[1]周靖宇,单慧明,成官迅.基于支持向量机的全乳纹理分析预测乳腺癌新辅助化疗疗效的价值[J].医学信息,2022,35(19):60-64.[doi:10.3969/j.issn.1006-1959.2022.19.016]
　ZHOU Jing-yu,SHAN Hui-ming,CHENG Guan-xun.Prediction of Neoadjuvant Chemotherapy Outcomes of Breast Cancer Based on Whole Breast Texture with Support Vector Machine[J].Journal of Medical Information,2022,35(19):60-64.[doi:10.3969/j.issn.1006-1959.2022.19.016]
点击复制

基于支持向量机的全乳纹理分析预测乳腺癌新辅助化疗疗效的价值()
分享到：

医学信息[ISSN:1006-1959/CN:61-1278/R]

卷:
35卷
期数:
2022年19期
页码:
60-64
栏目:
论著
出版日期:
2022-10-01

文章信息/Info

Title:
Prediction of Neoadjuvant Chemotherapy Outcomes of Breast Cancer Based on Whole Breast Texture with Support Vector Machine
文章编号:
1006-1959（2022）19-0060-05
作者:
周靖宇; 单慧明; 成官迅
（北京大学深圳医院医学影像科，广东 深圳 518000）
Author(s):
ZHOU Jing-yu; SHAN Hui-ming; CHENG Guan-xun
(Department of Radiology,Peking University Shenzhen Hospital,Shenzhen 518000,Guangdong,China)
关键词:
乳腺癌; 新辅助化疗; 纹理特征; 支持向量机
Keywords:
Breast cancer; Neoadjuvant chemotherapy; Texture features; Support vector machine
分类号:
R737.9
DOI:
10.3969/j.issn.1006-1959.2022.19.016
文献标志码:
A
摘要:
目的 探讨基于支持向量机的全乳纹理分析预测局部进展晚期乳腺癌新辅助化疗疗效（NAC）的应用价值。方法 纳入2013年9月-2019年9月北京大学深圳医院收治的乳腺癌患者152例，均行3次动态增强磁共振检查（治疗前标记为E1、NAC 2个周期后标记为E2、NAC 4个周期后标记为E3），完成8个周期标准NAC后接受根治性乳房切除。根据病理Miller-Payne评级分为组织学显著反应组（MHR）与组织学非显著反应组（NMHR）。从动态增强磁共振图像中提取基于灰度直方的均值、方差、一致性、偏度、峰度以及灰度共生矩阵的能量、熵、惯性矩、相关性、逆差距的患侧全乳纹理特征，在R project构建径向基函数内核的支持向量机模型预测乳腺癌NAC疗效，其中训练病例114例，测试病例38例，分析模型预测准确度及受试者工作特征曲线、曲线下面积。结果 152例病例中MHR组54例，NMHR组98例。两组不同年龄、月经状态、家族史、肿瘤分期、分子分型及化疗方案比较，差异无统计学意义（P＞0.05）；NMHR组逆差距（E1、E2）、灰度一致性（E2、E3）及比值（E3/E1）、灰度偏度比值（E2/E1、E3/E1）和峰度比值（E3/E1）高于MHR组，而惯性矩（E2）低于MHR组，差异有统计学意义（P＜0.05）；逆差距（E1、E2）、惯性矩（E2）、灰度一致性（E2、E3）及比值（E3/E1）、灰度偏度比值（E2/E1、E3/E1）和峰度比值（E3/E1）单独预测NAC疗效的AUC值介于0.5797~0.6419，诊断效能较低。E1+E2、E1+E3、E1+E2+E3三种组合模型中，E1+E2+E3组合的预测效能最高，在训练组准确度为77.20%，曲线下面积为0.90，在测试组准确度为71.10%，曲线下面积为0.76。结论 基于支持向量机的全乳纹理可以预测乳腺癌新辅助化疗疗效。
Abstract:
Objective To explore the application value of whole breast texture analysis based on support vector machine in predicting the efficacy of neoadjuvant chemotherapy (NAC) in locally advanced breast cancer.Methods A total of 152 patients with breast cancer admitted to Peking University Shenzhen Hospital from September 2013 to September 2019 were enrolled. All patients underwent three dynamic contrast-enhanced magnetic resonance examinations (before treatment marked as E1, after 2 cycles of NAC marked as E2, after 4 cycles of NAC marked as E3), and underwent radical mastectomy after 8 cycles of standard NAC. According to the pathological Miller-Payne rating, the patients were divided into major histological response group and non-major histological response group. The mean, variance, consistency, skewness, kurtosis based on the gray square and the energy, entropy, moment of inertia, correlation and inverse gap of the gray level co-occurrence matrix were extracted from the dynamic enhanced magnetic resonance image. The texture features of the affected side of the whole breast were extracted. The support vector machine model with radial basis function kernel was constructed in R project to predict the NAC efficacy of breast cancer, including 114 training cases and 38 test cases. The prediction accuracy of the model and the receiver operating characteristic curve and area under the curve were analyzed.Results Among 152 patients, there were 54 cases of MHR group and 98 cases of NMHR group. There was no significant difference in age, menstrual status, family history, tumor stage, molecular classification and chemotherapy regimen between the two groups (P>0.05). The inverse difference (E1, E2), gray consistency (E2, E3) and ratio (E3/E1), gray skewness ratio (E2/E1, E3/E1) and kurtosis ratio (E3/E1) in the NMHR group were higher than those in the MHR group, while the moment of inertia (E2) was lower than that in the MHR group, the difference was statistically significant (P<0.05). The AUC values of inverse difference (E1, E2), moment of inertia (E2), gray consistency (E2, E3) and ratio (E3/E1), gray skewness ratio (E2/E1, E3/E1) and kurtosis ratio (E3/E1) alone in predicting NAC efficacy ranged from 0.5797 to 0.6419, with low diagnostic efficiency. Among the three combined models of E1+E2, E1+E3 and E1+E2+E3, the combination of E1+E2+E3 had the highest prediction efficiency, with an accuracy of 77.20% in the training group and an area under the curve of 0.90, the accuracy of the test group was 71.10% and the area under the curve was 0.76.Conclusion Whole breast texture based on support vector machine can predict the efficacy of neoadjuvant chemotherapy for breast cancer.

