Vicomtech participates in the AGAR project, which aims to improve the remote assessment and monitoring of elderly people through conversational assistants.

There are many elderly people who, in a situation of dependency, decide to remain at home. However, in their situation, very specific and careful monitoring is required to facilitate and ensure a good quality of life.

For this reason, the AGAR project proposes to use a secure conversational assistant for the remote assessment and monitoring of dependent elderly people who decide to stay at home. The aim is that a secure conversational assistant incorporated into the home can be used to carry out personalised assessments and monitoring based on the profile of the dependent elderly person. It also incorporates a decision support component for professionals based on the profile of the person and the action protocols.

Aim of the project

The aim of the AGAR project is to advance in the development of technology to facilitate the assessment and personalised monitoring of elderly people in a situation of dependency at home. This will be carried out through:

• The implementation of a secure and personalised conversational assistant for the assessment of elderly people.
• The development of a decision support system for the professional monitoring team based on the care protocols according to the profiles of the people.

Technologies developed and applied during the project

The following technologies have been used throughout the project:

1) Automatic speech detection and wake-up-word.

A speech recognition model adapted to the AGAR use case has been developed to facilitate voice interaction. The assistant also has a wake-up word or wake-up phrase to identify whether the user is addressing the system or another interlocutor. The recognition model will only start working if the system detects that the user has said the wake-up word.

2) Personalised interaction

A number of skills or capabilities have been developed that allow expert staff to assign personalised information to the profile of each of the residents in their care. Thanks to these capabilities, each user can consult personalised information and, at the same time, the privacy of all users is maintained.

3) Reminders

The assistant has been given the ability to start the conversation without the need for the user to initiate an interaction first. The aim of this capability is to be able to schedule personalised reminders for each user, such as taking medication or visiting the doctor.

4) Decision support platform

A platform has been developed to help the professional team of the residence to monitor the residents and make decisions based on the profile of each one of them and the established action protocols.

Vicomtech's role

Vicomtech's role has focused on designing, developing, integrating and testing both the conversational assistant and the decision support platform. The project consortium consists of ALCAD, Caser Residencial and Vicomtech.

Thanks to the AGAR project, elderly people in a situation of dependency who are at home will be able to receive the same monitoring and assessment as those who are physically in the residence. Advances like this demonstrate that technology applied to health in geriatrics will be able to facilitate and improve the quality of life of all these people.

The AGAR project has been funded by the Basque Government within the framework of the Hazitek programme.