Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study

Abstract Background The clinical application of artificial intelligence (AI) models based on breast ultrasound static images has been hindered in real-world workflows due to operator-dependence of standardized image acquisition and incomplete view of breast lesions on static images. To better exploi...

Full description

Saved in:
Bibliographic Details
Main Authors: Jie Han, Yuanjing Gao, Ling Huo, Dong Wang, Xiaozheng Xie, Rui Zhang, Mengsu Xiao, Nan Zhang, Meng Lei, Quanlin Wu, Lu Ma, Chao Sun, Xinyi Wang, Lei Liu, Shuzhen Cheng, Binghui Tang, Liwei Wang, Qingli Zhu, Yong Wang
Format: Article
Language:English
Published: BMC 2025-06-01
Series:Cancer Imaging
Subjects:
Online Access:https://doi.org/10.1186/s40644-025-00892-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850111693089144832
author Jie Han
Yuanjing Gao
Ling Huo
Dong Wang
Xiaozheng Xie
Rui Zhang
Mengsu Xiao
Nan Zhang
Meng Lei
Quanlin Wu
Lu Ma
Chao Sun
Xinyi Wang
Lei Liu
Shuzhen Cheng
Binghui Tang
Liwei Wang
Qingli Zhu
Yong Wang
author_facet Jie Han
Yuanjing Gao
Ling Huo
Dong Wang
Xiaozheng Xie
Rui Zhang
Mengsu Xiao
Nan Zhang
Meng Lei
Quanlin Wu
Lu Ma
Chao Sun
Xinyi Wang
Lei Liu
Shuzhen Cheng
Binghui Tang
Liwei Wang
Qingli Zhu
Yong Wang
author_sort Jie Han
collection DOAJ
description Abstract Background The clinical application of artificial intelligence (AI) models based on breast ultrasound static images has been hindered in real-world workflows due to operator-dependence of standardized image acquisition and incomplete view of breast lesions on static images. To better exploit the real-time advantages of ultrasound and more conducive to clinical application, we proposed a whole-lesion-aware network based on freehand ultrasound video (WAUVE) scanning in an arbitrary direction for predicting overall breast cancer risk score. Methods The WAUVE was developed using 2912 videos (2912 lesions) of 2771 patients retrospectively collected from May 2020 to August 2022 in two hospitals. We compared the diagnostic performance of WAUVE with static 2D-ResNet50 and dynamic TimeSformer models in the internal validation set. Subsequently, a dataset comprising 190 videos (190 lesions) from 175 patients prospectively collected from December 2022 to April 2023 in two other hospitals, was used as an independent external validation set. A reader study was conducted by four experienced radiologists on the external validation set. We compared the diagnostic performance of WAUVE with the four experienced radiologists and evaluated the auxiliary value of model for radiologists. Results The WAUVE demonstrated superior performance compared to the 2D-ResNet50 model, while similar to the TimeSformer model. In the external validation set, WAUVE achieved an area under the receiver operating characteristic curve (AUC) of 0.8998 (95% CI = 0.8529–0.9439), and showed a comparable diagnostic performance to that of four experienced radiologists in terms of sensitivity (97.39% vs. 98.48%, p = 0.36), specificity (49.33% vs. 50.00%, p = 0.92), and accuracy (78.42% vs.79.34%, p = 0.60). With the WAUVE model assistance, the average specificity of four experienced radiologists was improved by 6.67%, and higher consistency was achieved (from 0.807 to 0.838). Conclusion The WAUVE based on non-standardized ultrasound scanning demonstrated excellent performance in breast cancer assessment which yielded outcomes similar to those of experienced radiologists, indicating the clinical application of the WAUVE model promising.
format Article
id doaj-art-59d218f0ce2447a1a5c97352cbd2f5c5
institution OA Journals
issn 1470-7330
language English
publishDate 2025-06-01
publisher BMC
record_format Article
series Cancer Imaging
spelling doaj-art-59d218f0ce2447a1a5c97352cbd2f5c52025-08-20T02:37:35ZengBMCCancer Imaging1470-73302025-06-0125111710.1186/s40644-025-00892-yWhole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter studyJie Han0Yuanjing Gao1Ling Huo2Dong Wang3Xiaozheng Xie4Rui Zhang5Mengsu Xiao6Nan Zhang7Meng Lei8Quanlin Wu9Lu Ma10Chao Sun11Xinyi Wang12Lei Liu13Shuzhen Cheng14Binghui Tang15Liwei Wang16Qingli Zhu17Yong Wang18Department of Ultrasound, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeKey Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Breast Center, Peking University Cancer Hospital & InstituteYizhun Medical AI Co., Ltd CNUniversity of Science and Technology BeijingDepartment of Ultrasound, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeKey Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Breast Center, Peking University Cancer Hospital & InstituteCenter for Data Science, Peking UniversityCenter for Data Science, Peking UniversityYizhun Medical AI Co., Ltd CNDepartment of Ultrasound, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeKey Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Breast Center, Peking University Cancer Hospital & InstituteDepartment of Ultrasound, Nanchang People’s HospitalDepartment