P008 What type of evidence do we need to develop guidelines for diagnostic imaging?
Bibliographic record
Abstract
Background Diagnostic imaging (DI) is used several ways in patient management, and the evidence required for each of these roles is somewhat different. This presentation will focus on the evidence needed to develop guidelines for the use of DI in primary diagnosis. Context GRADE states that randomised control trials of patient outcomes are the highest level of evidence for assessing diagnostic tests but also that accuracy can be used as a proxy for outcomes. DI guidelines provide two basic types of information: whether DI is indicated in a particular clinical situation and what is the best DI modality to use. In choosing a modality the accuracy of different DI modalities is important. However, the question of whether DI is indicated in a given clinical situation is at least as important, and in determining this, accuracy is less important. Best Practice The type of evidence which is needed for this question relates to whether DI will affect the management of the patient. If the information that DI provides is not relevant to the management of the patient then DI is not indicated. If the pre-test probability of the diagnosis is very low or very high then DI is also not indicated. Lessons When developing guidelines for DI first consider whether the type of information DI can provide is important in patient management. If it is, clinical decision rules are important in assessing whether the pre-test probability justifies its use. Accuracy only becomes important in determining which imaging modality to recommend.
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 | 0.111 | 0.577 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.007 | 0.007 |
| Bibliometrics | 0.013 | 0.010 |
| Science and technology studies | 0.003 | 0.008 |
| Scholarly communication | 0.013 | 0.019 |
| Open science | 0.010 | 0.006 |
| Research integrity | 0.026 | 0.018 |
| Insufficient payload (model declined to judge) | 0.020 | 0.012 |
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".