A115 ACCURACY OF OPTICAL DIAGNOSIS IN ENDOSCOPY AT A TERTIARY ACADEMIC CENTER
Bibliographic record
Abstract
Abstract Background Optical diagnosis relies on the ability of the endoscopist to visualize normal and abnormal patterns on the epithelial surface of the gastrointestinal tract. With ongoing technologic advances in image-enhanced and magnifying endoscopy, there has been more attention given to improving our ability to visually evaluate and classify lesions as this can help guide decisions around resection techniques. However, the accuracy of optical diagnosis of epithelial lesions remains under investigated. Aims To analyse (1) the presence or absence of descriptive details (size, gross morphology, and classification systems used) of lesions of interest within the endoscopy report and (2) the accuracy of the optical diagnosis, when stated, as compared to the final pathology report. Methods This is a single-centre retrospective chart review and quality improvement initiative conducted at St. Michael’s Hospital, Toronto, Ontario. All patients who had polypectomy performed between January 1st, 2019 and December 31st, 2020 for polyp(s) > 10mm in size, were eligible for study inclusion. Patients were excluded if polyps did not meet the size criteria, the polyp was not resected, or absent documentation. Descriptive statistics were conducted. Results 2100 patients had polypectomies during the study period. 714 patients with 833 polyps >10mm in size were included in the data analysis. Estimated size was reported for 93% of polyps, gross morphology for 68%, and a classification system for 72%. All three description parameters were reported for 52% of polyps. Predicted pathology was recorded in 41% of polyps. When documented, the accuracy of optical diagnosis was 71%. Conclusions In our study, the presence of key descriptive details attributed to polyps at the time of polypectomy was lower than expected. In addition, an optical diagnosis was documented in less than half of the time. Finally, the overall accuracy of optical diagnosis was lower than predicted potentially related to the underreporting and underutilization the important predictors of submucosal invasion. Considering that making real-time endoscopic diagnoses has major implications on treatment decisions, it is imperative that we work on these skills to improve patient outcomes (increasing R0 resection rates, decreasing recurrence and avoiding unnecessary surgeries). Through this project, recommendations will be made regarding the implementation of synoptic reporting in addition to guiding future quality improvement initiatives. Funding Agencies None
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 | 0.036 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".