Machine Learning to Detect Motor Agitation in People With Dementia Using Real-Time Location Data
Bibliographic record
Abstract
OBJECTIVES: Motor agitation is a neuropsychiatric symptom of dementia that signals distress and poses a safety risk. Development of digital phenotypes for this symptom could enable better detection, monitoring, and personalized care. Our aim is to establish whether data from a location-monitoring system, alone or combined with clinical assessments, can be used in a machine learning model to classify 8-hour shifts as motor agitation or normal motor activity in individuals with dementia. DESIGN: A prospective, longitudinal observational study. SETTING: A psychogeriatric unit for individuals with behavioral and psychological symptoms of dementia. PARTICIPANTS: 47 older adults with dementia. MEASUREMENTS: Motor agitation was rated on a shift-by-shift basis by clinical staff using the Pittsburgh Agitation Scale. Information about movement in the unit was extracted from a real-time clinical safety system. Models were developed to predict the presence of motor agitation using combinations of clinical and location-based features. RESULTS: Our findings indicate that using location-based trajectory metrics with clinical measures significantly improves the predictability of motor agitation beyond individual baselines. Our best model achieves an area under the curve of the receiver operating characteristic of 0.81 in classifying presence of motor agitation. Explainability analysis identified that trajectory-based features were the most important features for correctly classifying motor agitation. CONCLUSIONS: Location data can be used to distinguish clinically important motor agitation from normal motor activities on a shift-by-shift basis. Behavioral analytics derived from location data are informative for clinical assessment of neuropsychiatric symptoms in dementia.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".