Article type :

1

Author :

Zayid Hazim Jasim,Wathiq Neamah Ajeel Hassan,Ruqayah Ali Mohsen Issa,Mezead Rahem Turke Hendi,Muhammad Hassan Abdul Razzaq Kazem

Volume :

2

Issue :

7

Abstract :

Abstract In recent years the algorithms of machine learning were used for brainsignals identification as a useful technique for diagnosing diseases likeAlzheimer's and epilepsy. In this paper, the Electroencephalogram (EEG)signals are classified using an optimized Quantum neural network (QNN) afternormalizing these signals. The wavelet transform (WT) and the independentcomponent analysis (ICA) were utilized for feature extraction. Thesealgorithms were used to reduce the dimensions of the data, which is an input tothe optimized QNN for the purpose of performing the classification processafter the feature extraction process. This research uses an optimized QNN, aform of feedforward neural network (FFNN), to recognize the EEG signals.The Particle swarm optimization (PSO) algorithm was used to optimize thequantum neural network, which improved the training process of the system'sperformance. The optimized QNN provided us with somewhat faster and morerealistic results. According to simulation results, the total classification for ICAis 82.4 percent, while the total classification for WT is 78.43 percent; fromthese results, using the ICA for feature extraction is better than using WT