Preliminary data from a multicenter Italian study: use of EUS-Elastography and Contrast in the differential diagnosis of SELs
Bibliographic record
Abstract
Aims Distinguishing gastrointestinal subepithelial lesions (GI-SELs) poses a clinical challenge because Endoscopic Ultrasound (EUS) is adept at detecting them but may fall short in providing effective differentiation. New methodologies have been introduced which provide further details and potential prognostic information. Elastography (EUS-E) allows to carry out a qualitative and semi-quantitative assessment of tissue stiffness, but for now only a few studies have examined its role in the diagnosis of SELs. Recent findings indicate that contrast agents enhance the diagnostic accuracy of EUS (CE-EUS) for SELs. The purpose of the study is to examine the performance of the EUS-E and CE-EUS in differentiating GI-SELs, and in particular gastrointestinal stromal tumors (GISTs). Methods From March 2021 to June 2023, all patients referred to 4 Italian Institutions for EUS-guided fine needle biopsy (FNB) of GI-SELs who agreed to participate in the study were prospectively enrolled. Endosonographic patterns were compared with the analysis of the histological samples taken through FNB or subsequent endoscopic/surgical resection. Sensitivity (Se), specificity (Sp) and accuracy (Ac) of an increasing number of variables in the diagnosis of GIST were then evaluated. Results We present the data of the first 56 enlisted patients. The cohort had a balanced M:F ratio of 1, with a median age of 65 years (range: 26-85). Lesions’ median size was 29 mm (12-90), 38 were detected in the stomach, while 9 in the duodenum and 9 in the esophagus. All lesions were solid and 51 of them originated from the muscolaris propria, 52 were hypoechoic. Qualitative elastography (based on color) revealed homogeneous blue (stiff) lesion in 40 cases, 1 green (soft) lesion and 15 mixed lesions. During contrast infusion, 40 lesions showed hyper-enhancement, 9 hypo-enhancement and 7 iso-enhancement; contrast distribution was evaluated as inhomogeneous in 32 cases. Final diagnosis showed 38 GISTs, 10 leiomyomas and 8 among lipomas, NETs and others. The basic characteristics of EUS (origin 4 th layer and hypo-echogenicity) have a Se, Sp and Ac of 94,7%, 22,2% and 71,5% respectively; qualitative elastography showed Se, Sp and Ac of 71,1%, 44,4% and 62,5% respectively; elastography plus CE-hyperenhancement corresponded to Se, Sp and Ac of 61,1%, 66,7% and 63,0%; and if contrast diffusion was also evaluated as inhomogeneous Se, Sp and Ac were 41,7%, 83,3% and 55,5% respectively. [ 1 ] [ 2 ] [ 3 ] [ 4 ] Conclusions Our findings indicate that EUS alone is highly sensitive and reasonably accurate in detecting GISTs, albeit lacking specificity. However, when dealing with a lesion that raises suspicion for GIST, employing certain enhancing EUS techniques can significantly enhance specificity. This can be of help in the decision-making process, especially where a biopsy is not technically easy to obtain, but, in the concrete suspicion of a GIST, it is of fundamental importance to obtain it. Publication History Article published online: 15 April 2024 © 2024. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
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 | 0.008 | 0.023 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
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".