Streamlining Endometriosis MRI Reporting: Automated Extraction of #Enzian Scores from Pelvic MRI Reports Using Local and Online LLMs

Joana Kostova, Caroline Reinhold, Jonas D. Stief, Nemanja Avramović, Stoyan Benev, Marina Bucher, Toni Dimov, Andrei Dragnea, Elena Kiyomi Lanza de Cristoforis, Elena Lanza de Cristoforis, Abigail Martin Mens, Katharina Ochs, Ivan Rakovic, Pavel Stegarescu, Eliane Pauli, Jean‐Pierre Pelage, Jean-Pierre Pelage, Christian Houbois, C Houbois, Khashayar Namdar, Dominik Deniffel
In: European Journal of Radiology Artificial Intelligence · 2026 · vol. 7 , pp. 100116 · doi:10.1016/j.ejrai.2026.100116 · W7169566133
OA: gold CC0
🔓 Open OA copy View on OpenAlex View at publisher
⚙ AI-generated summary by gemini-2.5-flash-lite, 2026-07-21 ⓘ

Online LLMs accurately extracted #Enzian scores from endometriosis MRI reports, outperforming novice radiologists and suggesting utility as training aids.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

Abstract

Background The #Enzian classification standardizes endometriosis reporting but requires specialized training. This study evaluated whether large language models (LLMs) can accurately extract #Enzian scores from MRI reports to support novice radiologists and clinicians. Methods This retrospective study included 186 pelvic MRI reports (2022–2025). Two uroradiologists established report-based reference #Enzian scores for all cases via consensus. Twelve LLMs—six online (Gemini 2.5 Pro, Claude Sonnet 4.0, GPT-5, o3, Grok 4, Claude Opus 4.1) and six local (Llama 3.1 8B, Aya Expanse 8B, MedGemma 27B, DeepSeek R1 32B, Llama 3.3 70B, GPT-OSS 120B)—were tested with one-shot prompting. A 50-case subset was additionally scored by 12 trainees without #Enzian experience as novice benchmark. Accuracy for exact #Enzian score matching was assessed against the expert reference; 95% bootstrap confidence intervals were calculated for accuracies, and logistic regression with Dunnett-adjusted simultaneous inference and cluster bootstrapping was used to compare LLM and trainee performance. Results In the full 186-report cohort, online LLMs achieved 93.8% [95% CI: 92.4–95.1] to 95.8% [95% CI: 94.7–96.8] accuracy vs. expert reference, while local models reached 72.7% [95% CI: 70.0–75.5] to 92.5% [95% CI: 91.1–93.8]. In the 50-case subset, trainees achieved 87.4% [95% CI: 84.4–89.9] accuracy. Online LLMs reached 88.9% [95% CI: 86.1–91.3] to 92.4% [95% CI: 90.1–94.7], with three models significantly outperforming trainees: Gemini 2.5 Pro (OR 1.81, p<0.001), Grok 4 (OR 1.60, p=0.012), and o3 (OR 1.57, p=0.034). Local models achieved 61.7% [95% CI: 56.4–67.0] to 87.1% [95% CI: 83.9–90.0] and were significantly inferior to trainees (p<0.001) except GPT-OSS 120B (OR 0.98, p=1.00). Online LLM API costs for the full cohort ranged from $3.82 to $42.26. Conclusions Online LLMs accurately extracted #Enzian scores from MRI reports with performance exceeding that of novice radiologists, suggesting potential as training aids and for standardized reporting in surgical planning. Out-of-the-box local models demonstrated limited performance.

My notes (saved in your browser only)

Citation neighborhood (2-hop)

Papers in the corpus that this work cites (lower rings, blue) and that cite this one (upper rings, green). Dot size scales with the paper's in-corpus citation count — bigger dot = more influential within the endo/adeno field. Click a dot to open that paper. Outer rings show 2-hop neighbours — papers reached through the immediate citers/citees. [ collapse to 1-hop ]

References (24)

Source provenance

openalex
last seen: 2026-10-15T06:00:39.803327+00:00
License: CC0 · commercial use OK