Development of a machine learning model for automatic data extraction from breast cancer pathology reports
Bibliographic record
Abstract
Data extraction from medical records is crucial for clinical research, with current methods relying on human annotation. Natural Language Processing (NLP) and Machine Learning-based approaches show promise. We develop and evaluate an NLP pipeline constructed by selecting among four candidate models; ClinicalBERT, PubMedBERT, BioMedRoBERTa and Mistral-Nemo LLM to automate data extraction of 1,795 breast cancer pathology reports obtained from the Providence Health Services Authority in British Columbia. We also explore the effect of further pre-training the BERT-based models using the SQuAD question-answering dataset. Accuracy was evaluated by comparing model output and human annotation. PubMedBERT pre-trained on SQuAD proved to be the best performing model, achieving an overall accuracy of 97.4%. 30 of the 32 FOIs had an accuracy greater than 95.0%. Our model outperformed a previous rule-based algorithm (95.6%). Our findings demonstrate how a high-performing question-answering NLP pipeline for breast cancer pathology can provide a scalable approach to high-fidelity extraction of clinicopathologic features, thereby enhancing research efficiency and improving clinical outcomes.
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 | 0.004 | 0.009 |
| 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 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".