Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study

Abstract BackgroundBreast ultrasound is essential for evaluating breast nodules, with Breast Imaging Reporting and Data System (BI-RADS) providing standardized classification. However, interobserver variability among radiologists can affect diagnostic accuracy. Large language...

Full description

Saved in:
Bibliographic Details
Main Authors: Su Miaojiao, Liang Xia, Zeng Xian Tao, Hong Zhi Liang, Cheng Sheng, Wu Songsong
Format: Article
Language:English
Published: JMIR Publications 2025-06-01
Series:JMIR Medical Informatics
Online Access:https://medinform.jmir.org/2025/1/e70924
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849688124116959232
author Su Miaojiao
Liang Xia
Zeng Xian Tao
Hong Zhi Liang
Cheng Sheng
Wu Songsong
author_facet Su Miaojiao
Liang Xia
Zeng Xian Tao
Hong Zhi Liang
Cheng Sheng
Wu Songsong
author_sort Su Miaojiao
collection DOAJ
description Abstract BackgroundBreast ultrasound is essential for evaluating breast nodules, with Breast Imaging Reporting and Data System (BI-RADS) providing standardized classification. However, interobserver variability among radiologists can affect diagnostic accuracy. Large language models (LLMs) like ChatGPT-4 have shown potential in medical imaging interpretation. This study explores its feasibility in improving BI-RADS classification consistency and malignancy prediction compared to radiologists. ObjectiveThis study aims to evaluate the feasibility of using LLMs, particularly ChatGPT-4, to assess the consistency and diagnostic accuracy of standardized breast ultrasound imaging reports, using pathology as the reference standard. MethodsThis retrospective study analyzed breast nodule ultrasound data from 671 female patients (mean 45.82, SD 9.20 years; range 26‐75 years) who underwent biopsy or surgical excision at our hospital between June 2019 and June 2024. ChatGPT-4 was used to interpret BI-RADS classifications and predict benign versus malignant nodules. The study compared the model’s performance to that of two senior radiologists (≥15 years of experience) and two junior radiologists (<5 years of experience) using key diagnostic metrics, including accuracy, sensitivity, specificity, area under the receiver operating characteristic curve, P ResultsChatGPT-4 achieved an overall BI-RADS classification accuracy of 96.87%, outperforming junior radiologists (617/671, 91.95% and 604/671, 90.01%, PPPP ConclusionsIntegrating ChatGPT-4 into an image-to-text–LLM workflow improves BI-RADS classification accuracy and supports radiologists in breast ultrasound diagnostics. These results demonstrate its potential as a decision-support tool to enhance diagnostic consistency and reduce variability.
format Article
id doaj-art-e30849d533a245199f8ddfe9fa480d24
institution DOAJ
issn 2291-9694
language English
publishDate 2025-06-01
publisher JMIR Publications
record_format Article
series JMIR Medical Informatics
spelling doaj-art-e30849d533a245199f8ddfe9fa480d242025-08-20T03:22:07ZengJMIR PublicationsJMIR Medical Informatics2291-96942025-06-0113e70924e7092410.2196/70924Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective StudySu Miaojiaohttp://orcid.org/0009-0000-5251-1003Liang Xiahttp://orcid.org/0009-0005-2451-7535Zeng Xian Taohttp://orcid.org/0009-0000-7875-7079Hong Zhi Lianghttp://orcid.org/0000-0001-8995-9995Cheng Shenghttp://orcid.org/0000-0002-4761-6203Wu Songsonghttp://orcid.org/0009-0006-7590-0259 Abstract BackgroundBreast ultrasound is essential for evaluating breast nodules, with Breast Imaging Reporting and Data System (BI-RADS) providing standardized classification. However, interobserver variability among radiologists can affect diagnostic accuracy. Large language models (LLMs) like ChatGPT-4 have shown potential in medical imaging interpretation. This study explores its feasibility in improving BI-RADS classification consistency and malignancy prediction compared to radiologists. ObjectiveThis study aims to evaluate the feasibility of using LLMs, particularly ChatGPT-4, to assess the consistency and diagnostic accuracy of standardized breast ultrasound imaging reports, using pathology as the reference standard. MethodsThis retrospective study analyzed breast nodule ultrasound data from 671 female patients (mean 45.82, SD 9.20 years; range 26‐75 years) who underwent biopsy or surgical excision at our hospital between June 2019 and June 2024. ChatGPT-4 was used to interpret BI-RADS classifications and predict benign versus malignant nodules. The study compared the model’s performance to that of two senior radiologists (≥15 years of experience) and two junior radiologists (<5 years of experience) using key diagnostic metrics, including accuracy, sensitivity, specificity, area under the receiver operating characteristic curve, P ResultsChatGPT-4 achieved an overall BI-RADS classification accuracy of 96.87%, outperforming junior radiologists (617/671, 91.95% and 604/671, 90.01%, PPPP ConclusionsIntegrating ChatGPT-4 into an image-to-text–LLM workflow improves BI-RADS classification accuracy and supports radiologists in breast ultrasound diagnostics. These results demonstrate its potential as a decision-support tool to enhance diagnostic consistency and reduce variability.https://medinform.jmir.org/2025/1/e70924
spellingShingle Su Miaojiao
Liang Xia
Zeng Xian Tao
Hong Zhi Liang
Cheng Sheng
Wu Songsong
Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study
JMIR Medical Informatics
title Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study
title_full Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study
title_fullStr Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study
title_full_unstemmed Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study
title_short Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study
title_sort using a large language model for breast imaging reporting and data system classification and malignancy prediction to enhance breast ultrasound diagnosis retrospective study
url https://medinform.jmir.org/2025/1/e70924
work_keys_str_mv AT sumiaojiao usingalargelanguagemodelforbreastimagingreportinganddatasystemclassificationandmalignancypredictiontoenhancebreastultrasounddiagnosisretrospectivestudy
AT liangxia usingalargelanguagemodelforbreastimagingreportinganddatasystemclassificationandmalignancypredictiontoenhancebreastultrasounddiagnosisretrospectivestudy
AT zengxiantao usingalargelanguagemodelforbreastimagingreportinganddatasystemclassificationandmalignancypredictiontoenhancebreastultrasounddiagnosisretrospectivestudy
AT hongzhiliang usingalargelanguagemodelforbreastimagingreportinganddatasystemclassificationandmalignancypredictiontoenhancebreastultrasounddiagnosisretrospectivestudy
AT chengsheng usingalargelanguagemodelforbreastimagingreportinganddatasystemclassificationandmalignancypredictiontoenhancebreastultrasounddiagnosisretrospectivestudy
AT wusongsong usingalargelanguagemodelforbreastimagingreportinganddatasystemclassificationandmalignancypredictiontoenhancebreastultrasounddiagnosisretrospectivestudy