The role of cognitive functions in the diagnosis of bipolar disorder: A machine learning model

Background: Considering the clinical heterogeneity of the bipolar disorder, difficulties are encountered in making the correct diagnosis. Although a number of common findings have been found in studies on the neurocognitive profile of bipolar disorder, the search for a neurocognitive endophenotype has failed. The aim of this study is to separate bipolar disorder patients from healthy controls with higher accuracy by using a broader neurocognitive evaluation and a novel machine-learning algorithm. Methods: Individuals who presented to the Bipolar Outpatient Clinic of the Medical Faculty of Eskisehir Osmangazi University and met the inclusion criteria of the research are included in the study. Six neurocognitive tests from the CANTAB test battery were used for neurocognitive evaluation, Polyhedral Conic Functions algorithm was used to classify the participants. Results: Bipolar disorder patients differentiated from healthy controls with an accuracy of 78 %. Discussion: Our study presents a prediction algorithm that separates bipolar disorder from healthy controls with high accuracy by using CANTAB neurocognitive battery.

Automated summary

This study achieved high accuracy in differentiating bipolar disorder patients from healthy controls by utilizing a broader neurocognitive evaluation and a novel machine-learning algorithm.