Portal de Periódicos da CAPES

Algorithm for Medical Diagnostic Support Using Machine and Deep Learning for Depressive Disorder Based on Electroencephalogram Readings

2023; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-031-32213-6_23

1865-0937

Lady L. González, Giovana F. Vaca, M. Figueroa, Adriana E. Estrella, Evelyn G. González, Carolina Cadena-Morejón, Diego Almeida-Galárraga, Andrés Tirado-Espín, Jonathan Cruz-Varela, Fernando Villalba-Meneses,

Functional Brain Connectivity Studies

Depression is one of the most common mental disorders affecting 121 million people worldwide. Depression is more than a low mood and those who suffer from it can experience a lack of interest in daily activities, lack of concentration, low energy, feelings of worthlessness and in the worst cases, it can lead to suicide. For this reason, correct detection of the disorder is essential to reduce the number of cases of misdiagnosed people. In addition to psychological analysis, EEG signals are also one of the tools that help in the detection of mental disorders, such as depressive disorder. Therefore, the purpose of this study is to develop an algorithm for the detection of depressive disorder based on the classification of EEG signals. For this purpose, machine learning was used with the Welch method and four different classifiers, which are: LDA, LR, KNN and RFC. Also was used neural network that combines (IC-RNN) and (C-DRNN). Despite working with few data from only 26 depressed patients and 29 healthy patients, it could be obtained an accuracy of 57%.