As part of the PD Assist+ initiative, which is funded by TRiNNOVATION OWL, inIT is developing an AI-powered voice assistant to support Parkinson's care. The aim is to improve communication between patients and healthcare professionals by facilitating the use of natural language, digital document management, and specialised information.
Digital assistance for a complex disease
Caring for people with Parkinson's disease is challenging as it requires continuous support from interdisciplinary teams. Excellent care requires collaboration between clinics, therapy facilities, general practitioners, pharmacies and others. The PD Assist+ project provides new impetus for the technology-supported optimisation of care.
The project aims to improve communication between all parties involved by providing an intuitive, voice-based interface and making medical information more accessible.
Language models as an interface to care
Under the direction of Prof. Dr. Volker Lohweg, the inIT working group 'Discrete Systems' is set to further develop the existing PD Assist app for Parkinson's patients. Christoph-Alexander Holst (research group leader) and Patrick Gaudl (research associate) will lead the development.
The language assistance is based on a specialised large language model (LLM) that has been trained using Parkinson's-specific data. Combined with a retrieval-augmented generation (RAG) architecture, it will be embedded in the PD Assist app. This will enable users to access content from the digital patient archive, as well as general information about the disease.
Testing in everyday clinical practice
The new functionality is being tested in cooperation with the University Clinic for Neurology and Neurogeriatrics at the Lippe Clinic and the Parkinson's Network OWL+. The aim is to investigate whether the application is perceived as providing relief in everyday life and whether it can address key care challenges.
Funding from TRiNNOVATION OWL enables the project team to further develop the application in close cooperation with practice partners and to evaluate it in a targeted manner.
Prospects for regional healthcare
‘With PD Assist+, we aim to demonstrate the tangible benefits of AI-based assistance systems in improving patient care,’ explains Christoph-Alexander Holst. ‘Close cooperation with Parkinsonnetz OWL+ is particularly important to us, as it allows us to incorporate practical needs into the development process.’
The further development of the app will not only strengthen regional Parkinson's care, but also emphasise inIT's role as an innovation partner in healthcare.
TRiNNOVATION OWL is a project of the Federal-State Initiative Innovative University, which is funded by the Federal Ministry of Education, Research, Technology and Space (BMBF).