参考文献/References:

[1]Granzier RWY,van Nijnatten TJA,Woodruff HC,et al.Exploring Breast Cancer Response Prediction to Neoadjuvant Systemic Therapy Using MRI-based Radiomics: A Systematic Review[J].Eur J Radiol,2019,121:108736.[2]孙赛花,周纯武,赵莉芸,等.动态增强磁共振成像纹理分析预测乳腺癌新辅助化疗疗效[J].中华肿瘤杂志,2017,39(5):344-349.[3]Carin AJ,Molière S,Gabriele V,et al.Relevance of Breast MRI in Determining The Size and Focality of Invasive Breast Cancer Treated by Mastectomy: A Prospective Study[J].World J Surg Oncol,2017,15(1):128.[4]Henderson SA,Muhammad Gowdh N,Purdie CA,et al.Breast Cancer: Influence of Tumour Volume Estimation Method at MRI on Prediction of Pathological Response to Neoadjuvant Chemotherapy[J].Br J Radiol,2018,91(1087):20180123.[5]Kim JH,Ko ES,Lim Y,et al.Breast Cancer Heterogeneity: MR Imaging Texture Analysis and Survival Outcomes[J].Radiology,2017,282(3):665-675.[6]Scalco E,Rizzo G.Texture Analysis of Medical Images for Radiotherapy Applications[J].Br J Radiol,2017,90(1070):20160642.[7]Gastounioti A,Conant EF,Kontos D.Beyond Breast Density: AReview on The Advancing Role of Parenchymal Texture Analysis in Breast Cancer Risk Assessment[J].Breast Cancer Res,2016,18(1):91.[8]Reig B,Heacock L,Geras KJ,et al.Machine Learning in Breast MRI[J].J Magn Reson Imaging,2020,52(4):998-1018.[9]Tsougos I,Vamvakas A,Kappas C,et al.Application of Radiomics and Decision Support Systems for Breast MR Differential Diagnosis[J].Comput Math Methods Med,2018,2018:7417126.[10]Pinker K,Chin J,Melsaether AN,et al.Precision Medicine and Radiogenomics in Breast Cancer: New Approaches toward Diagnosis and Treatment[J].Radiology,2018,287(3):732-747.[11]单慧明,周靖宇,谢婷婷,等.MRI影像学特征预测乳腺癌新辅助化疗疗效的可行性[J].中国医学影像学杂志,2019,27(12):905-909,913.[12]Pirooznia M,Deng Y.SVM Classifier - AComprehensiveJava Interface for Support Vector Machine Classification of Microarray Data[J].BMC Bioinformatics,2006,7 Suppl 4(Suppl 4):S25.[13]Aghaei F,Tan M,Hollingsworth AB,et al.Applying ANewQuantitative Global Breast MRI Feature Analysis Scheme to Assess Tumor Response to Chemotherapy[J].J Magn Reson Imaging,2016,44(5):1099-1106.[14]Johansen R,Jensen LR,Rydland J,et al.Predicting Survival and Early Clinical Response to Primary Chemotherapy for Patients with Locally Advanced Breast Cancer Using DCE-MRI[J].J Magn Reson Imaging,2009,29(6):1300-1307.[15]Teruel JR,Heldahl MG,Goa PE,et al.Dynamic Contrast-Enhanced MRI Texture Analysis for Pretreatment Prediction of Clinical and Pathological Response to Neoadjuvant Chemotherapy in Patients with Locally Advanced Breast Cancer[J].NMR Biomed,2014,27(8):887-896.[16]Hattangadi J,Park C,Rembert J,et al.Breast Stromal Enhancement on MRI Is Associated with Response to Neoadjuvant Chemotherapy[J].AJR Am J Roentgenol,2008,190(6):1630-1636.[17]Kim SY,Cho N,Shin SU,et al.Contrast-Enhanced MRI After Neoadjuvant Chemotherapy of Breast Cancer: Lesion-to-Background Parenchymal Signal Enhancement Ratio for Discriminating Pathological Complete Response from Minimal Residual Tumour[J].Eur Radiol,2018,28(7):2986-2995.[18]王琼瑶,何友全,彭小玲.基于改进遗传算法的支持向量机参数优化方法[J].计算机与现代化,2015(3):33-36.