of Ultrasound, Nanchang People’s HospitalDepartment of Ultrasound, Nanchang People’s HospitalCenter for Data Science, Peking UniversityDepartment of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Ultrasound, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeAbstract Background The clinical application of artificial intelligence (AI) models based on breast ultrasound static images has been hindered in real-world workflows due to operator-dependence of standardized image acquisition and incomplete view of breast lesions on static images. To better exploit the real-time advantages of ultrasound and more conducive to clinical application, we proposed a whole-lesion-aware network based on freehand ultrasound video (WAUVE) scanning in an arbitrary direction for predicting overall breast cancer risk score. Methods The WAUVE was developed using 2912 videos (2912 lesions) of 2771 patients retrospectively collected from May 2020 to August 2022 in two hospitals. We compared the diagnostic performance of WAUVE with static 2D-ResNet50 and dynamic TimeSformer models in the internal validation set. Subsequently, a dataset comprising 190 videos (190 lesions) from 175 patients prospectively collected from December 2022 to April 2023 in two other hospitals, was used as an independent external validation set. A reader study was conducted by four experienced radiologists on the external validation set. We compared the diagnostic performance of WAUVE with the four experienced radiologists and evaluated the auxiliary value of model for radiologists. Results The WAUVE demonstrated superior performance compared to the 2D-ResNet50 model, while similar to the TimeSformer model. In the external validation set, WAUVE achieved an area under the receiver operating characteristic curve (AUC) of 0.8998 (95% CI = 0.8529–0.9439), and showed a comparable diagnostic performance to that of four experienced radiologists in terms of sensitivity (97.39% vs. 98.48%, p = 0.36), specificity (49.33% vs. 50.00%, p = 0.92), and accuracy (78.42% vs.79.34%, p = 0.60). With the WAUVE model assistance, the average specificity of four experienced radiologists was improved by 6.67%, and higher consistency was achieved (from 0.807 to 0.838). Conclusion The WAUVE based on non-standardized ultrasound scanning demonstrated excellent performance in breast cancer assessment which yielded outcomes similar to those of experienced radiologists, indicating the clinical application of the WAUVE model promising.https://doi.org/10.1186/s40644-025-00892-yBreast neoplasmsUltrasonographyVideoDiagnosisDeep learningArtificial intelligence
spellingShingle Jie Han
Yuanjing Gao
Ling Huo
Dong Wang
Xiaozheng Xie
Rui Zhang
Mengsu Xiao
Nan Zhang
Meng Lei
Quanlin Wu
Lu Ma
Chao Sun
Xinyi Wang
Lei Liu
Shuzhen Cheng
Binghui Tang
Liwei Wang
Qingli Zhu
Yong Wang
Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study
Cancer Imaging
Breast neoplasms
Ultrasonography
Video
Diagnosis
Deep learning
Artificial intelligence
title Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study
title_full Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study
title_fullStr Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study
title_full_unstemmed Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study
title_short Whole-lesion-aware network based on freehand ultrasound video for breast cancer assessment: a prospective multicenter study
title_sort whole lesion aware network based on freehand ultrasound video for breast cancer assessment a prospective multicenter study
topic Breast neoplasms
Ultrasonography
Video
Diagnosis
Deep learning
Artificial intelligence
url https://doi.org/10.1186/s40644-025-00892-y
work_keys_str_mv AT jiehan wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT yuanjinggao wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT linghuo wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT dongwang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT xiaozhengxie wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT ruizhang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT mengsuxiao wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT nanzhang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT menglei wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT quanlinwu wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT luma wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT chaosun wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT xinyiwang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT leiliu wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT shuzhencheng wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT binghuitang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT liweiwang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT qinglizhu wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy
AT yongwang wholelesionawarenetworkbasedonfreehandultrasoundvideoforbreastcancerassessmentaprospectivemulticenterstudy