More information
Description
Mental state in clinical research is assessed subjectively through discussion and rating scales. Initial research confirms that objectively collected data from sensors can be a gamechanger in detection of episodes of bipolar disorder (BD). However, the key barriers to use sensors in BD monitoring remain open: (i) lack of easily adaptable computational methods for BD episodes prediction; (ii) lack of reliable benchmark datasets for training of the algorithms. BIPOLAR has access to two large digital anonymised data sets already collected from sensors of BD patients which will guide the research and experimental development. BIPOLAR aims at the development of highly novel computational intelligence methods for sensor-based, semi-supervised and uncertainty-aware prediction of BD episodes.
Scientific uniqueness of BIPOLAR consists of delivering a software prototype with a set of accurate computational methods for early prediction of bipolar episodes based on the conjoint use of fuzzy logic, statistical process control and semi-supervised learning. Results of BIPOLAR will be made available under open access licence and will have significant societal and economic impact.
The target communication group ranges from an expert-audience to the main public. An important target group of BIPOLAR are psychiatrists and small and medium-sized enterprises (SMEs) who may particularly benefit from the prototypes developed in BIPOLAR. Intended long-term objective of BIPOLAR would be the introduction in the clinical psychiatric practice objective quantification of important behavioral manifestations of depressive, mixed, and bipolar states, thus helping clinicians to objectively measure the degree of behavioral deviations from the norm in individual patients. Clinical application of the software that will grow from BIPOLAR package could lead to early identification of impending shift from depressive or euthymic states to manic or mixed states.
Summary of project results
Mental state in clinical research is assessed subjectively through discussion and rating scales. Initial research confirmed that objectively collected data from sensors can be a gamechanger in detection of episodes of bipolar disorder (BD). BIPOLAR („Bipolar disorder prediction with sensor-based semi-supervised learning”) project was needed to address the key barriers to use sensors in BD monitoring.
BIPOLAR developed highly novel computational intelligence methods for sensor-based, semi-supervised and uncertainty-aware prediction of BD episodes. Scientific uniqueness of BIPOLAR consisted of delivering a software prototype with a set of accurate computational methods for early prediction of bipolar episodes based on the conjoint use of fuzzy logic, statistical process control and semi-supervised learning.
The main exploitable result of the project are the open-source BIPOLAR packages published via GitHub and project''s webpage (http://bipolar.ibspan.waw.pl). All results of BIPOLAR were made available under open access licence, thus are expected to have significant societal and economic impacts, and to contribute significantly to a wider adoption of sensors in the patient''s natural setting and improve the diagnosis and monitoring of mental health. In particular, the outcomes of BIPOLAR will provide its end-beneficiaries (doctors and patients) options for early intervention on prodromal symptoms between outpatient visits in BD patients.
Intended long-term objective of the developed computational methods for mental state monitoring is the introducion in the clinical psychiatric practice objective quantification of important behavioral manifestations of depressive, mixed, and bipolar states, thus helping clinicians to objectively measure the degree of behavioral deviations from the norm in individual patients with subsequent option to monitor their dynamics (toward improvement or worsening) during treatment with antidepressants, antipsychotics, mood stabilizers, or various combinations among them at the individual-patient level. The potential social impacts are related with the suicide prevention as well as with the early diagnosis and prevention of both depressive and manic episodes when the patient is in euthymic (therapeutically normalized) state.