相似文献/References:

[1]屈欣然.DCE-MRI对局部进展期乳腺癌新辅助化疗的临床指导价值研究[J].医学信息,2018,31(04):34.[doi:10.3969/j.issn.1006-1959.2018.04.012]
　QU Xin-ran.Clinical Value of DCE-MRI in Neoadjuvant Chemotherapy for Locally Advanced Breast Cancer[J].Journal of Medical Information,2018,31(19):34.[doi:10.3969/j.issn.1006-1959.2018.04.012]
[2]董 云.SPECT/CT 双时相99mTc- MIBI断层显像 在乳腺癌诊断中的应用[J].医学信息,2018,31(05):147.[doi:10.3969/j.issn.1006-1959.2018.05.053]
　DONG Yun.Application of SPECT/CT Dual-phase 99mTc-MIBI Tomography in the Diagnosis of Breast Cancer[J].Journal of Medical Information,2018,31(19):147.[doi:10.3969/j.issn.1006-1959.2018.05.053]
[3]王文重,伊 扬,董晓昕,等.食管癌患者新辅助化疗前后生理指标变化及意义[J].医学信息,2022,35(09):116.[doi:10.3969/j.issn.1006-1959.2022.09.029]
　WANG Wen-zhong,YI Yang,DONG Xiao-xin,et al.Changes and Significance of Physiological Indexes in Esophageal Cancer Patients Before and After Neoadjuvant Chemotherapy[J].Journal of Medical Information,2022,35(19):116.[doi:10.3969/j.issn.1006-1959.2022.09.029]
[4]李建辉,祝旭龙.乳腺癌的治疗管理——把好生活质量关[J].医学信息,2018,31(13):1.[doi:10.3969/j.issn.1006-1959.2018.13.001]
[5]杨玉峰,李 强,曲国红,等.乳腺癌术前巴德枪穿刺检测ER、PR的临床研究[J].医学信息,2018,31(13):163.[doi:10.3969/j.issn.1006-1959.2018.13.050]
　YANG Yu-feng,LI Qiang,QU Guo-hong,et al.Clinical Study on Preoperative ER and PR Detection of Bard Gun Puncture in Breast Cancer[J].Journal of Medical Information,2018,31(19):163.[doi:10.3969/j.issn.1006-1959.2018.13.050]
[6]刘雪静,路 红,徐熠琳,等.乳腺影像与容积导航技术初步应用的临床分析[J].医学信息,2022,35(11):161.[doi:10.3969/j.issn.1006-1959.2022.11.043]
　LIU Xue-jing,LU Hong,XU Yi-lin,et al.Clinical Analysis of the Preliminary Application of Breast Imaging and Volume Navigation Technology[J].Journal of Medical Information,2022,35(19):161.[doi:10.3969/j.issn.1006-1959.2022.11.043]
[7]刘在波,修文娟.背阔肌皮瓣联合假体在乳腺癌改良根治术后乳房重建中的应用[J].医学信息,2019,32(02):127.[doi:10.3969/j.issn.1006-1959.2019.02.036]
　LIU Zai-bo,XIU Wen-juan.Application of Latissimus Dorsi Flap Combined with Prosthesis in Breast Reconstruction after Modified Radical Mastectomy for Breast Cancer[J].Journal of Medical Information,2019,32(19):127.[doi:10.3969/j.issn.1006-1959.2019.02.036]
[8]孟庆国,王佳雪,王斯佳.马来酸吡咯替尼对乳腺癌术后患者心脏功能的影响[J].医学信息,2019,32(02):152.[doi:10.3969/j.issn.1006-1959.2019.02.045]
　MENG Qing-guo,WANG Jia-xue,WANG Si-jia.Effect of Pyrrolidine Maleate on Cardiac Function in Patients after Breast Cancer Operation[J].Journal of Medical Information,2019,32(19):152.[doi:10.3969/j.issn.1006-1959.2019.02.045]
[9]刘 曼,厉玛倩倩,徐润润,等.延续性护理对乳腺癌术后化疗患者生活质量与 健康知识水平的影响[J].医学信息,2018,31(16):178.[doi:10.3969/j.issn.1006-1959.2018.16.057]
