Real-time estimation and detection of non-linearity in bio-signals using wireless brain-computer interface

Area of Science:

• Biomedical Engineering
• Signal Processing
• Computational Intelligence

Background:

• Physiological signals such as electroencephalogram (EEG) and electrocardiogram (ECG) often contain non-linear parameters.
• These non-linearities increase signal complexity, hindering accurate analysis and interpretation.
• Developing methods to linearize these signals is crucial for improved diagnostic capabilities.

Purpose of the Study:

• To present efficient, simple, and accurate techniques for removing non-linear parameters from physiological signals.
• To reduce the complexity of EEG and ECG signals for enhanced analysis.
• To explore the utility of transformation techniques, fuzzy logic, and adaptive neuro-fuzzy inference systems (ANFIS) in bio-signal processing.

Main Methods:

• Discrete Walsh-Hadamard Transform (DWHT) for signal transformation.
• Application of fuzzy logic control principles.
• Development of an Adaptive Neuro-Fuzzy Inference System (ANFIS) model.

Main Results:

• Demonstrated successful removal of non-linear parameters from EEG and ECG signals.
• Achieved simplification and reduction in the complexity of analyzed bio-signals.
• Validated the effectiveness of DWHT, fuzzy logic, and ANFIS for bio-signal processing.

Conclusions:

• The proposed methods offer efficient, simple, and accurate approaches to bio-signal analysis.
• Linearization of physiological signals through parameter removal enhances analytical outcomes.
• DWHT, fuzzy logic, and ANFIS are effective tools for processing complex bio-signals like EEG and ECG.