VVT en de kansen van AI

VVT en de kansen van AI

About

VVT en de kansen van AI

The challenges in elderly care are well known: an ageing population will double the demand for care by 2040, staff shortages are growing, healthcare costs are increasing and treatments grow more complex. So the question is: can data-driven technologies help us stay healthy for longer, detect illness earlier, stay at home longer, enable us to receive care remotely and to make sure care workers can be deployed as efficiently as possible on tasks where human input is indispensable?

VVT is a Dutch abbreviation for the nursing, care and home assistance sector. Innovations in the sector have lead to the use of sensors for a broad range of alerts. These sensors produce data and the data enables us to explore how to further innovate, using AI. By using data-driven technologies, we aim to find new solutions for the challenges the healthcare system is facing.

Objectives
In this research project, we have teamed up with care professionals, data scientists, legal experts and tech companies to combine our knowledge and expertise. The project focuses on two pillars; the first is developing use cases such as fall prevention, where an AI model is being developed using elderly care data to predict when clients may be more susceptible to falling. If using AI can prevent that client from falling in the first place, that is a win for both the client and the caregiver; the client doesn’t get hurt, the caregiver does not need to spend time on the injuries caused by the fall. Another use case focuses on providing support and improving advice to the care dispatcher on the validity and urgency of sensor alerts.

The second and overarching objective is to develop a practical toolkit for the organization of structural data supply and research. The toolkit aims to facilitate pathways for future consortia and to encourage continuous innovation in elderly care – thus increasing the impact for the sector as a whole.

Challenges
Care facilities were initially reluctant to share data, their main concern understandably being any privacy issues that could occur for their clients. Our first goal was to build trust that we can sufficiently anonymize this data, in which we succeeded. Subsequently we were able to show care facilities what we can achieve using the data, which sparked their curiosity.

We have ensured that exported data sharing takes place in a secure and responsible manner, complying with all relevant legal frameworks. In the next phase we will work on drawing up a system architecture that allows real-time data to be shared with the same security requirements.

VVT en de kansen van AI will be ongoing until September 2025.