　LIU Man,LIMA Qian-qian,XU Run-run,et al.Effect of Continuous Nursing on Quality of Life and Health Knowledge of Postoperative Chemotherapy Patients with Breast Cancer[J].Journal of Medical Information,2018,31(19):178.[doi:10.3969/j.issn.1006-1959.2018.16.057]
[10]满枋霖,葛俊宏,林晓月,等.p21活化激酶在乳腺癌中的作用研究[J].医学信息,2019,32(06):35.[doi:10.3969/j.issn.1006-1959.2019.06.013]
　MAN Fang-lin,GE Jun-hong,LIN Xiao-yue,et al.The Role of p21 Activated Kinase in Breast Cancer[J].Journal of Medical Information,2019,32(19):35.[doi:10.3969/j.issn.1006-1959.2019.06.013]
[11]付大鹏.多西他赛联合表柔比星新辅助化疗治疗乳腺癌的效果[J].医学信息,2020,33(05):149.[doi:10.3969/j.issn.1006-1959.2020.04.048]
　FU Da-peng.Evaluation of Docetaxel Combined with Epirubicin Neoadjuvant Chemotherapy for Breast Cancer[J].Journal of Medical Information,2020,33(19):149.[doi:10.3969/j.issn.1006-1959.2020.04.048]
[12]陈甜甜,潘跃银,沈夏波.新辅助内分泌治疗与新辅助化疗用于乳腺癌中的短期疗效Meta分析[J].医学信息,2020,33(14):67.[doi:10.3969/j.issn.1006-1959.2020.14.020]
　CHEN Tian-tian,PAN Yue-yin,SHEN Xia-bo.Meta-analysis of Short-term Efficacy of Neoadjuvant Endocrine Therapy and Neoadjuvant Chemotherapy in Breast Cancer[J].Journal of Medical Information,2020,33(19):67.[doi:10.3969/j.issn.1006-1959.2020.14.020]
[13]张 虎,张 涛,徐 芳,等.MRI在乳腺癌诊治中的研究进展[J].医学信息,2022,35(10):62.[doi:10.3969/j.issn.1006-1959.2022.10.015]
　ZHANG Hu,ZHANG Tao,XU Fang,et al.Research Progress of MRI in Diagnosis and Treatment of Breast Cancer[J].Journal of Medical Information,2022,35(19):62.[doi:10.3969/j.issn.1006-1959.2022.10.015]
[14]关利平,李建文.凝血相关指标与乳腺癌新辅助化疗疗效预测的相关性研究[J].医学信息,2022,35(16):159.[doi:10.3969/j.issn.1006-1959.2022.16.041]
　GUAN Li-ping,LI Jian-wen.Study on Correlation Between Coagulation- related Indexes and Efficacy Prediction of Neoadjuvant Chemotherapy for Breast Cancer[J].Journal of Medical Information,2022,35(19):159.[doi:10.3969/j.issn.1006-1959.2022.16.041]
[15]杨季春,卿伯华,杨民正.动态增强MRI预测乳腺癌新辅助化疗后病理完全缓解的准确性[J].医学信息,2024,37(22):102.[doi:10.3969/j.issn.1006-1959.2024.22.025]
　YANG Jichun,QING Bohua,YANG Minzheng.Accuracy of Dynamic Contrast-enhanced MRI in Predicting Pathological Complete Response After Neoadjuvant Chemotherapy for Breast Cancer[J].Journal of Medical Information,2024,37(19):102.[doi:10.3969/j.issn.1006-1959.2024.22.025]
[16]樊 婷,刘新兰２.乳腺癌组织中TOP2A蛋白表达临界值及其与蒽环紫杉类新辅助化疗疗效相关性研究[J].医学信息,2025,38(01):79.[doi:10.3969/j.issn.1006-1959.2025.01.014]
　FAN Ting,LIU Xinlan.Critical Value of TOP2A Protein Expression in Breast Cancer Tissues and its Correlation with the Efficacy of Anthracycline Neoadjuvant Chemotherapy[J].Journal of Medical Information,2025,38(19):79.[doi:10.3969/j.issn.1006-1959.2025.01.014]

更新日期/Last Update: 1900-01-01