|Table of Contents|

Computer aided diagnosis of breast lesions based on images of breast DBT

Journal Of Modern Oncology[ISSN:1672-4992/CN:61-1415/R]

Issue:
2022 01
Page:
120-125
Research Field:
Publishing date:

Info

Title:
Computer aided diagnosis of breast lesions based on images of breast DBT
Author(s):
AI Hua1; NIU Shuxian1; CAO Yan1; DONG Yue2; YU Tao2; LUO Yahong2; JIANG Xiran1
1.Department of Biomedical Engineering,China Medical University,Liaoning Shenyang 110122，China;2.Department of Medical Image,Cancer Hospital of China Medical University,Liaoning Cancer Hospital & Institute,Liaoning Shenyang 110042,China.
Keywords:
breast DBT; computer aided diagnosis; radiomics; deep learning
PACS:
R737.9
DOI:
10.3969/j.issn.1672-4992.2022.01.026
Abstract:
Objective:To evaluate the role of the computer aided diagnosis algorithms for digital breast tomosynthesis(DBT) imaging in differentiating benign and maligant breast lesions.Methods:190 patients with early breast cancer screening(February 2018 to February 2020) in Liaoning Provincial Cancer Hospital Affiliated to China Medical University were enrolled.Handcrafted features and deep learning neural network features were extracted,selected and fused from DBT images.Logistic regression(LR) classification models were established.A nomogram model for predicting the benign and maligant breast lesions were constructed and evaluated through ROC curve,calibration curve and DCA curve.Results:Three features were selected from the fusion features.The fusion feature model performed better.Unsupervised cluster analysis and box plot showed that the selected features had good discrimination performance.The AUC values of the constructed nomogram based on the fusion features were 0.985(95%CI 0.956~1.000,SEN=0.970,SPE=0.929) and 0.984(95%CI 0.956~1.000,SEN=0.909,SPE=0.931) in the training and test cohorts,respectively.Decision curve analysis(DCA) showed that our nomogram had good clinical application value.Conclusion:Building computer model based on the feature fusion method from radiomics and deep learning could assist clinicians to improve the ability of differentiating benign and malignant breast lesions on DBT images.

References:

[1] MCGUIRE A,BROWN JAL,MALONE C,et al.Effects of age on the detection and management of breast cancer ［J］.Cancers,2015,7(2):908-929.
[2] CHEN WQ,ZHENG RS,BAADE PD,et al.Cancer statistics in China,2015 ［J］.CA:A Cancer Journal for Clinicians,2016,66(2):115-132.
[3] ZHAO H,ZHENG SS,JIN J,et al.Value of mammography X-ray images in diagnosing breast diseases ［J］.Chinese Journal of Medical Imaging,2012,20(7):511-514.
[4] WANG XY,LIU Y.Research progress of multimodal imaging in breast cancer research ［J］.Shanxi Med J,2020,49(9):1100-1104.
[5] LAMBIN P,RIOS-VELAZQUEZ E,LEIJENAAR R,et al.Radiomics:extracting more information from medical images using advanced feature analysis ［J］.European Journal of Cancer,2012,48(4):441-446.
[6] LECUN Y,BENGIO Y,HINTON G.Deep learning ［J］.Nature,2015,521(7553):436-444.
[7] ZHOU FY,JIN LP,DONG J.Review of convolutional neural network ［J］.Chinese Journal of Computers,2017,40(6):1229-1251.
[8] ZEILER MD,FERGUS R.Visualizing and understanding convolutional networks［J］.Lecture Notes in Computer Science,2014,8689(1):818-833.
[9] PAUL R,HAWKINS SH,SCHABATH MB,et al.Predicting malignant nodules by fusing deep features with classical radiomics features ［J］.J Med Imaging(Bellingham),2018,5(1):011021.
[10] HUA WQ,XIAO TH,JIANG XR,et al.Lymph-vascular space invasion prediction in cervical cancer:Exploring radiomics and deep learning multilevel features of tumor and peritumor tissue on multiparametric MRI ［J］.Biomedical Signal Processing and Control,2020,58:101869.1-101869.7.
[11] VAN GRIETHUYSEN JJM,FEDOROV A,PARMAR C,et al.Computational radiomics system to decode the radiographic phenotype ［J］.Cancer Research,2017,77(21):e104-e107.
[12] KETKAR N.Deep learning with python:A hands-on introduction ［M］.Berkeley,CA:Apress,2017:97-111.
[13] KIM J,LEE JK,LEE KM.Accurate image super-resolution using very deep convolutional networks［C］.IEEE:IEEE Conference on Computer Vision and Pattern Recognition,2016:1646-1654.
[14] AFSHAR P,MOHAMMADI A,PLATANIOTIS KN,et al.From handcrafted to deep-learning-based cancer radiomics:Challenges and opportunities ［J］.IEEE Signal Processing Magazine,2019,36(4):132-160.
[15] ANTROPOVA N,HUYNH BQ,GIGER ML.A deep feature fusion methodology for breast cancer diagnosis demonstrated on three imaging modality datasets ［J］.Medical Physics,2017,44(10):5162-5171.
[16] SAUERBREI W,ROYSTON P,BINDER H.Selection of important variables and determination of functional form for continuous predictors in multivariable model building ［J］.Statistics in Medicine,2007,26(30):5512-5528.
[17] JIAN YY.Clinical significance of early screening for breast cancer ［J］.Women's Health Research,2016,1(19):168-169.
[18] TAGLIAFICO AS,VALDORA F,MARISCOTTI G,et al.An exploratory radiomics analysis on digital breast tomosynthesis in women with mammographically negative dense breasts ［J］.Breast,2018,40:92-96.
[19] FUSCO R,VALLONE P,FILICE S,et al.Radiomic features analysis by digital breast tomosynthesis and contrast enhanced dual-energy mammography to detect malignant breast lesions ［J］.Biomedical Signal Processing and Control,2019,53:101568.1-101568.13.
[20] HE ZY,YAO GH.Advanced technique of full-field digital mammography:contrast-enhanced dual-energy digital mammography ［J］.Oncoradiology,2013,22(2):129-131.
[21] LAO JW,CHEN YS,LI ZC,et al.A deep learning-based radiomics model for prediction of survival in glioblastoma multiforme ［J］.Scientific Reports,2017,7(1):10353.
[22] FU J,ZHONG XR,LI N,et al.Deep learning based radiomic features for improving neoadjuvant chemoradiation resonse prediction in locally advanced rectal cancer ［J］.Physics in Medicine & Biology,2020,65(7):075001.

Memo

Memo:
国家重点研发计划项目（编号：2016YFC1303002）；中国公益研究专项基金（编号：201402020）；National Natural Science Foundation of China(No.81872363）；国家自然科学基金（编号：81872363）；中青年科技创新人才支持计划项目（编号：RC170497）；辽宁省肿瘤医院-大连理工大学“医-工交叉研究基金”项目（编号：LD202029）；中国医科大学健康医疗大数据研究课题（编号：HMB201903101）；辽宁省医工结合创新驱动发展策略
Last Update: 2021-12-02