PD11 First Comprehensive Assessment Of Artificial Intelligence At Canada’s Drug Agency: Evidence Review Methods, Lessons Learned, And Next Steps
Bibliographic record
Abstract
Introduction Canada’s Drug Agency (CDA-AMC) conducted a health technology assessment of RapidAI for detecting ischemic stroke and hemorrhagic stroke to test and learn from its first comprehensive assessment of an artificial intelligence (AI)-enabled health technology. Methods The assessment included a review evaluating the effectiveness, accuracy, and cost-effectiveness of RapidAI for detecting ischemic and hemorrhagic stroke, alongside an implementation review capturing digital infrastructure considerations. Ethics and equity considerations were integrated throughout, informed by literature, patient engagement, and expert input. Checklists and other AI or digital health tools were applied. The Health Technology Expert Review Panel (HTERP), an advisory body to CDA-AMC, reviewed the evidence and developed recommendations on the appropriate use of RapidAI for stroke detection, considering the following domains: unmet clinical need, clinical value, economic considerations, impacts on health systems, and distinct social and ethical considerations. Results Patient input highlighted speed and accuracy in stroke diagnosis. Low certainty clinical evidence suggested that using the AI functionalities of RapidAI to assist diagnoses may result in clinically important time reductions. Its effects on other clinical outcomes were very uncertain. Ethical and equity considerations have implications across the technology life cycle when using RapidAI for detecting stroke; however, little relevant information was identified from the literature. We found no relevant economic evaluations. The implementation review identified key considerations for AI-enabled health technologies for decision-makers. Given the evidence gaps and uncertainty, HTERP could not recommend for or against the use of RapidAI for stroke detection. Conclusions Our appraisal and deliberative processes identified evidence limitations that may be common across many AI-enabled health technologies, identifying challenges that need to be addressed in their evaluation. Based on this experience, for AI evaluations CDA-AMC plans to add AI-specific implementation and other considerations to its evidence reviews and to consider a broader range of information sources.
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 | 0.091 | 0.247 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.005 | 0.008 |
| Bibliometrics | 0.021 | 0.017 |
| Science and technology studies | 0.003 | 0.003 |
| Scholarly communication | 0.015 | 0.006 |
| Open science | 0.006 | 0.006 |
| Research integrity | 0.007 | 0.007 |
| Insufficient payload (model declined to judge) | 0.018 | 0.003 |
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".