Exploring the Applicability of Using Natural Language Processing to Support Nationwide Venous Thromboembolism Surveillance: Model Evaluation Study
Bibliographic record
Abstract
BACKGROUND Conducting public health surveillance for venous thromboembolism (VTE) at a national scale is important for measuring the disease burden and the impact of prevention measures. Integrating natural language processing (NLP) into VTE surveillance may be an efficient and accurate option in establishing a sustainable and cost-effective national surveillance system. OBJECTIVE We evaluated the performance of the VTE identification instance of IDEAL-X, an NLP tool, in automatically classifying cases of VTE from “reading” unstructured text from diagnostic imaging records. METHODS Accessing imaging records from pilot surveillance systems for VTE from Duke University and the University of Oklahoma Health Sciences Center (OUHSC) during 2012–2014, we used a VTE identification model of IDEAL-X to classify cases of VTE that had previously been manually classified according to pre-defined criteria. The performance measures (and 95% confidence intervals [CI]) calculated were accuracy, sensitivity, specificity, and positive and negative predictive values. RESULTS The VTE model of IDEAL-X “read” 1591 records from Duke University and 1487 records from OUHSC for a total of 3078 records. The combined performance measures were 93.7% accuracy (95% CI: 93.7%–93.8%), 96.3% sensitivity (95% CI: 96.2%–96.4%), 92.0% specificity (95% CI: 91.9%–92.0%), 89.1% positive predictive value (95% CI: 89.0%–89.2%), and 97.3% negative predictive value (95% CI: 97.3%–97.4%). The sensitivity was higher at Duke University (97.9%, 95% CI: 97.8%–98.0%) than at OUHSC (93.8%, 95% CI: 93.5%–94.0%), but the specificity was higher at OUHSC (96.3%, 95% CI: 96.2%–96.3%) than at Duke University (90.4%, 95% CI: 90.1%–90.5%). CONCLUSIONS The VTE model of IDEAL-X accurately classified cases of VTE from pilot surveillance systems from 2 separate states and health systems. NLP is a promising tool in the design and implementation of an automated national surveillance system for VTE.
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 | 0.022 | 0.050 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".