TD‐P‐32: EVALUATING DISEASE PROGRESSION AND THE IMPACT OF TREATMENTS OR INTERVENTIONS IN PEOPLE WITH DEMENTIA, MCI OR PHYSICAL IMPAIRMENT USING 3D SKELETON MOVEMENT DATA AND MACHINE LEARNING
Bibliographic record
Abstract
Motion-based technologies (MBT) like the Microsoft Kinect and Nintendo Wii can facilitate beneficial group activities for people with dementia, MCI or physical impairments. At the same time, these technologies provide data such as 3D skeleton representations of participants, allowing the objective evaluation of disease progression or the impact of treatments or interventions through analysis of physical movements over time in community and home-based settings. We propose a novel set of objective metrics that utilize 3D skeleton tracking data and machine learning to help understand disease progression and intervention efficacy in people with dementia, MCI or physical impairments. We used the Kinect SDK to capture skeleton data from 60-minute bowling sessions with a group of participants twice per week for 12 weeks. We manually annotated data captured by a second video camera to provide ground-truth labelling for the skeleton data. We employed k-means for pre-processing, and developed a feature vector representing the movement speed of the skeleton data and the movement ranges of body parts over time. We identified poses (e.g., arm raised) using deep learning to evaluate participants’ walking speed, arm swing range, arm swing speed, and upper body stability across each trial to create an objective measure of movement over time as a result of the bowling intervention. 14 participants (average 10/session) provided a total of over 6.5 million frames of skeleton data over 24 sessions (average ∼276,000 frames/session). Walking speed and arm swing speed increased by an average of 6.7% and 14% respectively over the first four sessions then stayed constant for the remainder of the trial. Arm swing range, similarly, increased on average from 82 degree to 114 degrees respective to the legs. Finally, upper body stability, or deviation from a fully upright position, increased 13% over the sessions. Our proposed set of measures, derived from 3D skeleton movement data captured using MBT during a group exercise intervention, identified a change in physical capabilities in participants with dementia, MCI or physical impairments over 12-weeks. This suggests that these metrics, once validated against best-practices clinical evaluations, may provide a means of objectively evaluating the impact of physical interventions.
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 | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| 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".