Share：
Share this content in WeChat
X
[Chinese][PDF]2994
• Special Focus •
Research on automatic classification of breast MRI images based on deep learning
MA Mingming  QIN Naishan  JIANG Yuan  ZHANG Yaofeng  ZHANG Xiaodong  WANG Xiaoying 

Cite this article as: MA M M, QIN N S, JIANG Y, et al. Research on automatic classification of breast MRI images based on deep learning[J]. Chin J Magn Reson Imaging, 2024, 15(1): 55-60. DOI:10.12015/issn.1674-8034.2024.01.009.


[Abstract] Objective To train and verify an automatic image classification model based on deep learning and program.Materials and Methods A total of 862 breast MRI images were collected from the picture archiving and communication system (PACS) system from January 2010 to November 2020 (dataset I). The images were divided into three categories: T2WI, fat-suppressed (FS) T2WI, apparent diffusion coefficient (ADC). A deep learning model of sequence differentiation was trained with the dataset I. Another group of 377 breast MRI images from February 2013 to April 2020 (data set Ⅱ) were collected and divided into three categories: no-contrast (NoC), contrast enhanced early (CEearly), contrast enhanced (CE) according to the phase characteristics of dynamic contrast-enhanced (DCE). A deep learning model of phase differentiation of DCE was trained with the dataset Ⅱ. A third group of 95 breast MRI images (data set Ⅲ) were collected from October 2021 to December 2021 for independent validation of the classification models (different sequences and different phases of DCE). Then the diffusion weighted imaging (DWI) parameters were classified by the program in the data set Ⅲ (DWI-high and DWI-low). Using the classification results of radiologists on images as the gold standard, according to image sequence, enhancement characteristics and parameters on as the gold standard, the confusion matrix was used to evaluate the classification performance of the model.Results In the sequence classification model, the overall prediction accuracy was 92.0%, and the prediction accuracy for each sequence of ADC, T2WI and FS T2WI were 100.0%, 84.9%, and 100.0%, respectively. In the DCE classification model, the overall prediction accuracy was 90.4%, and the prediction accuracy for each sequence of NoC, CEearly, and CE were 89.7%, 39.2%, and 95.7%,respectively. The program's classification of DWI-high and DWI-low was exactly the same as that of the radiologist.Conclusions Using deep learning model and program technology to classify the image sequence, phase and parameters of multi-parameter breast MRI, the output results are highly consistent with the classification results of physicians, which basically meet the clinical needs.
[Keywords] breast tumor；automated classification of images；deep learning；artificial intelligence；magnetic resonance imaging

MA Mingming1   QIN Naishan1   JIANG Yuan1   ZHANG Yaofeng2   ZHANG Xiaodong1   WANG Xiaoying1*  

1 Department of Radiology, Peking University First Hospital, Beijing 100034, China

2 Beijing Smart Tree Medical Technology Co., Ltd., Beijing 100011, China

Corresponding author: WANG X Y, E-mail: wangxiaoying@bjmu.edu.cn

Conflicts of interest   None.

Received  2023-08-06
Accepted  2024-01-09
DOI: 10.12015/issn.1674-8034.2024.01.009
Cite this article as: MA M M, QIN N S, JIANG Y, et al. Research on automatic classification of breast MRI images based on deep learning[J]. Chin J Magn Reson Imaging, 2024, 15(1): 55-60. DOI:10.12015/issn.1674-8034.2024.01.009.

