Multimodal Detection of Agitation in People With Dementia in Clinical Settings: Observational Pilot Study
Bibliographic record
Abstract
Background: Dementia is a progressive neurodegenerative condition that affects millions worldwide, often accompanied by agitation and aggression (AA), which contribute to patient distress and increased health care burden. Existing assessment methods for AA rely heavily on caregiver reporting, introducing subjectivity and inconsistency. Objective: This study proposes a novel, multimodal system for predicting AA episodes in individuals with severe dementia, integrating wearable sensor data and privacy-preserving video analytics. Methods: A pilot study involving 10 participants was conducted at Ontario Shores Mental Health Institute. The system combines digital biomarkers collected from the EmbracePlus (Empatica Inc) wristband with video-based behavioral monitoring. Facial features in video frames were anonymized using a masking tool, and a deep learning model was used for AA detection. To determine optimal performance, various machine learning and deep learning models were evaluated for both wearable and video data streams. Results: The Extra Trees model achieved up to 99% accuracy for personalized wristband data, while the multilayer perceptron model performed best in general models with 98% accuracy. For video analysis, the gated recurrent unit model achieved 95% accuracy and 99% area under the curve, and the long short-term memory model demonstrated superior response time for real-time use. Importantly, the system predicted AA episodes at least 6 minutes in advance in all participants based on wearable data. Conclusions: The findings demonstrate the system's potential to autonomously and accurately detect and predict AA events in real-time. This approach represents a significant advancement in the proactive management of behavioral symptoms in dementia care.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".