An intelligent video-monitoring system to detect falls: a proof of concept
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Purpose To face the challenges raised by the high incidence of falls among older adults, the intelligent video-monitoring system (IVS), a fall detection system that respects privacy, was developed. Most fall detection systems are tested only in laboratories. The purpose of this paper is to test the IVS in a simulation context (apartment-laboratory), then at home. Design/methodology/approach This study is a proof of concept including two phases: a simulation study to test the IVS in an apartment-laboratory (29 scenarios of activities including falls); and a 28-day pre-test at home with two young occupants. The IVS’s sensitivity (Se), specificity (Sp), accuracy (A) and error rate (E) in the apartment-laboratory were calculated, and functioning at home was documented in a logbook. Findings For phase 1, results are: Se =91.67 per cent, Sp =99.02 per cent, A=98.25 per cent, E=1.75. For phase 2, the IVS triggered four false alarms and some technical dysfunctions appeared (e.g. computer screen never turning off) that are easily overcome. Practical implications Results show the IVS’s efficacy at automatically detecting falls at home. Potential issues related to future installation in older adults’ homes were identified. This proof of concept led to recommendations about the installation and calibration of a camera-based fall detection system. Originality/value This paper highlights the potentialities of a camera-based fall detection system in real-world contexts and supports the use of the IVS to help older adults age in place.
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.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 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 it