Biomarkers and their use to detect the pathological profile of depression in mammals

摘要

To provide a system for detecting a depressive symptom of a subject animal.SOLUTION: A system comprises: a storage device for storing information for identifying a classifier from signals obtained by measuring in time series, using brain activity detection means, signals showing brain activity in a plurality of predetermined areas in a brain of a mammal at a resting state in a health state and depression state, the classifier being generated so that, from functional connectivity between the predetermined areas extracted using a sparse canonical correlation analysis related to a disease label for a depressive symptom using machine learning, the disease label for the depressive symptom is discriminated on the basis of a weighted sum of the functional connectivity by characteristic selection using sparse logistic regression, the selected functional connectivity including at least one selected from a first functional connectivity between a left dorsolateral prefrontal cortex and a left precuneus and left posterior cingulate cortex and a second functional connectivity between a left inferior frontal gyrus opercular part and a right dorsomedial prefrontal cortex and right supplementary motor area; and discrimination means for generating a classification result for a depressive symptom of a subject animal using the classifier.SELECTED DRAWING: Figure 5