A multimodal approach to depression diagnosis: insights from machine learning algorithm development in primary care

European Archives of Psychiatry and Clinical Neuroscience, Год журнала: 2025, Номер unknown

Опубликована: Март 10, 2025

Abstract General practitioners play an essential role in identifying depression and are often the first point of contact for patients. Current diagnostic tools, such as Patient Health Questionnaire-9, provide initial screening but might lead to false positives. To address this, we developed a two-step machine learning model called Clinical 15 , trained on cohort 581 participants using nested cross-validation framework. The integrates self-reported data from validated questionnaires within study sample patients presenting general practitioners. demonstrated balanced accuracy 88.2% incorporates traffic light system: green healthy, red depression, yellow uncertain cases. Gaussian mixture clustering identified four subtypes, including Immuno-Metabolic cluster characterized by obesity, low-grade inflammation, autonomic nervous system dysregulation, reduced physical activity. algorithm all immuno-metabolic depressed, although 22.2% (30.8% across whole dataset) were categorized uncertain, leading light. biological characterization monitoring their clinical course may be used differential risk stratification future. In conclusion, provides highly sensitive specific tool support GPs diagnosing depression. Future improvements integrate further markers longitudinal data. tool’s utility needs evaluation through randomized controlled trial, which is currently being planned. Additionally, assessing whether actively algorithm’s predictions into treatment decisions will critical its practical adoption.

Язык: Английский

Interaction of chronic diseases and levels of mastery on the course of depression

Journal of Psychosomatic Research, Год журнала: 2024, Номер 189, С. 112000 - 112000

Опубликована: Дек. 7, 2024

Язык: Английский