Titre :

Performance of an Open-Source Large Language Model in Extracting Information from Free-Text Radiology Reports

Auteur(s) :

Le Guellec, Bastien [Auteur]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Departement de Neuroradiologie [Lille]
Lefevre, Alexandre [Auteur]
Departement de Neuroradiologie [Lille]
Geay, Charlotte [Auteur]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Shorten, Lucas [Auteur]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Bruge, Cyril [Auteur]
Departement de Neuroradiologie [Lille]
Hacein-Bey, Lotfi [Auteur]
University of California [Davis] [UC Davis]
Amouyel, Philippe [Auteur]
Facteurs de Risque et Déterminants Moléculaires des Maladies liées au Vieillissement - U 1167 [RID-AGE]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Pruvo, Jean-Pierre [Auteur]
Lille Neurosciences & Cognition (LilNCog) - U 1172
Plateformes Lilloises en Biologie et Santé - UAR 2014 - US 41 [PLBS]
215018|||Departement de Neuroradiologie [Lille] (VALID)
Kuchcinski, Gregory [Auteur]
Plateformes Lilloises en Biologie et Santé - UAR 2014 - US 41 [PLBS]
215018|||Departement de Neuroradiologie [Lille] (VALID)
Lille Neurosciences & Cognition (LilNCog) - U 1172
Hamroun, Aghiles [Auteur]
Facteurs de Risque et Déterminants Moléculaires des Maladies liées au Vieillissement - U 1167 [RID-AGE]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]

Titre de la revue :

Radiology: Artificial Intelligence

Nom court de la revue :

Radiol Artif Intell

Numéro :

6

Pagination :

e230364

Éditeur :

RSNA

Date de publication :

2024-05-08

ISSN :

2638-6100

Mot(s)-clé(s) en anglais :

Large Language Model (LLM)
Generative Pretrained Transformers (GPT)
Open Source
Information Extraction
Report
Brain
MRI

Discipline(s) HAL :

Sciences du Vivant [q-bio]

Résumé en anglais : [en]

Purpose To assess the performance of a local open-source large language model (LLM) in various information extraction tasks from real-life emergency brain MRI reports. Materials and Methods All consecutive emergency ...
Lire la suite >Purpose To assess the performance of a local open-source large language model (LLM) in various information extraction tasks from real-life emergency brain MRI reports. Materials and Methods All consecutive emergency brain MRI reports written in 2022 from a French quaternary center were retrospectively reviewed. Two radiologists identified MRI scans that were performed in the emergency department for headaches. Four radiologists scored the reports’ conclusions as either normal or abnormal. Abnormalities were labeled as either headache-causing or incidental. Vicuna (LMSYS Org), an open-source LLM, performed the same tasks. Vicuna’s performance metrics were evaluated using the radiologists’ consensus as the reference standard. Results Among the 2398 reports during the study period, radiologists identified 595 that included headaches in the indication (median age of patients, 35 years [IQR, 26–51 years]; 68% [403 of 595] women). A positive finding was reported in 227 of 595 (38%) cases, 136 of which could explain the headache. The LLM had a sensitivity of 98.0% (95% CI: 96.5, 99.0) and specificity of 99.3% (95% CI: 98.8, 99.7) for detecting the presence of headache in the clinical context, a sensitivity of 99.4% (95% CI: 98.3, 99.9) and specificity of 98.6% (95% CI: 92.2, 100.0) for the use of contrast medium injection, a sensitivity of 96.0% (95% CI: 92.5, 98.2) and specificity of 98.9% (95% CI: 97.2, 99.7) for study categorization as either normal or abnormal, and a sensitivity of 88.2% (95% CI: 81.6, 93.1) and specificity of 73% (95% CI: 62, 81) for causal inference between MRI findings and headache. Conclusion An open-source LLM was able to extract information from free-text radiology reports with excellent accuracy without requiring further training.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Établissement(s) :

Université de Lille
Inserm
CHU Lille

Collections :

• Facteurs de risque et déterminants moléculaires des maladies liées au vieillissement (RID-AGE) - U1167
• Lille Neurosciences & Cognition (LilNCog) - U 1172
• Plateformes Lilloises en Biologie et Santé (PLBS) - UAR 2014 - US 41

Équipe(s) de recherche :

Troubles cognitifs dégénératifs et vasculaires
Santé publique et épidemiologie des maladies vasculaires

Date de dépôt :

2024-05-29T21:06:23Z
2024-12-04T14:55:35Z