Testing remote feedback using a virtual semi-automated educational tool for the detection of pancreatic tumour-vessel contact on staging CT
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
PURPOSE: Assessing tumour-vessel contact in pancreatic adenocarcinoma on CT is challenging for trainees and time-intensive for educators. Semi-automating feedback on this task may optimize radiologist time and standardize resident education. We hypothesized that residents who reviewed expert annotations of tumour-vessel contact would outperform those without feedback on an independent test set. METHODS: We retrospectively reviewed pre-operative staging CTs from 60 patients who underwent upfront surgical resection for pancreatic adenocarcinoma. Two resident groups (control and test) independently annotated tumour contact with the superior mesenteric artery. The test group received feedback-annotations from three expert radiologists-for the first 30 cases; the control group received none. Resident performance on the remaining 30 cases was compared against both surgical pathology and expert annotations. RESULTS: Test group residents demonstrated higher sensitivity than control group residents (mean sensitivity = 93 % vs. 79 %), with comparable specificity and accuracy relative to surgical pathology. While both groups performed similarly relative to expert consensus, the test group showed greater consistency in sensitivity (mean variation = 29 % vs. 46 %). CONCLUSION: Virtual expert feedback improved resident sensitivity in identifying tumour-vessel contact without compromising specificity or accuracy. These findings support the use of semi-automated educational tools to enhance radiology training efficiency and effectiveness.
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.
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 | 0.016 |
| 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 it