[1]
SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249. DOI: 10.3322/caac.21660.
[2]
LIU X Y, CHEN Y Y, CHENG A S, et al. Conceptualizing problems with symptoms, function, health behavior, health-seeking skills, and financial strain in breast cancer survivors using hierarchical clustering[J]. J Cancer Surviv, 2022, 16(4): 751-759. DOI: 10.1007/s11764-021-01068-w.
[3]
HE J, CHEN W Q, LI N, et al. China guideline for the screening and early detection of female breast cancer(2021, Beijing)[J]. Chin J Oncol, 2021, 43(4): 357-382. DOI: 10.3760/cma.j.cn112152-20210119-00061.
[4]
WAGNER M W, NAMDAR K, BISWAS A, et al. Radiomics, machine learning, and artificial intelligence-what the neuroradiologist needs to know[J]. Neuroradiology, 2021, 63(12): 1957-1967. DOI: 10.1007/s00234-021-02813-9.
[5]
LI Z W, LIU F, YANG W J, et al. A survey of convolutional neural networks: analysis, applications, and prospects[J]. IEEE Trans Neural Netw Learn Syst, 2022, 33(12): 6999-7019. DOI: 10.1109/TNNLS.2021.3084827.
[6]
WANG Y X, TAN H N. Research and application progresses of artificial intelligence in breast cancer imaging[J]. Chin J Magn Reson Imag, 2023, 14(11): 177-182. DOI: 10.12015/issn.1674-8034.2023.11.030.
[7]
CHEN X X, WANG X M, ZHANG K, et al. Recent advances and clinical applications of deep learning in medical image analysis[J/OL]. Med Image Anal, 2022, 79: 102444 [2023-08-05]. https://doi.org/10.1016/j.media.2022.102444. DOI: 10.1016/j.media.2022.102444.
[8]
OUYANG H, LIU G Y, BAI Y P, et al. Clinical application progress of MRI-based radiomics in gliomas[J]. Chin J Magn Reson Imag, 2020, 11(1): 74-76. DOI: 10.12015/issn.1674-8034.2020.01.017.
[9]
ZHAO W W, SUN J, ZHU J Q. Research progress of artificial intelligence in MRI diagnosis of glioma[J]. Chin J Magn Reson Imag, 2021, 12(8): 88-90. DOI: 10.12015/issn.1674-8034.2021.08.019.
[10]
GROSSMANN P, NARAYAN V, CHANG K, et al. Quantitative imaging biomarkers for risk stratification of patients with recurrent glioblastoma treated with bevacizumab[J]. Neuro Oncol, 2017, 19(12): 1688-1697. DOI: 10.1093/neuonc/nox092.
[11]
LI Y H, JIA W D. Research progress of radiomics with artificial intelligence in precise diagnosis and treatment of hepatocellular carcinoma[J]. Chin J Hepatol, 2020, 28(11): 905-909. DOI: 10.3760/cma.j.cn501113-20201022-00573.
[12]
CONTI A, DUGGENTO A, INDOVINA I, et al. Radiomics in breast cancer classification and prediction[J]. Semin Cancer Biol, 2021, 72: 238-250. DOI: 10.1016/j.semcancer.2020.04.002.
[13]
ZHENG X Y, YAO Z, HUANG Y N, et al. Deep learning radiomics can predict axillary lymph node status in early-stage breast cancer[J/OL]. Nat Commun, 2020, 11(1): 1236 [2023-08-05]. https://pubmed.ncbi.nlm.nih.gov/32144248/. DOI: 10.1038/s41467-020-15027-z.
[14]
ZHOU L Q, WU X L, HUANG S Y, et al. Lymph node metastasis prediction from primary breast cancer US images using deep learning[J]. Radiology, 2020, 294(1): 19-28. DOI: 10.1148/radiol.2019190372.
[15]
WANG X Y, XIE T S, LUO J R, et al. Radiomics predicts the prognosis of patients with locally advanced breast cancer by reflecting the heterogeneity of tumor cells and the tumor microenvironment[J/OL]. Breast Cancer Res, 2022, 24(1): 20 [2023-08-05]. https://pubmed.ncbi.nlm.nih.gov/35292076/. DOI: 10.1186/s13058-022-01516-0.
[16]
ZHANG X D, SUN Z N, REN X, et al. Development of a ResNet-based CADx software for classification diagnosis in chest X-ray images[J]. Radiol Pract, 2019, 34(9): 952-957. DOI: 10.13609/j.cnki.1000-0313.2019.09.004.
[17]
