Comparison of a Natural Language Processing Model and Large Language Models for Extracting Incidental Lung Nodule Data

Radiology Cardiothoracic Imaging 2026

View at publisher
Authors: João Martins da Fonseca, Alysson Roncally Carvalho, Rosana Souza Rodrigues, Marco Aurélio Veiga Conrado, Rodrigo Basilio, Thiago Machuca, Bruno Hochhegger
DOI: 10.1148/ryct.250210
Cited by: 1
References: 6

Abstract

A task-specific natural language processing model demonstrated higher accuracy and reliability when compared with seven general-purpose large language models for extracting incidental lung nodule data from unstructured chest CT radiology reports.

Fields of Study & Subjects

Lung Cancer Diagnosis and Treatment COVID-19 diagnosis using AI Radiology practices and education

Publication details

Journal / source
Radiology Cardiothoracic Imaging
Publisher
Radiological Society of North America (RSNA)
Volume
8
Issue
1
Pages
e250210–e250210
Publication date
1 February 2026
Language
EN