New Fast ApEn and SampEn Entropy Algorithms Implementation and Their Application to Supercomputer Power Consumption

Tomčala, Jiří

doi:10.3390/e22080863

Cited by 26 publications

(21 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The parameter r was defined as 0.2 sd ( s ) where sd ( s ) is the standard deviation of the time series ( s ), as recommended by previous work. 41 An individual’s global entropy was calculated as the average of their regional entropies.…”

Section: Methodsmentioning

confidence: 99%

Larger lesion volume in people with multiple sclerosis is associated with increased transition energies between brain states and decreased entropy of brain activity

Tozlu

Card²,

Jamison

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Neuronal loss, demyelination, and inflammation in the brains of people with multiple sclerosis (pwMS) can cause impairment or disability across a range of domains. However, the mapping between severity or location of this brain tissue damage and an individual's impairment remains challenging. Advanced neuroimaging techniques may enable us to better understand the neuropathological or compensatory mechanisms in MS and quantify the link between the brain's functional activity and its structural backbone. One recently developed approach uses network control theory to characterize the brain's energetic landscape using an individual's structural connectome and brain activity dynamics. While brain activity dynamics have previously been analyzed in MS cohorts, investigating how structural network damage, one of the hallmarks of MS, relates to shifts in activity dynamics has not yet been performed. Aims: We apply brain activity dynamics analysis and network control theory (NCT) to functional time series and structural connectomes in healthy controls (HC) and pwMS with and without disability and compare the three groups' resulting metrics. We also assessed entropy of the three groups' brain activity time series. Our hypotheses were that pwMS with disability would have increased brain state transition energies compared to pwMS without disability and controls, and that greater energies would be related to greater lesion burden and decreased entropy of brain activity. Materials and Methods: Ninety-nine pwMS and 19 HC were included in our study; individuals underwent MRI, including anatomical, diffusion and resting-state functional scans. Expanded Disability Status Scale (EDSS) was used to assess disability; 66 pwMS with EDSS<2 were considered as not having disability. Commonly recurring brain states were identified using k-means clustering of the regional brain activity time series. NCT was used to compute the energy required to transition between each pair of states or to remain in the same state, taking into account an individual's structural connectome. The entropy of brain activity time series was quantified and correlated with overall transition energy and lesion volume. Finally, logistic regression with ridge regularization was used to classify the groups using transition energies and demographics/clinical information and the model's feature importance assessed. Results: Six brain states were identified, including pairs of states with high and low visual (VIS+/-), ventral attention (VAN)+/- and somatomotor (SOM+/-) network activity. Global entropy was negatively correlated with lesion volume (r=-0.20, p=0.03) as was global transition energy (r=-0.18, p=0.04) in pwMS. Smaller transition energies were associated with pwMS without disability, while larger transition energies were associated with pwMS with disability. Discussion: We conjecture that functional compensation of brain activity in pwMS without disability results in decreased transition energies between states compared to controls, but, as this mechanism diminishes, transition energies increase and disability occurs. We also conjecture that larger volumes of MS-related lesions result in greater energetic demands to transition between brain states which then leads to lower overall entropy of brain activity.

show abstract

Section: Methodsmentioning

confidence: 99%

Larger lesion volume in people with multiple sclerosis is associated with increased transition energies between brain states and decreased entropy of brain activity

Tozlu

Card²,

Jamison

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…When x is mostly selfsimilar, then [ ] and [ ] sequences are very close and thus is high. The ApEn is defined in equation ( 12) [103]:…”

Section: F Nonlinear Methodsmentioning

confidence: 99%

Feature Extraction Methods for Electroretinogram Signal Analysis: A Review

2021

View full text Add to dashboard Cite

Feature extraction is an essential aspect of Electroretinogram (ERG) signal analysis. The extracted features are beneficial to analyze the signal further and compress the signal for storage or transmission purposes. Various methods have been widely employed to extract the characteristics of ERG signals. Methods based on the time-domain, frequency-domain, time-frequency domain and nonlinear and chaotic feature extraction techniques have been used to extract features that characterize ERG signals. This paper reviews several feature extraction methods applied to ERG and compares their performance under different conditions to provide guidance to select the most appropriate feature extraction method based on the performance.

show abstract

“…To quantify the properties of latent dimensions, we have used a well-known information theory based algorithm suitable for time series datasets, called 'Fast Sample Entropy' (Tomčala, 2020). To compute Fast Sample Entropy, we have used the 'FastSampEn' function in the 'TSEntropies' package in R (Tomcala, 2018).…”

Section: Entropy Of the Latent Dimensionsmentioning

confidence: 99%

Genomic Surveillance of COVID-19 Variants with Language Models and Machine Learning

Nagpal

Pal

Ashima

et al. 2021

Preprint

View full text Add to dashboard Cite

The global efforts to control COVID-19 are threatened by the rapid emergence of novel variants that may display undesirable characteristics such as immune escape or increased pathogenicity. The current approaches to genomic surveillance do not allow early prediction of emerging variations. Here, we derive Dimensions of Concern (DoC) in the latent space of SARS-CoV-2 mutations and demonstrate their potential to provide a lead time for predicting the increase of new cases in 9 countries across the globe. We learned unsupervised word embeddings from 3,09,060 spike protein coding sequences deposited on GISAID database until April, 2021. We discovered that "blips" in the latent dimensions of embeddings are associated with mutations. We modeled the temporal occurrence of blips and their relationships with the number of new cases in the following months for these countries. Certain dimensions demonstrated a consistent leading relationship between the occurrence of blips and the number of new cases in the following months, thus labeled as potential Dimensions of Concern, DoCs. We validated the predictive importance of DoCs by performing Random Forest-based feature selection and modeling in a temporally split training, validation, testing regime. Twelve dimensions achieved statistical significance and achieved an R-squared of 37% for prediction of number of new cases in the following month. Biological exploration of DoCs revealed that dimensions 3 and 12 captures 3-mers CGG, ACG and CAC that are associated with known variants L452R, K417T and Q677H respectively. Learning and tracking DoCs is extensible to related challenges such as pandemic preparedness, immune escape, pathogenicity modeling and antimicrobial resistance.

show abstract

New Fast ApEn and SampEn Entropy Algorithms Implementation and Their Application to Supercomputer Power Consumption

Cited by 26 publications

References 5 publications

Larger lesion volume in people with multiple sclerosis is associated with increased transition energies between brain states and decreased entropy of brain activity

Larger lesion volume in people with multiple sclerosis is associated with increased transition energies between brain states and decreased entropy of brain activity

Feature Extraction Methods for Electroretinogram Signal Analysis: A Review

Genomic Surveillance of COVID-19 Variants with Language Models and Machine Learning

Contact Info

Product

Resources

About