MétaCan
Menu
Back to cohort
Record W7117235256 · doi:10.1002/alz70858_100300

Spatio‐temporal digital markers of behavioural and psychological symptoms to support clinical dementia care

2025· article· en· W7117235256 on OpenAlexaff
Leia C Shum, Yasser Karam, Zain Ul Hasan, Shehroz S. Khan, Andrea Iaboni

Bibliographic record

VenueAlzheimer s & Dementia · 2025
Typearticle
Languageen
FieldComputer Science
TopicContext-Aware Activity Recognition Systems
Canadian institutionsUniversity of TorontoToronto Rehabilitation Institute
Fundersnot available
KeywordsDementiaDiseaseReal-time locating systemCognitionPatient careHealth care

Abstract

fetched live from OpenAlex

BACKGROUND: Real-time location systems (RTLS) are increasingly used within nursing homes and seniors' residential settings, primarily as safety and nurse call systems. Beyond location monitoring, RTLS collects rich information about movement over time, which can be processed to derive clinical insight into resident behaviours and health, as well as measure impact in interventions. The Space-Time Indices for Clinical Support (STICS) project aims to derive clinical markers and phenotypes of behavioural health in people with dementia to support longitudinal monitoring. This presentation will provide an overview of our work on two behavioural health markers: motor agitation and rest-activity rhythms. METHODS: In the STICS pilot, location data was collected from a clinical RTLS system installed on a 20-bed secure inpatient dementia care unit. Weekly health and behavioural assessments supplemented and provided clinical labels for machine learning models. RTLS data was used to build motor agitation detection models based on shift-by-shift Pittsburgh Agitation Scale scores and generate six rest-activity 'profiles' using hierarchical clustering methods. RESULTS: 47 people with dementia participated in the study, with a mean of 7 weeks of location data per person. Participants were 45% female, had a mean MMSE of 5/30 (range 0-23), and a mean baseline NPI of 39 (range 0-110). 15 participants used gait aids and 10 of 13 wheelchair users could self-propel. Models distinguished motor agitation from normal motor activities with a best AUROC of 0.81. SHAP explainability analysis determined that 17 of the top 20 model features were RTLS-based, with movement speed and total distance being key predictors. Using unsupervised deep learning, six digital phenotypes for rest-activity were identified of which two had well-regulated circadian rhythms. The remainder included a cluster with high night activity, one with a high degree of day-to-day instability, one with a high time in bed over the day and night, and one marked by severe rhythm disturbance. CONCLUSIONS: RTLS systems are a low-effort method to objectively track changes in resident movements and behaviours longitudinally. We demonstrated that RTLS-derived digital markers can describe behavioural symptoms in dementia, although further validation in long-term care settings is needed.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.005
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.004
Threshold uncertainty score0.014

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.005
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0020.002
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0000.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0040.001

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.

Opus teacher head0.082
GPT teacher head0.348
Teacher spread0.266 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueAlzheimer s & DementiaSame topicContext-Aware Activity Recognition SystemsFrench-language works237,207