切换至 "中华医学电子期刊资源库"

中华乳腺病杂志(电子版) ›› 2024, Vol. 18 ›› Issue (06) : 353 -360. doi: 10.3877/cma.j.issn.1674-0807.2024.06.005

论著

超声影像组学对致密型乳腺背景中非肿块型乳腺癌的诊断价值
邱琳1, 刘锦辉2, 组木热提·吐尔洪1, 马悦心1, 冷晓玲2,()   
  1. 1.830011 乌鲁木齐,新疆医科大学附属肿瘤医院超声诊断科
    2.523059 东莞,南方医科大学附属第十医院(东莞市人民医院)超声科
  • 收稿日期:2023-10-17 出版日期:2024-12-01
  • 通信作者: 冷晓玲

Diagnostic value of ultrasound radiomics in non-mass breast cancer in dense breasts

Lin Qiu1, Jinhui Liu2, Tuerhong Zumureti1, Yuexin Ma1, Xiaoling Leng2,()   

  1. 1.Department of Ultrasound,Affiliated Tumor Hospital of Xinjiang Medical University, Urumqi 830011, China
    2.Department of Ultrasound,Tenth Affiliated Hospital of Southern Medical University/Dongguan People’s Hospital, Dongguan 523059, China
  • Received:2023-10-17 Published:2024-12-01
  • Corresponding author: Xiaoling Leng
引用本文:

邱琳, 刘锦辉, 组木热提·吐尔洪, 马悦心, 冷晓玲. 超声影像组学对致密型乳腺背景中非肿块型乳腺癌的诊断价值[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 353-360.

Lin Qiu, Jinhui Liu, Tuerhong Zumureti, Yuexin Ma, Xiaoling Leng. Diagnostic value of ultrasound radiomics in non-mass breast cancer in dense breasts[J/OL]. Chinese Journal of Breast Disease(Electronic Edition), 2024, 18(06): 353-360.

目的

探讨超声影像组学特征对检出致密型乳腺背景中的非肿块型乳腺癌的诊断价值。

方法

回顾性分析2017 年1 月1 日到2023 年1 月30 日东莞市人民医院及新疆医科大学附属肿瘤医院619 例致密型乳腺背景中的非肿块型病变(NML)的二维超声图像,采用7 ∶3的比例进行随机分组,训练组434 例,验证组185 例,共提取848 个影像组学特征,采用最小绝对收缩和选择算子(LASSO)回归模型进行特征筛选,通过LASSO-Logistic 回归来建立影像组学模型,并与临床和超声特征进行整合构建联合模型。 通过比较受试者工作特征(ROC)曲线评价模型的诊断效能。 用校准曲线评估模型的一致性,用决策曲线分析(DCA)评估模型的临床价值,用DeLong 检验将其与其余模型进行比较。

结果

术后病理结果显示619 例乳腺NML 中,恶性304 例,良性315 例。 单因素和多因素Logistic 回归分析结果显示,年龄、病灶长度、细小钙化、周围结构扭曲、血流特征是恶性病变的独立预测因素(OR=1.053、8.197、0.701、3.479、1.195;95%CI:1.027~1.080, 4.895~14.154, 0.573~0.857,2.044~6.044, 1.536~2.408;P 均<0.050)。 共筛选出12 个非零系数的影像组学特征。 将筛选出临床指标和影像组学特征整合,创建了联合预测模型。 联合预测模型的训练组ROC 曲线下面积为0.89(95%CI:0.86~0.92),验证组曲线下面积为0.83(95%CI:0.78~0.89)。 DeLong 检验表明,联合模型与临床模型、超声模型、影像组学模型比较,差异有统计学意义(Z=-3.974、-3.338、-3.468,P 均<0.050)。 联合模型的DCA 曲线下面积最大,训练组为0.12,验证组为0.22。 校准曲线显示,与其他模型相比,联合模型在预测结果与真实病理结果具有更好的一致性。

结论

超声影像组学与临床指标的联合模型对于致密型乳腺背景中NML 的良恶性的鉴别具有较好的效能,可为乳腺癌的临床治疗决策提供支持。

Objective

To investigate the diagnostic value of ultrasound radiomics features for the detection of non-mass breast cancer in dense breasts.

