Design of a mammography X-ray image classification assistant system adapted to Chinese population

SUN Changjin; TONG Fei; WU Yi; WANG Yuting; LUO Junjie; GONG Yan; QIU Mingguo; QIAO Liang

doi:10.16016/j.2097-0927.202407001

您当前的位置：

首页 >

文章列表页 >

Design of a mammography X-ray image classification assistant system adapted to Chinese population

Biomedical Engineering | 更新时间：2025-01-10

- Design of a mammography X-ray image classification assistant system adapted to Chinese population
- Journal of Army Medical University Vol. 47, Issue 1, Pages: 92-99(2025)
- 作者机构：
  
  1.陆军军医大学（第三军医大学）生物医学工程与影像医学系：，医学图像学教研室，重庆
  2.陆军军医大学（第三军医大学）生物医学工程与影像医学系：，数字医学教研室，重庆
  3.陆军特色医学中心（第三军医大学大坪医院）医学工程科，重庆
- 作者简介：
  
  QIAO Liang， E-mail： bobq310@tmmu.edu.cn
- 基金信息：
  
  the Natural Science Foundation of Chongqing(CSTB2022NSCQ-MSX0173);the Project of Postgraduate Scientific Research Innovation of Chongqing(CYS23766);the Special Project of Science and Technology Innovation Capacity Promotion of Army Medical University(2023XJS13)
- DOI：10.16016/j.2097-0927.202407001
  CLC： R319;R322.66;R814.414
- Published：15 January 2025，
  
  Received：01 July 2024，
  
  Revised：30 October 2024，
- 稿件说明：
移动端阅览
孙常晋,童飞,吴毅,等.适应中国人群的乳腺钼靶X线图像分类辅助系统设计[J].陆军军医大学学报,2025,47(01):92-99.

SUN Changjin,TONG Fei,WU Yi,et al.Design of a mammography X-ray image classification assistant system adapted to Chinese population[J].J Army Med Univ,2025,47(1):92-99.
孙常晋,童飞,吴毅,等.适应中国人群的乳腺钼靶X线图像分类辅助系统设计[J].陆军军医大学学报,2025,47(01):92-99. DOI：10.16016/j.2097-0927.202407001.

SUN Changjin,TONG Fei,WU Yi,et al.Design of a mammography X-ray image classification assistant system adapted to Chinese population[J].J Army Med Univ,2025,47(1):92-99. DOI：10.16016/j.2097-0927.202407001.

摘要

目的

构建适应中国人群的乳腺钼靶图像分类辅助系统，探讨人工智能技术在国内辅助乳腺癌早期筛查的潜力。

方法

为复现当前文献中的主流深度学习方法，利用乳腺X线筛查数字数据库子集（curated breast imaging subset of digital database for screening mammography，CBIS-DDSM）、乳房X线图像分析学会数据库（mammographic image analysis society database，MIAS）等国际公开数据集分别进行模型训练，并在华教科技有限公司提供的中国人群乳腺钼靶图像数据集（Chinese breast mammography dataset，CBMD）上进行测试和模型性能比较；针对中国人群数据在公开数据集训练模型性能测试不理想的问题，结合中国人群数据特点，提出基于滑动窗口调窗机制的优化策略，设计二阶段迁移学习方法，以提升模型的整体性能，并进行系统研发。

结果

使用滑动窗口调窗机制及二阶段迁移学习后的CBMD训练模型，以中国人群数据集为测试集，其准确度从默认窗口下公开数据集训练模型的0.50提升至0.80，精度从0.54提升至0.82，灵敏度从0.52提升至0.80，F1值从0.52提升至0.80，AUC值从0.51提升至0.89。

结论

本研究引入滑动窗口调窗机制和二阶段迁移学习策略，显著提升了乳腺钼靶图像分类模型在中国人群数据集上的性能，初步达到适应中国人群的乳腺钼靶图像辅助分类的目的。

Abstract

Objective‍　‍To construct a mammography image classification assistant system suitable for Chinese population， and explore the potential of artificial intelligence technology to assist early screening of breast cancer in China. Methods‍　‍Curated breast imaging subset of digital database for screening mammography （CBIS-DDSM）， Mammographic image analysis society database （MIAS） and other international open datasets were used to conduct model training respectively in order to reproduce the mainstream in-depth learning methods in the current literature. The model was also tested on the Chinese breast mammography database （CBMD） provided by Huajiao Technology Co.， Ltd， and the performance was compared. Aiming at the problem that the Chinese population data are not ideal in the performance test of the open dataset training model， an optimization strategy based on the sliding window adjustment mechanism was implemented in combination with the characteristics of Chinese population data. Then a two-stage migration learning method was designed to improve the overall performance of the model， and then development of our system was carried out. Results‍　‍With the sliding window adjustment mechanism and the CBMD training model after two-stage transfer learning， the accuracy of our developed system was improved from 0.50 of the open datasets to 0.80， precision from 0.54 to 0.82， sensitivity from 0.52 to 0.80， F1 value from 0.52 to 0.80， and AUC value from 0.51 to 0.89 based on the Chinese population dataset as the test set. Conclusion‍　‍Through the introduction of sliding window adjustment mechanism and two-stage migration learning strategy， the performance of the breast molybdenum target image classification model has been significantly improved in the Chinese population dataset， and our system primarily achieves the purpose of assisting the classification of breast molybdenum target images for the Chinese population.

