Virtual Reality Companion for Dementia Patients in Long‐term Care: A Feasibility Study
Bibliographic record
Abstract
Abstract Background Persons with dementia (PWDs) develop progressive cognitive and physical decline which often results in placement in long‐term care (LTC). Within LTC, PWDs are at high risk for social isolation, as friends and family members may not visit often, a problem exacerbated by the COVID‐19 pandemic. Method This project is developing a novel application for immersive virtual reality (VR), in which an avatar (i.e. a visual representation of a person) acts as an autonomous artificial companion for PWDs in LTC. This avatar is programmed to listen to PWDs through microphones embedded in VR goggles, and provide verbal interaction with PWDs through embedded speakers. In this way, the avatar can provide a source of conversation and socialization, available at any time, for any duration. This engagement may improve quality of life, and reduce responsive and reactive dementia behaviours. Result Our virtual companion was trialed in 10 PWD (3 men; 7 women). Most PWD found the companion to be engaging, and this invited positive reminiscence in 3 PWD. VR goggles were generally well‐tolerated, though 2 PWD closed their eyes and/or did not respond. The ability to interact verbally in a conversational manner was easier in patients with milder dementia. There was some poor performance of the speech recognition software, which impacted the flow of the conversation. Significant verbal feedback was recorded on how to improve the avatar, and how to enhance the conversation. Conclusion An autonomous artificial companion presented in virtual reality (VR) is feasible, and may benefit persons with dementia who are at risk of social isolation. Future iterations will incorporate improved speech recognition and artificial intelligence (AI)‐guided conversation generator software.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".