Development and Evaluation of an Automated Protocol Recommendation System for Chest CT Using Natural Language Processing With CLEVER Terminology Word Replacement
Bibliographic record
Abstract
Purpose: To evaluate the clinical performance of a Protocol Recommendation System (PRS) automatic protocolling of chest CT imaging requests. Materials and Methods: 322 387 consecutive historical imaging requests for chest CT between 2017 and 2022 were extracted from a radiology information system (RIS) database containing 16 associated patient information values. Records with missing fields and protocols with <100 occurrences were removed, leaving 18 protocols for training. After freetext pre-processing and applying CLEVER terminology word replacements, the features of a bag-of-words model were used to train a multinomial logistic regression classifier. Four readers protocolled 300 clinically executed protocols (CEP) based on all clinically available information. After their selection was made, the PRS and CEP were unblinded, and the readers were asked to score their agreement (1 = severe error, 2 = moderate error, 3 = disagreement but acceptable, 4 = agreement). The ground truth was established by the readers’ majority selection, a judge helped break ties. For the PRS and CEP, the accuracy and clinical acceptability (scores 3 and 4) were calculated. The readers’ protocolling reliability was measured using Fleiss’ Kappa. Results: Four readers agreed on 203/300 protocols, 3 on 82/300 cases, and in 15 cases, a judge was needed. PRS errors were found by the 4 readers in 1%, 2.7%, 1%, and 0.7% of the cases, respectively. The accuracy/clinical acceptability of the PRS and CEP were 84.3%/98.6% and 83.0%/99.3%, respectively. The Fleiss’ Kappa for all readers and all protocols was 0.805. Conclusion: The PRS achieved similar accuracy to human performance and may help radiologists master the ever-increasing workload.
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 | 0.003 | 0.000 |
| 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 | 0.000 |
| 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".