentioning

confidence: 99%

Spectral and topological analyses of the cortical representation of the head position: Does hypnotizability matter?

et al. 2019

View full text Add to dashboard Cite

show abstract

“…We specifically chose descriptors that are simple summaries of PH barcodes so that we can interpret biological differences between networks. In other biological applications, barcodes have been successfully analyzed by their vectorization, for example, by using persistence images (58) or persistence landscapes (59,60) and classifying them using methods from machine learning; see, for example, (22,36,(61)(62)(63)(64). Here, this type of transformation is not suitable because of the variation of initial vasculature within a treatment group and small sample size common to mouse model experiments.…”

Section: Discussionmentioning

confidence: 99%

Multiscale topology characterizes dynamic tumor vascular networks

et al. 2022

View full text Add to dashboard Cite

Advances in imaging techniques enable high-resolution three-dimensional (3D) visualization of vascular networks over time and reveal abnormal structural features such as twists and loops, and their quantification is an active area of research. Here, we showcase how topological data analysis, the mathematical field that studies the “shape” of data, can characterize the geometric, spatial, and temporal organization of vascular networks. We propose two topological lenses to study vasculature, which capture inherent multiscale features and vessel connectivity, and surpass the single-scale analysis of existing methods. We analyze images collected using intravital and ultramicroscopy modalities and quantify spatiotemporal variation of twists, loops, and avascular regions (voids) in 3D vascular networks. This topological approach validates and quantifies known qualitative trends such as dynamic changes in tortuosity and loops in response to antibodies that modulate vessel sprouting; furthermore, it quantifies the effect of radiotherapy on vessel architecture.

show abstract

“…Thus, they might be more useful in discrete decision-making tasks such as clustering and classification. In fact, topological data analysis has begun to be more useful in deep learning ( Chen et al, 2019 ) and in identifying shared common features in time series ( Wang et al, 2015 , 2018 ).…”

Section: Lack Of Localizationmentioning

confidence: 99%