Psychobiological markers to improve diagnosis and to predict affective episodes in bipolar disorder

There is an urgent need for better tools to improve and facilitate diagnosis of bipolar disorder and to detect and prevent affective episodes at the earliest possible stage.

We hypothesize that the longitudinal interaction between three psychobiological factors (sleep patterns, motor activity, and decision-making abilities) contains information to distinguish patients with bipolar disorder and to detect upcoming affective episodes. In this project, we will first optimize an existing decision-making battery for longitudinal testing.

Thereafter, we will use a smartphone app, a smartwatch, and artificial intelligence algorithms to characterize the interaction between the psychobiological factors at the individual level for 12 weeks in healthy volunteers, patients with bipolar disorder, and patients with recurrent depressive episodes.

We will demonstrate that bipolar patients can be distinguished based on the longitudinal interaction between the psychobiological factors.

Finally, we will follow up the patients with bipolar disorder for 1-year and use the artificial intelligence algorithm trained on each patient´s data to predict upcoming changes in affective states at the individual level.

This project will allow us to develop an app that can be used in the clinic to improve diagnosis and to predict affective episodes in bipolar disorder.

As such, this project will move the field of psychiatry towards precision medicine and improve the care of patients with bipolar disorder.