Depression Diagnosis Using Linear Features Based on EEG Signals

Depression is one of the major public health problem affecting society and has become a world issue. It is creating a negative impact on social and economical behavior on the society. It is defined as mental illness of human being causing mood swing, low appetite, health issues etc. Clinical diagnosis used interview questions as well as some laboratory tests for diagnosing depression. There are many limitations in laboratory test as accuracy is not achieved up to the mark. A cost effective and non-invasive for detecting depression is required for accurate diagnosis. Our study aims to use EEG signals for depression detection using a machine learning approach. This paper aims to classify the depression and healthy subjects by using EEG signal. This includes a machine learning based depression diagnosis using EEG frequency bands in the form of linear features contributing for early detection. In this research EEG frequency bands, the linear features are used for classification of depressed & non depressed subject.