BIOPOTENSIAL SIGNALLARNI (EMG, EEG) QAYTA ISHLASHDA MASHINAVIY O‘QITISH ALGORITMLARINING SAMARADORLIGI

Аннотация

This paper analyzes the efficiency of machine learning algorithms in processing biopotential signals, specifically electromyography (EMG) and electroencephalography (EEG) signals. The study demonstrates the effectiveness of algorithms such as SVM, random forest, and convolutional neural networks (CNN) in classifying signals and detecting pathological patterns. The complex nature of EMG and EEG signals, along with their susceptibility to noise and artifacts, complicates the signal processing procedure. Therefore, feature extraction methods, including RMS, spectral energy, and dominant frequency, play a crucial role in improving the accuracy and sensitivity of these algorithms. The results of this research confirm that machine learning algorithms exhibit high efficiency in processing biopotential signals and can be practically applied in human-computer interfaces, neuroprosthetic devices, and neurorehabilitation procedures. This study provides a scientific foundation for the development of artificial intelligence-based systems in the fields of medicine and biomedical engineering. Keywords: Biopotential signals, Electromyography (EMG), Electroencephalography (EEG), Machine learning, Convolutional neural networks (CNN), Feature extraction, Signal classification, Medical diagnostics.