App based on intangible cultural heritage to prevent memory loss in the elderly
The UPC’s Soft Computing research group is leading the LONG-REMI project for designing an app for reminiscence therapy. The project is being developed jointly by researchers from the UPC, the Universitat Autònoma de Barcelona and the Instituto Politécnico de Coimbra (Portugal).
Sep 16, 2021
Intangible cultural heritage - such as oral literature, music and dance - contributes to the development of identity as part of our collective imagination. They are deeply rooted elements of our cultural identity that can have a very positive impact in the context of reminiscence therapy to prevent cognitive impairment.
The LONG-REMI project aims to develop an AI-based app that allows for periodical reminiscence therapy that is tailored to patients’ needs, based on shared heritage. This is a simple and intuitive tablet app that displays photos, videos or questions related to proverbs, tongue-twisters and songs, which act as stimuli to trigger memories.
The app, which will be available at the end of the year, aims to prevent cognitive impairment and avoid the effects of severe memory loss, and also to improve patients’ mood and overall health. The app can be used in patients with or without cognitive impairment and will allow to conduct reminiscence therapy in a more agile, comfortable and economical way than conventional treatments, which require monitoring by specialised staff and thus result in more expensive therapies that are occasionally received only by patients with dementia.
The project is led by the researcher Angela Nebot, who leads the Soft Computing research group (SOCO) at the Intelligent Data Science and Artificial Intelligence Research Center (IDEAI) of the Universitat Politècnica de Catalunya · BarcelonaTech (UPC), and involves the researcher Francisco Múgica, also from SOCO, and researchers from the Foundation for Health and Ageing of the Universitat Autònoma de Barcelona (UAB) and the Instituto Politécnico de Coimbra, Portugal.
Artificial intelligence for emotion recognition
The UPC researchers are working to develop the algorithm that will guide reminiscence therapy and adapt it to each user’s specific needs based on cognitive impairment. AI-based facial recognition techniques that use computer vision and machine learning will assess dynamically patients’ emotions while they are engaged in the activities, especially by analysing the shape and position of their eyes and mouth.
As Nebot explains, “the patient’s basic emotions registered - happiness, sadness, indifference and anger - provide highly valuable information and will constantly guide the algorithm in searching for activities that best trigger memories in real time. The algorithm will leverage the elements that are significant to the patient and that evoke positive memories and will provide related information to achieve the best results from the therapy session”, adds the researcher, who is also a professor at the Barcelona School of Informatics (FIB).
“Cognitive stimulation therapies are of growing interest and are recommended in people with cognitive impairment and early-stage dementia to maintain cognitive skills, functional abilities and quality of life. It becomes crucial to develop new cognitive stimulation tools in clinical practice for older people with cognitive impairment, based on new technologies”, adds Antoni Salvà, the director of the UAB’s Foundation for Health and Ageing.
For Sara Domenech, the head of research in the cognitive area at the UAB’s Foundation for Health and Ageing, “using intangible cultural heritage as the topic for reminiscence therapy gives participants a sense of belonging and identity; additionally, therapy becomes pleasant and tailored to their tastes and preferences.”
The app will be tested by 60 older people from Spain and Portugal in October. In Barcelona, the trials will be conducted at the Baix Guinardó Home for the Elderly, the Nou Horitzó Association and the Pere Virgili Health Park day hospital. Ten people without cognitive impairment, ten with mild cognitive impairment and ten with mild dementia will take part. The trials will consist of three individual 45-minute sessions to assess participants’ response to the stimuli presented by the app.
The LONG-REMI project is funded, within the framework of the PSL-POCTEP programme (Programme for a Long-Lived Society), by the CSIC foundation and the European Union’s ERDF.
The LONG-REMI project aims to develop an AI-based app that allows for periodical reminiscence therapy that is tailored to patients’ needs, based on shared heritage. This is a simple and intuitive tablet app that displays photos, videos or questions related to proverbs, tongue-twisters and songs, which act as stimuli to trigger memories.
The app, which will be available at the end of the year, aims to prevent cognitive impairment and avoid the effects of severe memory loss, and also to improve patients’ mood and overall health. The app can be used in patients with or without cognitive impairment and will allow to conduct reminiscence therapy in a more agile, comfortable and economical way than conventional treatments, which require monitoring by specialised staff and thus result in more expensive therapies that are occasionally received only by patients with dementia.
The project is led by the researcher Angela Nebot, who leads the Soft Computing research group (SOCO) at the Intelligent Data Science and Artificial Intelligence Research Center (IDEAI) of the Universitat Politècnica de Catalunya · BarcelonaTech (UPC), and involves the researcher Francisco Múgica, also from SOCO, and researchers from the Foundation for Health and Ageing of the Universitat Autònoma de Barcelona (UAB) and the Instituto Politécnico de Coimbra, Portugal.
Artificial intelligence for emotion recognition
The UPC researchers are working to develop the algorithm that will guide reminiscence therapy and adapt it to each user’s specific needs based on cognitive impairment. AI-based facial recognition techniques that use computer vision and machine learning will assess dynamically patients’ emotions while they are engaged in the activities, especially by analysing the shape and position of their eyes and mouth.
As Nebot explains, “the patient’s basic emotions registered - happiness, sadness, indifference and anger - provide highly valuable information and will constantly guide the algorithm in searching for activities that best trigger memories in real time. The algorithm will leverage the elements that are significant to the patient and that evoke positive memories and will provide related information to achieve the best results from the therapy session”, adds the researcher, who is also a professor at the Barcelona School of Informatics (FIB).
“Cognitive stimulation therapies are of growing interest and are recommended in people with cognitive impairment and early-stage dementia to maintain cognitive skills, functional abilities and quality of life. It becomes crucial to develop new cognitive stimulation tools in clinical practice for older people with cognitive impairment, based on new technologies”, adds Antoni Salvà, the director of the UAB’s Foundation for Health and Ageing.
For Sara Domenech, the head of research in the cognitive area at the UAB’s Foundation for Health and Ageing, “using intangible cultural heritage as the topic for reminiscence therapy gives participants a sense of belonging and identity; additionally, therapy becomes pleasant and tailored to their tastes and preferences.”
The app will be tested by 60 older people from Spain and Portugal in October. In Barcelona, the trials will be conducted at the Baix Guinardó Home for the Elderly, the Nou Horitzó Association and the Pere Virgili Health Park day hospital. Ten people without cognitive impairment, ten with mild cognitive impairment and ten with mild dementia will take part. The trials will consist of three individual 45-minute sessions to assess participants’ response to the stimuli presented by the app.
The LONG-REMI project is funded, within the framework of the PSL-POCTEP programme (Programme for a Long-Lived Society), by the CSIC foundation and the European Union’s ERDF.