43 NATURAL LANGUAGE PROCESSING MODEL-BASED SOLUTION FOR LABELING BRAIN METASTASIS IN RADIOLOGY REPORTS
Bibliographic record
Abstract
Abstract Brain Cancer Canada Travel Award Recipient Brain metastases (BM) far exceed primary CNS tumours and constitute the majority workload for neuro-oncology care providers. Canadian Cancer Registry captures ~2800 BM identified at the primary cancer diagnosis in stage IV cancer patients, while we estimated that ~10,000 Canadians are diagnosed with BM annually. We aim to develop a natural language processing (NLP) algorithm to scan radiology reports on cancer patients to capture BM diagnoses as they occur. Using the population-based cancer registry data in Alberta Canada, we identified a cancer cohort diagnosed between 2012–2019, with follow-up reports up to 2022. All radiology reports at and post-cancer diagnosis that are related to brain/head were identified for this cohort. A subset of 1817 samples was then manually labeled for BM (yes/no) as a training dataset. We trained two BioBERT models, one using the “Findings” section and the other using the “Impressions” section of these reports. We predicted the BM label as yes when the probability estimate of either model is greater than a threshold. For testing, 1585 reports containing both the “Findings” and “Impressions” sections were selected. These reports were independently labeled by an annotator for external model evaluation. Using a threshold of 0.4, our ensembled model achieved 94.3% accuracy on BM predictions. Our model yielded an F1 score of 0.788, positive predictive value of 0.702, and sensitivity of 0.898. Model performance is currently being evaluated with radiology reports in Ontario, with results expected by the end of May.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".