A Software for Identification and Characterization of Theta Rhythms in the Hippocampus

Abstract

Characterizing theta rhythms in the hippocampus provides a window into understanding memory processing. An inquiry that arises when an animal sustains a pathological state is how theta rhythms are affected. In pathological states like epilepsy or Alzheimers, these rhythms change in specific ways. Statistically robust changes in these rhythms could serve as potential biomarkers, indicating the severity of the animals condition and the effectiveness of a drug. However, this understanding depends on how the data is analyzed. There are currently no standard criteria for recognizing theta dominance in experimental recordings. To address this, we have developed novel MATLAB-based software with an easy-to-use graphical user interface which enables identifying and analyzing theta rhythms in a standard way. We discuss the softwares functionality and its underlying algorithms. The algorithms were developed using previously acquired EEG/LFP data recorded from the hippocampus of a mouse kindling model of epilepsy. Two primary analyses were conducted to test the softwares functionality: first, comparing theta rhythms during the baseline period versus during spontaneous recurrent seizures; second, analyzing the timing of theta rhythms relative to the seizure event. Our illustrative results indicate that our developed software can robustly identify theta events with statistically significant feature differences. Further, the examination presented here with two mice shows that while theta events can occur just before seizures, it takes tens of minutes post-seizure before theta rhythms occur again. Our software thus provides the user with the ability to robustly identify and characterize theta rhythms and their feature changes. Significance StatementTheta rhythms in the hippocampus are fundamental for spatial navigation and memory formation. Their observed changes during several pathological states such as epilepsy and Alzheimers make them highly interesting to be able to serve as biomarkers. However, their variability (in terms of the time of occurrence, duration, and frequency range) makes them challenging to quantify. The absence of available tools for the automatic extraction of these rhythms from extended datasets significantly hampers processing efficiency and reduces accuracy and consistency. Clinicians and researchers often manually inspect their data to identify these rhythms, a process that is not only time-consuming but also inherently subjective. We have thus developed a MATLAB software for precise, automatic analysis of theta rhythms in EEG/LFP recordings.