关键词

乳腺钼靶深度学习X射线计算机辅助

Keywords

mammographydeep learningX-Raycomputer assistance

references

高艳多，阎炯，赵胜，等. 1990—2019年中国女性乳腺癌发病和死亡趋势的年龄-时期-队列模型分析［J］. 中国预防医学杂志， 2022， 23（12）： 909-916. DOI： 10.16506/j.1009-6639.2022.12.005http://dx.doi.org/10.16506/j.1009-6639.2022.12.005.

GAO Y D， YAN J， ZHAO S， et al. Trends in incidence and mortality of female breast cancer in China from 1990 to 2019 using age-period-cohort analysis model［J］. Chin Prev Med， 2022， 23（12）： 909-916. DOI： 10.16506/j.1009-6639.2022.12.005http://dx.doi.org/10.16506/j.1009-6639.2022.12.005.

XU Y Y， GONG M Y， WANG Y， et al. Global trends and forecasts of breast cancer incidence and deaths［J］. Sci Data， 2023， 10（1）： 334. DOI： 10.1038/s41597-023-02253-5http://dx.doi.org/10.1038/s41597-023-02253-5.

郭亚梅，蒋凌燕，段正凡. 乳腺钼靶X线摄影诊断乳腺癌的应用价值分析［J］. 国际全科医学， 2023.4（1）：83-85. DOI：10.37155/2717-5669-0401-28http://dx.doi.org/10.37155/2717-5669-0401-28.

GAO Y M， JIANG L Y，DUAN Z F， et al. The value of mammography in the diagnosis of breast cancer［J］. Int J Gen Med， 2023.4（1）：83-85. DOI：10.37155/2717-5669-0401-28http://dx.doi.org/10.37155/2717-5669-0401-28.

LIAO J W， GUI Y， LI Z L， et al. Artificial intelligence-assisted ultrasound image analysis to discriminate early breast cancer in Chinese population： a retrospective， multicentre， cohort study［J］. EClinicalMedicine， 2023， 60： 102001. DOI： 10.1016/j.eclinm.2023.102001http://dx.doi.org/10.1016/j.eclinm.2023.102001.

FARAH L， DAVAZE-SCHNEIDER J， MARTIN T， et al. Are current clinical studies on artificial intelligence-based medical devices comprehensive enough to support a full health technology assessment： A systematic review［J］. Artif Intell Med， 2023， 140： 102547. DOI： 10.1016/j.artmed. 2023.102547http://dx.doi.org/10.1016/j.artmed.2023.102547.

MRAČKO A， VANOVČANOVÁ L， CIMRÁK I. Mammography datasets for neural networks-survey［J］. J Imaging， 2023， 9（5）： 95. DOI： 10.3390/jimaging9050095http://dx.doi.org/10.3390/jimaging9050095.

KIM C M， CHUNG K， PARK R C. Anomaly detection model of mammography using YOLOv4-based histogram［J］. Pers Ubiquitous Comput， 2023， 27（3）： 1233-1244. DOI： 10.1007/s00779-021-01598-1http://dx.doi.org/10.1007/s00779-021-01598-1.

LEE S E， HAN K， KIM E K. Application of artificial intelligence-based computer-assisted diagnosis on synthetic mammograms from breast tomosynthesis： comparison with digital mammograms［J］. Eur Radiol， 2021， 31（9）： 6929-6937. DOI： 10.1007/s00330-021-07796-yhttp://dx.doi.org/10.1007/s00330-021-07796-y.

WANG Y Q， TANG L， CHEN P P， et al. The role of a deep learning-based computer-aided diagnosis system and elastography in reducing unnecessary breast lesion biopsies［J］. Clin Breast Cancer， 2023， 23（3）： e112-e121. DOI： 10.1016/j.clbc.2022.12.016http://dx.doi.org/10.1016/j.clbc.2022.12.016.

SIVIENGPHANOM S， GANDOMKAR Z， LEWIS S J， et al. Global radiomic features from mammography for predicting difficult-to-interpret normal cases［J］. J Digit Imag， 2023， 36（4）： 1541-1552. DOI： 10.1007/s10278-023-00836-7http://dx.doi.org/10.1007/s10278-023-00836-7.

RABIDAS R， MIDYA A， SADHU A， et al. Benign-malignant mass classification in mammogram using edge weighted local texture features［C］. Med Ima 2016： Comput Aided Diag-SPIE， 2016， 9785（1x）： 1-6. DOI： 10.1117/12.2216767http://dx.doi.org/10.1117/12.2216767.

