A neural network confirms that physical exercise reverses EEG changes in depressed rats

Area of Science:

• Neuroscience
• Behavioral Science

Background:

• Depression is a significant health concern with complex underlying mechanisms.
• Chronic stress is a known contributor to depressive symptoms.
• Physical exercise has shown potential therapeutic benefits for stress and depression.

Purpose of the Study:

• To investigate the utility of artificial neural networks (ANNs) in differentiating electroencephalogram (EEG) power density spectra between depressed and normal rats.
• To assess the impact of chronic physical exercise on stress-induced depression using EEG analysis and ANNs.

Main Methods:

• Rats were divided into four groups: chronic stress, chronic exercise, exercise with stress, and handling (control).
• Prefrontal cortical EEG, EMG, and EOG were recorded. EEG signals were digitized and analyzed for power spectral density using Fast Fourier Transform (FFT).
• An ANN with a specific architecture (30 first-layer neurons, majority-vote-taker second layer) was used to classify EEG data from REM, NREM, and awake states.

Main Results:

• The ANN achieved high accuracy in distinguishing depressed from normal rats' EEG: 99% in REM sleep, 95% in NREM sleep, and 81% in awake states.
• The ANN generally classified the EEG of rats subjected to exercise as normal, indicating a potential mitigating effect of exercise on stress-induced changes.
• The study demonstrates the feasibility of using ANNs for objective assessment of stress and depression via EEG analysis.

Conclusions:

• ANNs are effective tools for identifying depression-related EEG alterations in animal models.
• Chronic physical exercise may counteract the neurophysiological effects of chronic stress, potentially alleviating depressive states.
• This approach offers a promising avenue for objective biomarkers of depression and stress in preclinical research.