Wavelet feature extraction and J48 decision tree classification of auditory late response (ALR) elicited by transcranial magnetic stimulation

• Main Authors: W Azlan, Wan Amirah, Low, Yin Fen, Liew, Siaw Hong, Choo, Yun Huoy, Zakaria, Hazli
• Format: Article
• Language: English
• Published: Asian Research Publishing Network (ARPN) 2016
• Subjects: T Technology (General); TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utem.edu.my/id/eprint/17708/1/Wavelet%20Feature%20Extraction%20And%20J48%20Decision%20Tree%20Classification%20Of%20Auditory%20Late%20Response%20%28ALR%29%20Elicited%20By%20Transcranial%20Magnetic%20Stimulation.pdf; http://eprints.utem.edu.my/id/eprint/17708/; http://www.arpnjournals.org/jeas/research_papers/rp_2016/jeas_0516_4272.pdf

Nowadays, transcranial magnetic stimulation (TMS) has been used to treat major depression and migraine. Integrating transcranial magnetic stimulation and electroencephalogram (TMS - EEG) may provide beneficial information. This paper introduces the experimental design, experimental setup and experimental procedures to differentiate the repetitive transcranial magnetic stimulation (rTMS) and without TMS over N100 (N1) and P200 (P2) peaks with regards to auditory attention. New experimental design, setup and procedures are developed to elicit N1 and P2 through the recording of EEG signal with the excitation of neurons from TMS and pure tones. Wavelet transform is implemented as feature extraction for the selected data. Four features are used for the classification. The classification is based on J48 decision tree performed using WEKA to distinguish between without TMS and rTMS. The result between without TMS and rTMS (in attention condition) showed 98.85% accuracy meanwhile between without TMS and rTMS (no attention condition) showed 99.46% accuracy.