Charting Brain Dynamics using Normative Modeling

This research endeavor aims to delineate the normal range of cerebral functional activities by employing advanced federated machine learning algorithms on a large and geographically distributed dataset of Magneto/encephalography brain recordings. The resultant models will undergo validation against clinical datasets to facilitate the identification of biological biomarkers indicative of complex psychiatric disorders.

Globally, mental disorders are a leading cause of disability and ill health. According to the World Health Organization (WHO), approximately 1 in 8 people will be affected by mental or neurological disorders at some point in their lives. Depression is the most common mental disorder globally, affecting more than 280 million people. In the Netherlands, mental disorders are also a significant health issue. According to findings from the second Netherlands Mental Health Survey and Incidence Study (NEMESIS II), around 43% of the Dutch population will experience a mental disorder.

Despite societal urges, the diagnosis and treatment of mental disorders remain a big challenge due to their complex nature and the substantial heterogeneity in their symptoms. Mental disorders can present with a wide range of symptoms that can vary in severity, frequency, and duration, even among individuals with the same diagnosis. This can make it difficult for healthcare professionals to accurately diagnose mental disorders and determine the most effective treatment options for each individual. Normative modeling is a statistical method that can help decode the heterogeneity of symptoms in mental disorders. Using normative modeling, healthcare professionals can identify patterns and subtypes of symptoms that may be more closely associated with certain diagnoses. Such subtypes can also be useful for predicting treatment outcomes and developing personalized treatment plans.