arrow
返回

Exploration on Bubble Entropy

delete2025-07-28
delete0
delete
OA
AI
G
George Manis
D
Dimitrios Platakis
R
Roberto Sassi
DOI:10.1109/JBHI.2025.3593153delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Bubble entropy is a recently proposed entropy metric. Having certain advantages over popular definitions, bubble entropy finds its place in the research community map. It belongs to the family of entropy estimators which embed the signal into an m-dimensional space. Two are the main drawbacks for which those methods are criticized: the high computational cost and the dependence on parameters. Bubble entropy can be an answer to both, since computation can be performed in linear time and the dependence on parameters can be considered minimal in many practical situations. Popular entropy definitions, which are built over an embedding of the signal, mainly rely on two parameters: the size of the embedding space m and a tolerance r, which set a threshold over the distance between two points in the m-dimensional space to be considered similar. Bubble entropy totally eliminates the necessity to define a threshold distance, while it largely decouples the entropy estimation from the selection of the actual size of the embedding space in stationary conditions. Bubble entropy is compared to popular entropy definitions on theoretical and experimental basis. Theoretical analyses reveal significant advantages. Experimental analyses, comparing congestive heart failure patients and controls subjects, show that bubble entropy outperforms other popular, well established, entropy estimators in discriminating those two groups. Furthermore, machine learning-based feature ranking and experiments show that bubble entropy serves as a valuable source of features for AI decision-support algorithms.
Keyword:
Bubble entropy
biomedical engineering
HRV analysis
entropy estimator
embedding space
elimination of parameters
fast computation
permutation entropy
sample entropy
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

IEEE Journal of Biomedical and Health Informatics 封面图
IEEE Journal of Biomedical and Health Informatics
IF:
6.8
论文数:
4.5K
被引数:
2.0W

机构

U
università degli studi di milano
学者数:
1.1K
论文数: 415
被引数: 0
U
university of ioannina
学者数:
588
论文数: 294
被引数: 0
引用论文

引用论文

On the relation of KS entropy and permutation entropy
err2012-09-01
err0
errOAAI
errKarsten Keller; Anton M. Unakafov; Valentina A. Unakafova
err分享
err收藏
Permutation Entropy Based on Non-Uniform Embedding
err2018-08-17
err0
errOAAI
errMei Tao; Kristina Poskuviene; Nizar Faisal Alkayem; Maosen Cao; Minvydas Ragulskis
err分享
err收藏
err分享
err收藏
Assessment of Dispersion and Bubble Entropy Measures for Enhancing Preterm Birth Prediction Based on Electrohysterographic Signals
errSENSORS
IF3.5
err2021-09-10
err17
errOAAI
errNieto-del-Amor, Felix; Beskhani, Raja; Ye-Lin, Yiyao; Garcia-Casado, Javier; Diaz-Martinez, Alba; Monfort-Ortiz, Rogelio; Jose Diago-Almela, Vicente; Hao, Dongmei; Prats-Boluda, Gema
err分享
err收藏
err分享
err收藏
CEPS: An Open Access MATLAB Graphical User Interface (GUI) for the Analysis of Complexity and Entropy in Physiological Signals
err2021-03-08
err0
errOAAI
errDavid Mayor; Deepak Panday; Hari Kala Kandel; Tony Steffert; Duncan Banks
err分享
err收藏
学者 查看更多内容