Depression impacts many aspects of an individual, including growth, development, and family relationships. Worldwide prevalence rates of depression are between 2.2 to 10.4%. If those who suffer from depression are not diagnosed and treated in time, severe consequences can easily be caused and even lead to suicide. However, the biological causes of depression are largely unknown. Therefore, clinical diagnosis of depression is currently based on symptoms, equivalent to fever, rather than its underlying causes, which is equivalent to the measles virus.
The EEG signal contains large amounts of data about many areas of the brain and cannot be easily manipulated by the subject. Therefore, it offers a great potential for unbiased depression recognition.
In my talk we propose to use computer technology, in particular machine learning methods, to design several depressivity prediction models to predict depressivity and find potential biomarkers of depressivity disorder in EEG signals. Ultimately, these biomarkers could redefine our ideas of what depression actually is.
Last modified: Tuesday, 05-Mar-2019 09:08:01 NZDT
This page is maintained by the seminar list administrator.