Classification of Electroencephalograms Using Machine Learning AlgorithmsName : Dr. Jose Joaquín Rodríguez García
Affliation : Professor
University : University of Alicante
Country : Spain
This work is a compilation of applications of techniques and mathematical tools, together with the scientific basis that gives us the field of medicine, to electroencephalogram (EEG) data collected in healthy children and children with attention deficit and hyperactivity disorder (ADHD). The objective of this study is to automatically diagnose ADHD in children using the EEG data collected from 85 children with ADHD and 83 healthy children.
Materials & Methods: The Fourier Transform, signal filtering, Independent Component Analysis (ICA), discrete Wavelet Transform, k-nearest neighbors (KNN) algorithm and k-fold cross validation.
Results: The results obtained have been a classification whose accuracy has been higher than 80%.
Conclusions: There is a remarkable relationship between the EEG data from a patient and to be diagnosed as ADHD. It opens the possibility of repeating the same study with a greater number of patients and other different mental pathologies.
Biography: Jose Joaquín was born in Villarrobledo, a town located near Albacete in Spain. From an early age, he became interested in electronic devices, so when he finished high school he began to study telecommunications engineering in the University. After that he was working in Holland during a year and a half. He then began and finished his mathematics studies at his age of 26.