BAGHERI M H, ROTH H, KOVACS W, et al. Technical and clinical factors affecting success rate of a deep learning method for pancreas segmentation on CT[J]. Acad Radiol, 2020, 27(5): 689-695. DOI: 10.1016/j.acra.2019.08.014.
[18]
HOU Z H. Research on semantic SLAM based on uncertainty model and relocalization[D].Harbin: Harbin Institute of Technology, 2018. DOI: 10.7666/d.D01588144.
[19]
WU S T, ZHONG S H, LIU Y. Deep residual learning for image steganalysis[J]. Multimed Tools Appl, 2018, 77(9): 10437-10453. DOI: 10.1007/s11042-017-4440-4.
[20]
FANG J H, LI Q B, YU C X, et al. Deep learning for classification of multi? sequence MR images of the prostate[J]. Chin J Radiol, 2019, 53(10): 839-843. DOI: 10.3760/cma.j.issn.1005?1201.2019.10.009.
[21]
SUN Z N, CUI Y P, LIU X, et al. Deep learning for classification of range and phase of abdominal and pelvic CT scanning: a prospective validation study in clinical workflow[J]. Radiol Pract, 2021, 36(4): 551-555. DOI: 10.13609/j.cnki.1000-0313.2021.04.025.
[22]
E T Y, GUO X C, WANG K, et al. The feasibility of training classification model of different phases of upper abdominal DCE-MRI images based on deep learning method[J]. Radiol Pract, 2021, 36(8): 1052-1058. DOI: 10.13609/j.cnki.1000-0313.2021.08.020.
[23]
CHEN S H, MA K, ZHENG Y F. Med3D: transfer learning for 3D medical image analysis[J/OL]. arXiv, 2019 [2023-08-05]. https://arxiv.org/abs/1904.00625. DOI: 10.48550/arXiv.1904.00625.
[24]
STREICH G, VILLALBA M B, CID C, et al. Developing a real-world database for oncology: a descriptive analysis of breast cancer in Argentina[J/OL]. Ecancermedicalscience, 2022, 16: 1435 [2023-08-05]. https://pubmed.ncbi.nlm.nih.gov/36200013/. DOI: 10.3332/ecancer.2022.1435.
[25]
CHOI W J, CHA J H, KIM H H, et al. The accuracy of breast MR imaging for measuring the size of a breast cancer: analysis of the histopathologic factors[J/OL]. Clin Breast Cancer, 2016, 16(6): e145-e152 [2023-08-05]. https://pubmed.ncbi.nlm.nih.gov/27566960/. DOI: 10.1016/j.clbc.2016.07.007.
[26]
ROMINGER M, BERG D, FRAUENFELDER T, et al. Which factors influence MRI-pathology concordance of tumour size measurements in breast cancer?[J]. Eur Radiol, 2016, 26(5): 1457-1465. DOI: 10.1007/s00330-015-3935-5.
[27]
SONG Y Y, HAN X F, BAI N N. Correlation between DCE-MRI imaging features and biological prognostic factors in breast cancer with dynamic contrast-enhanced MRI[J]. Oncol Prog, 2021, 19(15): 1568-1572, 1601. DOI: 10.11877/j.issn.1672-1535.2021.19.15.15.
[28]
DORNELAS E C, KAWASSAKI C S, OLANDOSKI M, et al. A three-sequence dynamic contrast enhanced abbreviated MRI protocol to evaluate response to breast cancer neoadjuvant chemotherapy[J]. Magn Reson Imaging, 2023, 102: 49-54. DOI: 10.1016/j.mri.2023.04.005.
[29]
LI H Y, LIU J L. Diagnostic value of multi-parameter magnetic resonance imaging for benign and malignant breast lesions[J]. Oncol Prog, 2021, 19(14): 1429-1432, 1455. DOI: 10.11877/j.issn.1672-1535.2021.19.14.08.
[30]
WANG C H, WANG G Y, ZHANG Y F, et al. Differentiation of benign and malignant breast lesions using diffusion-weighted imaging with a fractional-order calculus model[J/OL]. Eur J Radiol, 2023, 159: 110646 [2023-08-05]. https://pubmed.ncbi.nlm.nih.gov/36577184/. DOI: 10.1016/j.ejrad.2022.110646.

PREV Differentiation of benign and malignant breast lesions using DWI with a fractional-order calculus model based on SMS technique
NEXT Evaluation of the therapeutic efficacy of acupuncture in children with autism using DTI
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn