A TRUSTworthy speech-based AI monitorING system for the prediction of relapse in individuals with schizophrenia

Información del proyecto

TRUSTING

Identificador del acuerdo de subvención: 101080251

DOI

10.3030/101080251

Fecha de la firma de la CE 5 Abril 2023

Fecha de inicio 1 Julio 2023

Fecha de finalización 31 Diciembre 2028

Financiado con arreglo a

• Health

Coste total

€ 5 999 510,00

Aportación de la UE

€ 5 999 510,00

Coordinado por

ACADEMISCH ZIEKENHUIS GRONINGEN
Netherlands

Periodic Reporting for period 1 - TRUSTING (A TRUSTworthy speech-based AI monitorING system for the prediction of relapse in individuals with schizophrenia)

Período documentado: 2023-07-01 hasta 2024-12-31

Resumen del contexto y de los objetivos generales del proyecto

Psychotic disorders tend to have a waxing and waning course, with vulnerable persons being at risk for relapse. Psychotic relapses can often be prevented when a warning is received in time, so that appropriate measures can be taken. The EU-funded TRUSTING project is working under the hypothesis that spontaneous speech holds the key to predicting whether psychosis is imminent. The approach used will involve asking people at risk of psychotic relapse to record short speech samples at regular intervals. These samples will be analysed using artificial intelligence, with the aim of recognising changes that signal subtle psychotic symptoms. Our goal is to develop a user-friendly, trustworthy application that can be used from home to deliver a message when speech deviations predictive of psychotic relapse are detected. More information about the project can be found on the project website: https://trusting-project.eu

Trabajo realizado desde el comienzo del proyecto hasta el final del período abarcado por el informe y los principales resultados hasta la fecha

Over the past 18 months, significant progress has been made in laying the scientific foundation for the TRUSTING project and developing the AI monitoring tool. A beta version of the application to be used for the monitoring tool has been created and tested by members of the TRUSTING service user board, consisting of both patients and clinicians. Their feedback is being incorporated to ensure the tool is both user-friendly and trustworthy. Further information to that end has been obtained by a completed user assessment survey among patients and clinicians, meant to inform the design of the tool.
Extensive research has been conducted to identify and validate speech features linked to relapse in psychosis across multiple languages and speech elicitation tasks. This work provides the foundation for the monitoring tool.
Additionally, the direct and indirect costs of schizophrenia have been investigated, as well as reimbursement criteria in public financing systems. Significant progress has been made in developing the value framework for the TRUSTING tool. These are important first steps in understanding the economic impact of the TRUSTING tool.
Simultaneously, preparations are well underway for the upcoming trial, which will test the tool in a large, multilingual population. The trial is scheduled to begin in July 2025.

Avances que van más allá del estado de la técnica e impacto potencial esperado (incluida la repercusión socioeconómica y las implicaciones sociales más amplias del proyecto hasta la fecha)

The results achieved so far in the project provide a crucial foundation for the development of the AI monitoring tool. These results are expected to contribute to the creation of innovative technology that addresses a currently unmet need: delivering equitable and just healthcare for individuals at risk of relapse. While the project is still in its early stages, key next steps include demonstrating the tool’s efficacy in real-world settings, assessing its health economic impact, and establishing a clear roadmap for commercialization.
Ongoing efforts will focus on refining the technology and addressing regulatory compliance, which are essential for translating the project’s results into meaningful societal and commercial impact.