Characterization of Hippocampal Local Field Potentials using Lyapunov Exponent Analysis and Unsupervised Machine...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Biophysics

Background:

• Hippocampal local field potentials (LFPs) reflect complex neural activity.
• Differentiating between basal and active brain states is crucial for understanding neural function and dysfunction.
• Existing methods may not fully capture the dynamic complexity of neural signals.

Purpose of the Study:

• To apply Lyapunov exponent (LE) analysis to characterize LFPs from hippocampal slices.
• To differentiate between basal and active hippocampal states using LE.
• To explore LE as a potential biomarker for neurophysiological and pathological conditions, including Alzheimer's disease.

Main Methods:

• Recordings of LFPs from hippocampal slices treated with kainic acid to induce active states.
• Signal pre-processing including standardization and bandpass filtering (4th-order Butterworth) for gamma oscillations.
• Lyapunov exponent (LE) analysis to assess signal dynamics and time-series clustering to identify state transitions.

Main Results:

• Positive LE values were identified, indicating chaotic dynamics predominantly in active hippocampal states.
• Time-series clustering successfully distinguished patterns in the progression from basal to active states.
• LE analysis demonstrated sensitivity to changes in hippocampal activity.

Conclusions:

• Lyapunov exponent analysis offers a novel approach to characterizing complex hippocampal dynamics.
• LE analysis can differentiate between basal and active brain states.
• LE analysis holds potential as an early diagnostic biomarker for neurodegenerative diseases like Alzheimer's.