COVID-19 vaccine evidence monitoring assisted by artificial Intelligence: An emergency system implemented by the Public Health Agency of Canada to capture and describe the trajectory of evolving pandemic vaccine literature
Bibliographic record
Abstract
Background: ), which contributed to situational awareness in Canada through a bibliographic repository used to support decision-making by the National Advisory Committee on Immunization. We describe the process by which this literature was identified and catalogued, and provide an overview of characteristics in the identified literature. Methods: To expedite the process, PHAC leveraged an artificial intelligence (AI) tool to assist in the screening and selection of relevant articles. Literature search results were initially screened by AI, then manually reviewed for relevance. Relevant articles were tagged using controlled vocabulary and stored in a bibliographic repository. This repository was analyzed to identify trends in vaccine research over time according to several key characteristics. Results: As of December 31, 2023, EXTRA's repository contained 19,050 articles relevant to PHAC's immunization mandate. The majority of these articles (63.9 %) were identified between August 2021 and January 2023, with an average of 20 relevant articles added daily during this period. Nearly 14,000 articles reported on mRNA vaccines. Safety outcomes were most frequently reported (n = 8,289), followed by immunogenicity (n = 7,269) and efficacy/effectiveness (n = 3,246). COVID-19 vaccine literature output started to decrease in mid-2023, two years after the initial dramatic increase in mid-2021. Conclusions: This hybrid (AI and human) approach was critical for PHAC situational awareness and the development of timely vaccine guidance in Canada during the COVID-19 pandemic. Given the volume of data and analyses required, the AI-augmented processes made this massive undertaking manageable. Analysis of COVID-19 vaccine research patterns supports projections of research volume, type, and rate that will help predict resourcing and information needs to plan future emergency vaccine guidance activities.
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 | 0.003 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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".