ki:elements

Benefits of Automated Speech Analysis in Psychiatry

Felix Menne, Lana Kambeitz-Ilankovic, Diana Immel & Julia Schräder

* Symposium organized for the DGPPN Congress 2024, Berlin (Germany)

Symposium Abstract

Compared to other areas of medicine, psychiatry lacks objective biomarkers that are not primarily based on subjective assessments. Novel precision medicine approaches could help with challenges such as predicting response to lithium or resistance to antidepressants. Among the growing number of digital biomarkers for the objective assessment of mental disorders, automated speech analysis offers great potential, as psychiatric symptoms are also reflected in linguistic changes. Advances in computational linguistics enable the use of automatic speech analysis as an objective method for measuring psychiatric symptoms.

Dr. Felix Menne (ki:elements GmbH) will give an overview of the latest developments in speech analysis in the psychiatric field. In addition, data from own projects that were developed together with clinical partners will be presented.

Dr. Lana Kambeitz-Ilankovic (University Hospital Cologne) will present the EU-funded project PRONIA (Personalised Prognostic Tools for Early Psychosis Management), in which new methods for the early detection of psychosis are being researched, including parameters for automated speech analysis.

Diana Immel, M.Sc. (University of Oldenburg) will present UBIDENZ (Ubiquitous Digital Empathic Therapy Assistance), a joint project funded by the BMBF in which the University of Oldenburg is researching an assistance system for people with depression for care after hospital discharge. Among other things, linguistic parameters were collected in order to draw conclusions about changes in symptoms.

Julia Schräder, M.Sc. (RWTH Aachen) will report on the EmoCon study funded by the DFG, in which the influence of unconscious emotional stimuli on the evaluation of emotional facial expressions and their connection with language parameters was investigated in depressed patients and healthy control subjects.

Share this article