Chaotic Dynamics in Tracing BAEP and its Application on Investigating Brainstem Malfunction

Area of Science:

• Neuroscience
• Dynamical Systems Theory
• Signal Processing

Background:

• Brainstem auditory evoked potential (BAEP) analysis is crucial for assessing auditory pathway function.
• Distinguishing between normal and abnormal BAEP signals is challenging due to noise.
• Understanding the underlying dynamics of BAEP may reveal subtle indicators of brainstem dysfunction.

Purpose of the Study:

• To investigate the chaotic dynamics of tracing BAEP.
• To identify brainstem malfunction using dynamic characteristics of BAEP.
• To differentiate between normal and malfunctioning brainstems based on BAEP signal complexity.

Main Methods:

• Utilized radial basis function neural network (RBFNN) and moving window average (MWA) for noise reduction in BAEP signals.
• Collected noisy BAEP data from infantile spasm (IS) patients (brainstem malfunction) and a control group.
• Employed phase projection and correlation dimension techniques to analyze BAEP chaotic dynamics.

Main Results:

• Cleaned BAEP signals showed stronger determination (lower D2) and more deterministic phase projections than noisy signals.
• BAEP trajectories were non-repeating, indicating fractal correlation dimensions.
• BAEP signals from the brainstem malfunction group exhibited higher chaoticity and higher D2 values compared to the control group.

Conclusions:

• BAEP signals are inherently chaotic, not deterministic, possessing rich dynamic properties.
• Tracing BAEP is more effective than noisy BAEP for diagnosing brainstem malfunction due to its deterministic nature.
• Brainstem malfunction is associated with increased non-order dynamics in BAEP signals.