Abstracts from the CanVECTOR 2022 Annual Conference October 14th, 2022
Bibliographic record
Abstract
Background: Pulmonary embolism (PE) can be misdiagnosed when emergency physicians do not consider PE as a potential diagnosis.Little research has focused on who should be tested for PE.Aim: To develop a predictive model which identifies underlying PE diagnosis in emergency department patients.Methods: Using linked population-level administrative data from Ontario (Canada), we analyzed the first patient emergency department visit in each calendar year between 2015 and 2019.Patients were classified as presenting with PE if they were diagnosed with PE on that visit, within the following 30 days if there was no hospitalization, within seven days of hospitalization without surgery, or else prior to surgery within seven days.We collected patient data on age, sex, cancer, prior DVT and PE, pregnancy, coronary artery disease, COPD, diabetes, stroke, anticoagulant use, atrial fibrillation, recent hospitalization, recent surgery, presenting complaint and triage score.Logistic regression and 10-fold cross-validation were used to derive a model which predicts the diagnosis of PE for the year 2015, and subsequently validated the model performance on each calendar year 2016-2019.Results: We analyzed 7,024,111 emergency department visits with 18,751 cases of PE.Mean age 48.6, 52% female, 3% history of cancer, 1.7% stroke, 1% prior PE, 1.6% prior DVT, 1.9% recent admission and 4.4% recent surgery.Our final model predicting a diagnosis of PE contained age, prior PE, prior DVT, presenting complaint and triage category.The area under the curve for the predicting PE was 82.8 in 2015, and 84.5, 81.3, 78.4 and 80.6 for each subsequent year.The optimal cut-off gave sensitivity estimates between 69.4%-74.4% and specificity estimates 73.7%-76.0%for each of the 5 years. Conclusion:We derived a model which accurately predicts the diagnosis of PE in emergency department patients, with the potential to trigger PE testing in the emergency department.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.004 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.627 | 0.372 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".