Improving the Reporting Quality of Studies on Information Extraction From Clinical Texts: Protocol for the Development of a Consensus-Based Reporting Guideline
Bibliographic record
Abstract
Background: Information extraction (IE) from clinical texts is increasingly important in health care; yet, reporting practices remain inconsistent. Existing guidelines do not fully address the unique challenges of IE studies. IE methods vary widely in their design, ranging from rule-based systems to advanced large language models, contributing to heterogeneity in reporting. While several reporting frameworks exist for applications of artificial intelligence in health care, they primarily focus on prediction modeling or clinical trials and associated protocols rather than text-based IE. Objective: This study aims to develop the Clinical Information Extraction (CINEX) guideline, a consensus-based reporting guideline for studies on clinical IE. Methods: The CINEX guideline is developed following an established guideline methodology, including a 3-round electronic Delphi (eDelphi) study with domain experts and a final in-person consensus meeting. The eDelphi process includes feedback loops and predefined consensus thresholds, with items rated on a 10-point scale for both relevance and maturity. The final consensus meeting is held as a hybrid workshop at the MEDINFO 2025 conference and focuses on finalizing the items that reached consensus. Results: Our results will provide a validated reporting guideline for studies on clinical IE. A preliminary set of 28 reporting items was drafted from a scoping review and existing frameworks. The draft guidelines include 5 key dimensions: information model, architecture, data, annotation, and outcome. This draft guideline will be refined through the eDelphi process. It is designed to be technology-agnostic and applicable across diverse IE approaches, including not only large language models but also traditional machine learning methods and rule-based and hybrid systems. Conclusions: The CINEX guideline provides structured, expert-validated guidance for reporting clinical IE studies, improving transparency, reproducibility, and comparability. The final guideline will be disseminated alongside an explanatory document to support adoption and implementation.
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 | 0.465 | 0.579 |
| Meta-epidemiology (narrow) | 0.006 | 0.007 |
| Meta-epidemiology (broad) | 0.010 | 0.019 |
| Bibliometrics | 0.019 | 0.018 |
| Science and technology studies | 0.007 | 0.009 |
| Scholarly communication | 0.014 | 0.012 |
| Open science | 0.009 | 0.013 |
| Research integrity | 0.015 | 0.016 |
| Insufficient payload (model declined to judge) | 0.025 | 0.013 |
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".