REHMAN S U， KHAN M A， MASOOD A， et al. BRMI-net： deep learning features and flower pollination-controlled regula falsi-based feature selection framework for breast cancer recognition in mammography images［J］. Diagnostics， 2023， 13（9）： 1618. DOI： 10.3390/diagnostics 13091618http://dx.doi.org/10.3390/diagnostics13091618.

LEE R S， GIMENEZ F， HOOGI A， et al. A curated mammography data set for use in computer-aided detection and diagnosis research［J］. Sci Data， 2017， 4： 170177. DOI： 10.1038/sdata.2017.177http://dx.doi.org/10.1038/sdata.2017.177.

SHEN L， MARGOLIES L R， ROTHSTEIN J H， et al. Deep learning to improve breast cancer detection on screening mammography［J］. Sci Rep， 2019， 9： 12495. DOI： 10.1038/s41598-019-48995-4http://dx.doi.org/10.1038/s41598-019-48995-4.

MOREIRA I C， AMARAL I， DOMINGUES I， et al. INbreast： toward a full-field digital mammographic database［J］. Acad Radiol， 2012， 19（2）： 236-248. DOI： 10.1016/j.acra.2011.09.014http://dx.doi.org/10.1016/j.acra.2011.09.014.

HEATH M， BOWYER K， KOPANS D， et al. Current status of the digital database for screening mammography［M］. Dordrecht： Springer Netherlands， 1998： 457-460. DOI： 10.1007/978-94-011-5318-8_75http://dx.doi.org/10.1007/978-94-011-5318-8_75.

SUCKLING J， PARKER J， ASTLEY S， et al. The Mammographic Image Analysis Society digital mammogram database［C］. Int Congr Ser， 1994：375-378. https：//www.repository.cam.ac.uk/handle/1810/250394https://www.repository.cam.ac.uk/handle/1810/250394.

彭芳芳，沈坤炜. 中西方女性乳腺密度与乳腺癌发病关系的研究进展［J］. 中国癌症防治杂志， 2020， 12（4）： 469-474. DOI： 10.3969/j.issn.1674-5671.2020.04.19http://dx.doi.org/10.3969/j.issn.1674-5671.2020.04.19.

PENG F F， SHEN K W. Research progress on the relationship between breast density and breast cancer in Chinese and western women［J］. Chin J Oncol Prev Treat， 2020， 12（4）： 469-474. DOI： 10.3969/j.issn.1674-5671.2020.04.19http://dx.doi.org/10.3969/j.issn.1674-5671.2020.04.19.

董誉. 乳腺钼靶图像中的肿块良恶性分类［D］. 武汉：华中科技大学， 2020： 8-36. DOI：10.27157/d.cnki.ghzku.2020.001367http://dx.doi.org/10.27157/d.cnki.ghzku.2020.001367.

DONG Y. Classification of benign and malignant masses in mammography［D］. Wuhan： Huazhong University of Science and Technology， 2020： 8-36. DOI：10.27157/d.cnki.ghzku.2020.001367http://dx.doi.org/10.27157/d.cnki.ghzku.2020.001367.

薛珂，徐辉，岳保荣，等. 我国女性乳腺体积密度及相关乳房特征定量研究［J］. 中华放射医学与防护杂志， 2024， 44（4）： 278-285. DOI： 10.3760/cma.j.cn112271-20240205-00045http://dx.doi.org/10.3760/cma.j.cn112271-20240205-00045.

XUE K， XU H， YUE B R， et al. Quantitative study on female breast bulk density and related breast characteristics in China［J］. Chin J Radiol Med Prot， 2024， 44（4）： 278-285. DOI： 10.3760/cma.j.cn112271-20240205-00045http://dx.doi.org/10.3760/cma.j.cn112271-20240205-00045.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Construction and evaluation of an intelligent diagnostic model based on enhanced CT images and Swin Transformer network for T staging of esophageal cancer

Establishment of a chest-CT-image-based classification deep learning model for pneumonia prediction

Intelligent segmentation of prostate zones in MR images based on Vgg16-Unet

Prostate cancer T-stage intelligent diagnosis based on MRI images and deep learning

An anemia screening tool based on deep learning with conjunctiva images

Related Author

CUI Huilin

WU Yi

NIAN Yongjian

CAO Ximei

MA Jinfeng

GUO Mei

YAO Jie

WU Wei

Related Institution

陆军军医大学（第三军医大学）第一附属医院：胸外科，心外科4山西省肿瘤医院普外科

陆军军医大学（第三军医大学）生物医学工程与影像医学系数字医学教研室

山西医科大学基础医学院组织胚胎学教研室

重庆知见生命科技有限公司，重庆大学附属三峡医院放射科

陆军军医大学（第三军医大学）第一附属医院放射科

⁰