Using Wearable Sensors to Measure and Predict Personal Circadian Lighting Exposure in Nursing Home Residents: Model Development and Validation
Bibliographic record
Abstract
Background: Lighting, especially circadian lighting, significantly affects people with dementia, influencing sleep patterns, daytime alertness, and behavioral symptoms such as agitation. Since individuals experience and respond to light differently, measuring personal lighting exposure is essential for understanding its impact on health. Without individual data, the connection between lighting and health outcomes remains unclear. Wearable sensors provide a practical way to track personal light exposure, helping researchers better assess its effects on circadian rhythms and overall well-being. Objective: This study aims to develop and validate both calibration and predictive models using wearable lighting sensors to assess individual circadian lighting exposure accurately. By leveraging machine learning techniques and empirical data, we seek to establish a reliable method for health care researchers and practitioners to investigate and optimize lighting conditions for improved circadian health in nursing homes, especially for residents with dementia. Methods: A combination of controlled laboratory experiments and on-site data collection was conducted using professional spectrophotometer measurements as ground truth. Calibration models were developed for photopic lux and correlated color temperature, while predictive models estimated circadian metrics such as circadian stimulus. The sensors and the developed models were implemented in a real-world health care research project about bright light therapy intervention at 2 assisted-living facilities. Results: The calibration models for photopic lux and correlated color temperature demonstrated strong accuracy, with an adjusted R² of 0.858 and 0.982, respectively, ensuring reliable sensor measurements. Predictive models for circadian stimulus were developed using both simple regression and machine learning techniques. The random forest model outperformed linear regression, achieving an adjusted R² of 0.915 and a cross-validation R² of 0.857, demonstrating high generalization capability. Upon the implementation of these models, significant individual variations in circadian light exposure were found in the study, highlighting the significance of customized lighting evaluations. These results confirm the effectiveness of wearable sensors, combined with the developed calibration and predictive modeling, in accurately assessing personal circadian light exposure and supporting lighting-related health care research. Conclusions: This study introduces an effective and scalable approach to circadian light assessment using wearable sensors and predictive modeling. By replacing labor-intensive and costly spectrometer measurements, the proposed methodology enables continuous, cost-effective monitoring in health care environments. However, challenges related to sensor wearability, durability, and user compliance were identified, underscoring the need for further sensor design refinements. Future research should focus on refining sensor integration, expanding case studies, and developing adaptive lighting interventions to enhance circadian health in vulnerable populations.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| 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".