Detecting and characterizing high-frequency oscillations in epilepsy: a case study of big data analysis
Liang Huang1, Xuan Ni2, William L Ditto3
1School of Physical Science and Technology , Lanzhou University , Lanzhou , Gansu 730000 , People's Republic of China.
Royal Society Open Science
|March 11, 2017
Summary
We developed a novel framework to analyze complex time series data, revealing that high-frequency oscillations in rat brain recordings exhibit intermittent behavior governed by scaling laws.
Area of Science:
- Data Science
- Neuroscience
- Complex Systems
Background:
- Analyzing massive, nonlinear, and non-stationary time series data presents significant challenges.
- Identifying dynamical anomalies within such datasets requires advanced analytical approaches.
Purpose of the Study:
- To develop and validate a robust framework for uncovering and analyzing dynamical anomalies in large-scale time series data.
- To characterize the behavior of high-frequency oscillations (HFOs) in rat electroencephalogram (EEG) recordings.
Main Methods:
- A three-step framework involving data preprocessing, empirical mode decomposition (EMD) with Hilbert transform, and statistical/scaling analysis.
- Application of the framework to a large database of rat EEG recordings to detect and characterize HFOs.
Main Results:
- The developed framework successfully identified dynamical anomalies in massive time series data.
- High-frequency oscillations (HFOs) in rat EEG data exhibit on-off intermittency.
- This intermittency can be precisely quantified using algebraic scaling laws.
Conclusions:
- The proposed framework is effective for analyzing complex dynamical systems and identifying anomalous behaviors.
- The findings on HFO intermittency provide new insights into brain dynamics.
- The framework's generalizability extends to diverse big data applications, including sensor and seismic data analysis.
