Identification of Predictive Biomarkers of Mood Relapses in Patients With Bipolar Disorder (Predibip)

Relapses in bipolar disorders are associated with a significant proportional functional impact, as well as worsening of the course of bipolar disorder, with impairment of the quality of functional remission, as well as the development of addictive, anxiety and suicidal comorbidities.The functional deficit and the instability of the mood disorder increase with thymic relapses. Currently, these relapses (transition from the state of remission, to a depressive or hyperthymic state) are difficult to predict and to treat because of the absence of correlation between the degree of severity of the stressful event (intensity associated stress) and the occurrence of relapse, taking into account the mediation of this relationship by the stress compensation / adaptation capacities, which are very individual. This project proposes to develop tools based on artificial intelligence technologies to monitor the level of stress and adaptation to life events as well as identifying relapse predictive factors of a patient by using portable and connected devices recording different physiological signals in order to alert him/her when there is a risk of relapse, thus anticipating therapeutic strategies.

clinical assessment via psychometric scales physiological data acquired automatically via the connected device (wristwatch, wristband)

number of relapses predicted by machine learning algorithms versus number of relapses highlighted by clinical assessment

Correlation between values of Heart rate variability and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between values of skin surface temperature and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between values of actimetry and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between ElectroDermal activity and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between sleep periods and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between social measures and the occurrence of a thymic relapse (number of relapses during the follow-up period).

The internet connection times (minute), including social networks, the number and duration (minute) of telephone calls and the number and length (number of characters) of text messages will be automatically recorded daily in order to provide parameters that help quantify the patient's social activity.

Correlation between values of clinical characteristics (age of onset) and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between values of clinical characteristics about thyme episode and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Predominant polarity (hypo/manic or depressive or without), polarity of last episode (hypo/manic or depressive), number of previous thymic episodes, duration of remission (in month).

Correlation between values of clinical characteristics (suicide attempts) and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between values of clinical characteristics (hospitalizations) and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between values of clinical characteristics (comorbid disorders) and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Correlation between values of clinical characteristics (medication) and the occurrence of a thymic relapse (number of relapses during the follow-up period).

Inclusion Criteria: Voluntary subjects responding to the diagnosis of bipolar I or II disorder according to the criteria of the DSM-5 Agreement to benefit from connected objects required for study and able to understand their use Age between 18 and75 years old Affiliated to the Social Security system (beneficiary or entitled) No opposition signed Exclusion Criteria: Subject included in clinical and/or therapeutic experimentation with exclusion period Primary psychiatric pathology other than bipolar type I or II disorder Patient hospitalized without consent Subject deprived of liberty by judicial or administrative decision Pregnant, parturient, breastfeeding woman Known allergy to one of the materials of the bracelets