Predicting air pollution spatial variation with street-level imagery
Bibliographic record
Abstract
Air pollution has important and widely accepted implications on health outcomes [1,2,3,4].Advances in spatial and temporal estimation have been instrumental both in establishing health effects [2] and estimating impact on the burden of disease [1,5].Yet, we currently lack information on air pollution levels in many areas globally [1], especially at high spatial resolution.Limited spatial and temporal coverage of air quality models is partially due to input data requirements of existing estimation approaches including physically-based e.g., dispersion models, and geostatistical e.g., land use regression (LUR) models [6].Advances in deep learning methods and their success in computer vision applications led to a growing interest in using images for estimating air pollution levels [6,7,8,9,10,11,12,13].The rationale behind this interest is that information on common inputs to traditional approaches (e.g.land use, traffic, and built environment features) is, at least partially, visible from street-level and satellite images.If utilized effectively, this approach has the potential for scaling up to global coverage at high spatial and temporal resolution with increasing availability of imagery at low cost.Building on existing work, for this study, we obtained LUR based mean annual NO 2 (200m resolution raster) and PM 2.5 (100m resolution raster) estimates for 2010 from four cities: London, New York, Los Angeles, and Vancouver (BC) [14,15,16,17,18].To obtain images, a 100m grid for each city was created using the boundary shape files covering LUR model outputs.For each of the grid centroid points, we obtained the nearest street-level panorama image available from Google Street View that
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 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".