PSYCOP


PSYCOP: The PSYchiatric Clinical Outcome Prediction cohort 

The PSYCOP Group is a research collaboration between Aarhus University Hospital / Central Denmark Region and Aarhus University, dedicated to advancing the field of precision psychiatry. People with mental illness experience a significantly reduced quality of life and lifespan, and current prediction studies rarely translate into clinical practice. The PSYCOP project addresses this gap by developing clinically applicable machine learning tools to improve treatment and patient outcomes.

At the heart of the project is the PSYchiatric Clinical Outcome Prediction (PSYCOP) cohort—a large retrospective dataset including all patients with at least one contact with psychiatric services in the Central Denmark Region between 2011 and 2020 (n=119,291). Drawing on comprehensive electronic health records (EHR)—covering diagnoses, treatments, clinical notes, discharge summaries, laboratory tests, and more—the project builds prediction models for a broad range of outcomes, such as diagnostic shifts, treatment response, medical comorbidity, and premature mortality. A key focus is on evaluation, feasibility, and implementation in real-world clinical psychiatry.

PSYCOP combines clinical expertise and cutting-edge data science. The group applies advanced machine learning and natural language processing methods to transform rich health record data into clinically meaningful predictions. By grounding the research in practical applicability, PSYCOP aims not only to advance psychiatric research but also to improve treatment planning, prognosis, and ultimately the quality of life and lifespan of individuals with mental illness.

At Aarhus University, the group is embedded in the Center for Humanities Computing (CHC) research environment. PSYCOP-PhD students are co-supervised by Professor Søren Dinesen Østergaard (Department of Clinical Medicine, Department of Affective Disorders) and Professor Kristoffer Nielbo (CHC), benefiting from a strong interdisciplinary environment with expertise in machine learning, NLP, and data modeling.


Project affiliation


Funding

Lundbeck Foundation, Novo Nordisk Foundation, the Danish Cancer Society, the Central Denmark Region Fund for Strengthening of Health Science and the Danish Agency for Digitisation Investment Fund for New Technologies.  


Services and Support

Contact us by submitting a ticket with the CHC frontoffice