SafeRespirator: Comprehensive Database for N95 Filtering Facepiece Respirator Leakage Detection Including Infrared, RGB Videos, and Quantitative Fit Testing
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
The COVID-19 pandemic underscored the challenges of performing mandatory Quantitative Fit Tests (QNFT) for healthcare professionals and the limitations of self-administered fit checks. To address this, it is crucial to develop faster and more efficient methods for detecting, locating, and quantifying Filtering Facepiece Respirators (FFRs) leakage, providing wearers with immediate feedback on their safety. Infrared (IR) technology, which relies on temperature variation analysis around the face seal, has proven effective for locating leakage but has not yet achieved automated quantification. This paper introduces a validated protocol for creating a comprehensive database to advance automatic leakage detection. The database includes synchronized and calibrated IR and RGB video data, along with QNFT results, collected from 62 participants wearing four different N95 FFR models in four distinct positions. High-performance IR and RGB cameras were used to precisely capture temperature variations, while a PortaCount® instrument served as the reference for fit quantification. Preliminary results using the MediaPipe approach with synchronized and calibrated RGB and IR videos demonstrate that precise tracking of the human face is achievable even with an FFR. The normalized cross-correlation methods further highlight the capability of IR imaging to accurately monitor and detect leakage. This breakthrough paves the way for real-time, automated detection of N95 FFR leakage, potentially deployable at operator workstations. This large, high-quality, open-access database is available to the scientific community to drive innovation in respiratory protection research and beyond.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| 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) | 0.000 | 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