Open-LBP-RF: A Clinical Note Dataset annotated with Lower Back Pain Risk Factors
Bibliographic record
Abstract
Abstract Choosing Wisely Canada identifies six red-flag risk factors that justify imaging for patients with lower back pain (LBP), yet existing datasets for studying these indicators remain publicly inaccessible, small, and manually annotated. We present Open-LBP-RF, the first publicly available clinical note dataset annotated for LBP imaging risk factors. Our annotations are generated using a structured prompting framework called R2D2-G (Role-play, Domain, Rules, Output format, Demonstration, Guidance), which guides large language models (LLMs) to label risk factors using chain-of-thought rationales in structured JSON format. We demonstrate that R2D2-G achieves a competitive weighted F1-score (0.89) on zero-shot risk factor classification on an expert-annotated set. Qualitative analysis shows it uncovers both valid clinical mentions and human labeling errors. We release the Open-LBP-RF dataset and the novel R2D2-G prompting framework to facilitate further work on LLM-based clinical annotation and risk-aware triage for lower back pain. Given the importance of public clinical annotations, our approach represents a path from prompt calibration on internal data to public dataset annotation at scale. Our dataset is available at: https://huggingface.co/datasets/Aman-J/OPEN-LBP-RF. This article is part of the Proceedings of the BioCreative IX Challenge and Workshop (BC9): Large Language Models for Clinical and Biomedical NLP at the International Joint Conference on Artificial Intelligence (IJCAI).
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 | 0.019 |
| 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 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.010 | 0.007 |
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".