Depression Episodes Detection in Unipolar and Bipolar Patients: A Methodology with Feature Extraction and Feature Selection with Genetic Algorithms Using Activity Motion Signal as Information Source

Table 5

Comparison with related works on identification of depression.

Work	Description
Averill et al. [24]	Examining the response of the depression treatment based on the simple activity and psychomotor speed measured through actigraphy identifying depressed episodes.
Garcia-Ceja et al. [15]	Analyzing data collected through actigraphy comparing different machine learning techniques to classify depressed subjects.
Gershon et al. [43]	Identifying activity patterns from locomotor activity collected by actigraphy extracting a series of principal components, discriminating depressive days from other states.
Koo et al. [23]	Identification of depressed patients through the combination of biomarkers related to executive dysfunctions, motor activity, and neurophysiological patterns activity, among others.