Innovating Works

AI-DIAGNOSE

Financiado
Machine learning for diagnosis of bipolar disorder detection of physiological d...
Machine learning for diagnosis of bipolar disorder detection of physiological digital biomarkers Bipolar disorder (BD) is a chronic and debilitating mental disorder, that affects 2-3% of the population. It impacts quality of life, cognition, and is a leading cause of suicide and all-cause mortalities. Most patients are taken... Bipolar disorder (BD) is a chronic and debilitating mental disorder, that affects 2-3% of the population. It impacts quality of life, cognition, and is a leading cause of suicide and all-cause mortalities. Most patients are taken into clinical care during acute episodes, which puts the burden on psychiatrists to make fast, yet accurate diagnostic decisions. However, unlike most medical conditions, psychiatric diagnoses are subjective. This paired with the complexity of its clinical presentation, BD is the most misdiagnosed and underdiagnosed psychiatric condition. More objective scales used in research lack clinical application, due to time constraints and high burden on the patient. AI-DIAGNOSE wants to disrupt the state of the art of BD diagnosis through a completely novel approach: developing an automatized and fast tool for objective detection of BD and psychotic symptoms based on physiological audiovisual biomarkers and machine learning (ML). The timing of the project is supported through recent evidence, from the host, the applicant, and others, showing that speech and eye movement are promising physiological biomarkers. In a pilot study, I found that ML algorithms based on speech patterns could predict the presence of psychiatric diagnosis, and differentiate patients with and without psychosis. Eye‐tracking datasets provide insights regarding information processing patterns, and have shown potential as diagnostic biomarkers. Although eye movement and speech patterns are promising biomarkers as they can be acquired fast and without putting high burden on the patient, they have not been combined yet for psychiatric diagnostic purposes. The project will be the first to develop such a multi-modal ML diagnostic tool for BD and psychosis in BD. We will test its accuracy against the research gold standard in the field within a large patient cohort (140 patients, 70 controls). If successful, this will a major step towards precision medicine within BD and psychiatry ver más
31/12/2025
UB
181K€
Duración del proyecto: 29 meses Fecha Inicio: 2023-07-18
Fecha Fin: 2025-12-31

Línea de financiación: concedida

El organismo HORIZON EUROPE notifico la concesión del proyecto el día 2023-07-18
Línea de financiación objetivo El proyecto se financió a través de la siguiente ayuda:
HORIZON-MSCA-2022-PF-01-01: MSCA Postdoctoral Fellowships 2022
Cerrada hace 2 años
Presupuesto El presupuesto total del proyecto asciende a 181K€
Líder del proyecto
UNIVERSIDAD DE BARCELONA No se ha especificado una descripción o un objeto social para esta compañía.
Total investigadores 328