SDA: a data-driven algorithm that detects functional states applied to the EEG of Guhyasamaja meditation

被引:1
|
作者
Mikhaylets, Ekaterina [1 ]
Razorenova, Alexandra M. [1 ,2 ]
Chernyshev, Vsevolod [1 ]
Syrov, Nikolay [3 ]
Yakovlev, Lev [3 ]
Boytsova, Julia [4 ]
Kokurina, Elena [4 ]
Zhironkina, Yulia [5 ]
Medvedev, Svyatoslav [4 ]
Kaplan, Alexander [3 ,6 ]
机构
[1] HSE Univ, Fac Comp Sci, Fac Econ Sci, Moscow, Russia
[2] Moscow State Univ Psychol & Educ, Ctr Neurocognit Res, MEG Ctr, Moscow, Russia
[3] Immanuel Kant Balt Fed Univ, Balt Ctr Neurotechnol & Artificial Intelligence, Kaliningrad, Russia
[4] Academician Natalya Bekhtereva Fdn, St Petersburg, Russia
[5] Save Tibet Fdn, Moscow, Russia
[6] Lomonosov Moscow State Univ, Lab Neurophysiol & Neurocomp Interfaces, Moscow, Russia
关键词
EEG; clustering; unsupervised data annotation; information value; meditation practice; Ward's method; functional states; change point detection; DYNAMICS;
D O I
10.3389/fninf.2023.1301718
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
The study presents a novel approach designed to detect time-continuous states in time-series data, called the State-Detecting Algorithm (SDA). The SDA operates on unlabeled data and detects optimal change-points among intrinsic functional states in time-series data based on an ensemble of Ward's hierarchical clustering with time-connectivity constraint. The algorithm chooses the best number of states and optimal state boundaries, maximizing clustering quality metrics. We also introduce a series of methods to estimate the performance and confidence of the SDA when the ground truth annotation is unavailable. These include information value analysis, paired statistical tests, and predictive modeling analysis. The SDA was validated on EEG recordings of Guhyasamaja meditation practice with a strict staged protocol performed by three experienced Buddhist practitioners in an ecological setup. The SDA used neurophysiological descriptors as inputs, including PSD, power indices, coherence, and PLV. Post-hoc analysis of the obtained EEG states revealed significant differences compared to the baseline and neighboring states. The SDA was found to be stable with respect to state order organization and showed poor clustering quality metrics and no statistical significance between states when applied to randomly shuffled epochs (i.e., surrogate subject data used as controls). The SDA can be considered a general data-driven approach that detects hidden functional states associated with the mental processes evolving during meditation or other ongoing mental and cognitive processes.
引用
收藏
页数:21
相关论文
共 44 条
  • [41] Analysis of Meditation vs. Sensory Engaged Brain States Using Shannon Entropy and Pearson's First Skewness Coefficient Extracted from EEG Data
    Davis, Joshua J. J.
    Kozma, Robert
    Schubeler, Florian
    SENSORS, 2023, 23 (03)
  • [42] Deep Learning Optimizes Data-Driven Representation of Soil Organic Carbon in Earth System Model Over the Conterminous United States
    Tao, Feng
    Zhou, Zhenghu
    Huang, Yuanyuan
    Li, Qianyu
    Lu, Xingjie
    Ma, Shuang
    Huang, Xiaomeng
    Liang, Yishuang
    Hugelius, Gustaf
    Jiang, Lifen
    Doughty, Russell
    Ren, Zhehao
    Luo, Yiqi
    FRONTIERS IN BIG DATA, 2020, 3
  • [43] A Sequential Pressure-Based Algorithm for Data-Driven Leakage Identification and Model-Based Localization in Water Distribution Networks
    Daniel, Ivo
    Pesantez, Jorge
    Letzgus, Simon
    Fasaee, Mohammad Ali Khaksar
    Alghamdi, Faisal
    Berglund, Emily
    Mahinthakumar, G.
    Cominola, Andrea
    JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT, 2022, 148 (06)
  • [44] Topological Filtering of Dynamic Functional Brain Networks Unfolds Informative Chronnectomics: A Novel Data-Driven Thresholding Scheme Based on Orthogonal Minimal Spanning Trees (OMSTs)
    Dimitriadis, Stavros I.
    Salis, Christos
    Tarnanas, Ioannis
    Linden, David E.
    FRONTIERS IN NEUROINFORMATICS, 2017, 11