OP049 Topic: AS04–Emerging Sciences, Methodologies, Big Data and Technology: AUTOMATIC RECRUITMENT MANEUVERS DETECTION IN MECHANICALLY VENTILATED PEDIATRIC PATIENT
Bibliographic record
Abstract
Aims & Objectives: Invasive ventilation is often needed for acute hypoxemic respiratory insufficiency in the pediatric intensive care unit (PICU). Recruitment maneuvers (RMs) are a common treatment, but little is known on their clinical impact. Our objective was to develop a computational tool able to detect RMs and further study their impact on oxygenation and clinical outcomes. Methods: Continuous data of every patient on our high-resolution database hospitalized in Sainte-Justine Hospital PICU between 2016-02-01 and 2023-12-30 was included. Data extraction, cleaning and analysis were executed using SQL and R languages. A sliding window function was used on the timeseries data to identify both RMs including intrapulmonary percussive ventilation and intermittent positive pressure breathing and PEEP incremental maneuvers such as PEEP titration and temporary PEEP increase. A random manual record verification was conducted on 30 patients per group to measure the sensitivity and the accuracy of the algorithm. Results: We identified 14 635 RMs on 635 patients and 324 patients who had a PEEP increment. The median age was 25 months (IQR 2.95-130), the median PELOD-2 score on admission was 9 (IQR 6-12) and the median length of stay was 11 days (IQR 5.41-23.50). Our model had a 100% sensitivity for both groups and an accuracy of 100% for RMs and 93% for PEEP increment. The 2 patients misclassified were on non-invasive ventilation. Conclusions: We are able to retrospectively identify 96.7% RMs episodes. This will be useful to study the clinical impact of RMs and to develop clinical decision support systems to identify patients who respond positively. Keywords: Detection Model, clinical decision support systems, hypoxemia, pediatric intensive care, Recruitment Maneuvers
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 | 0.005 | 0.014 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.004 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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".