Help elderly people and those with neurodegenerative diseases like Alzheimer's preserve their memories and deepen their connections with family and caregivers. That's the mission behind the IBM Italy Foundation's new pilot project with the IRCCS Casa Sollievo della Sofferenza in San Giovanni Rotondo (Foggia): ViTA, Virtual Training for Aging.
Using various technologies and Artificial Intelligence, the ViTA system collects memory fragments—photographs, texts, songs, videos tied to lived experiences—and organizes them into a map where each piece connects to the others, creating personalized memory pathways and stories built from events the patient recalls. It's a form of cognitive training that strengthens relationships by working with memory and improves the elderly person's well-being, countering sadness or agitation through technology.
We spoke with Pietro Leo, CTO for Artificial Intelligence & Chief Scientist of IBM Italy, and Francesco Giuliani, Director of IT Systems, Innovation and Research at IRCCS Casa Sollievo della Sofferenza in San Giovanni Rotondo.
How did the ViTA project begin?
IBM Italy Foundation approached Casa Sollievo della Sofferenza hospital with a proposal to explore how technology and digital tools could bring generations together around the theme of memory. Our system aims to strengthen that connection. The project focuses on "Aging"—an increasingly urgent topic—and addresses the challenge our society faces in serving a rapidly aging population. We decided to focus primarily on people showing the first signs of cognitive decline, which manifests precisely as a weakening ability to retain memories.
What technologies does it use, and how does it work?
ViTA brings together various technologies with one goal: to help collect and connect memories. Technically, it uses IBM's Watson Cognitive Computing platform with its Natural Language Processing capabilities. We also use open-source tools like a Knowledge Graph system to store and represent the memory map, along with other AI tools. From the user's perspective, it's an app on a tablet or other device that offers multiple ways to interact—voice, text, touch—to navigate the memory map and access stories stored in the memory graph.
How do you build a memory map together with Alzheimer's patients and those around them?
The memory map, which can include distant memories, recent ones, or reminders for the future, is built through interaction between the patient and their caregiver—typically a family member or assistant. The caregiver uses a system that lets them define the "atomic elements" of memory, enriching them with specific content like a photo or a song's audio and descriptive text. These atomic elements link together to form stories. Stories are pathways the caregiver constructs based on events the patient recalls. They can reinforce or strengthen a particular emotional state triggered by those specific memories. One possible evolution is expanding the initial "memory bank" to include elements the system captures through natural dialogue with the patient.
Do the effects last?
We currently have an initial trial underway to evaluate the system in two settings: home environments and care facilities. The first phase focuses on assessing acceptability and functionality, with the goal of scaling up in the next phase. Only then will we have evidence about the system's long-term effects.
What impact does it have on the patient's mood?
The patient's response to the system is central to our research hypothesis. We want to verify whether ViTA—with its organized memories presented as stories—can reduce agitation, for example. This builds on similar experiences tested on a larger scale in the United States and England, which have already shown some promise for this approach.
What effects are you seeing in people tested at Casa Sollievo della Sofferenza?
We can't answer that yet. We're currently collecting data and populating the memory database for participants who have consented to the trial. It's demanding work, done with the help of people with psychological expertise and caregiver support. To give you an idea: if we want to add a meaningful photo, it needs to be accompanied by essential information about where it was taken, how it connects to other memories, who was there, when it happened, what emotions it stirs, and so on. We expect to produce our first results by next autumn.
What is the social impact of the project, and where do you hope it leads?
With ViTA, technology aims to strengthen relationships between people. We want to make it easier for those being cared for and their caregivers to share knowledge and memories, fostering a genuine exchange. The moment you create the memory map is just as important as using it afterward. The social impact is significant when we think broadly—not just about the individual but about the community they live in. Systems like ViTA will work much like the web itself: not only preserving memory fragments (like the web pages we visit), but also making those fragments easier to share across generations. This supports inclusion and the transfer of experience between age groups. There's a useful English term: "ageism"—discrimination based on age. ViTA directly counters this by fostering connection across age groups and reducing that prejudice.
Source: StartUpItalia