Methods

We retrospectively analyzed 2D ultrasound images of 619 patients with non-mass breast lesions (NML) in dense breasts in Dongguan People’s Hospital and Affiliated Tumor Hospital of Xinjiang Medical University between January 1st, 2017 and January 30th,2023. They were randomized into two groups using a 7 ∶3 ratio(434 cases in the training group and 185 cases in the validation group). Totally 848 imaging features were extracted. The least absolute shrinkage and selection operator (LASSO) regression model was used to screen the features, and the radiomics model was built by the LASSO-Logistic regression. The joint model was constructed by integrating it with clinical and ultrasound features. The diagnostic efficacy of the model was evaluated by the receiver operating characteristic (ROC)curve. The consistency of the model was assessed with the calibration curve and the clinical value of the model was assessed by decision curve analysis (DCA). The DeLong test was used to compare those models.

Results

The postoperative pathological results showed that among 619 cases of breast NML, 304 cases were malignant and 315 cases were benign. The results of univariate and multivariate Logistic regression analyses indicated that age, lesion length, microcalcification, distortion of surrounding structures, and blood flow features were independent predictors of malignancy (OR=1.053,8.197,0.701,3.479,1.195;95%CI:1.027-1.080, 4.895-14.154, 0.573-0.857, 2.044-6.044, 1.536-2.408; all P<0.050). A total of 12 radiomics features with non-zero coefficients were screened out. The selected clinical factors and radiomics features were integrated to create a joint prediction model. The area under the ROC curve of the joint model was 0.89(95%CI:0.86-0.92) in the training group, and 0.83 (95%CI: 0.78-0.89) in the validation group. The area under the DCA curve of the joint model was the largest(0.12 in the training group and 0.22 in the validation group). The calibration curve showed that the joint model had a better consistency between the predicted results and the actual pathological results compared with other models. The DeLong test demonstrated that there were statistically significant differences comparing the joint model with other models (the clinical model, the ultrasound model and the radiomics model) (Z=-3.974,-3.338,-3.468;all P<0.050).

Conclusions

The joint model integrating clinical factors and radiomics features of ultrasound has good efficacy in identifying the benign and malignant nature of NML in dense breasts, which can provide guidance for clinical decision-making in breast cancer.

图1 1 例47 岁导管原位癌患者的ROI 勾画示意图
表1 训练组和验证组非肿块型乳腺病变患者的临床病理特征比较
图2 619 例非肿块型乳腺病变患者的病理类型分布 a、b 图分别显示良性、恶性病理类型
表2 训练组非肿块型乳腺病变患者的临床和超声特征与病变良恶性的关系单因素分析
表3 训练组非肿块型乳腺病变患者的临床和超声特征与病变良恶性关系的多因素分析(n=434)
表4 不同模型的鉴别效能
图3 临床模型、超声模型、影像组学模型和联合模型的ROC 曲线及校准曲线
表5 影像组学特征参数图
图4 训练集(a)和验证集(b)联合模型的决策曲线分析
[1]
Zhang J, Cai L, Chen L, et al. Re-evaluation of high-risk breast mammography lesions by target ultrasound and ABUS of breast nonmass-like lesions [J]. BMC Med Imaging, 2021, 21(1): 156.
[2]
Choe J,Chikarmane SA,Giess C S. Nonmass findings at breast US:definition,classifications, and differential diagnosis [J]. Radiographics,2020, 40(2): 326-335.
[3]
Zhang W,Xiao X,Xu X, et al. Non-mass breast lesions on ultrasound:feature exploration and multimode ultrasonic diagnosis [J]. Ultrasound Med Biol, 2018, 44(8): 1703-1711.
[4]
Wang ZL, Li N, Li M, et al. Non-mass-like lesions on breast ultrasound: classification and correlation with histology [J]. Radiol Med, 2015, 120(10): 905-910.
[5]
吴琳, 张丹. 女性乳腺超声分型的研究进展[J/CD]. 中华临床医师杂志(电子版), 2015, 9(15): 2926-2929.
[6]
奚群, 侯新燕, 崔丽丽, 等. 不同乳腺超声分型导管管径及腺体厚径变化趋势及其发生乳腺癌的危险性[J]. 河北医科大学学报,2014, 35(09): 1025-1028.
[7]
Ohuchi N,Suzuki A,Sobue T, et al. Sensitivity and specificity of mammography and adjunctive ultrasonography to screen for breast cancer in the Japan Strategic Anti-cancer Randomized Trial (J-START): a randomised controlled trial [J]. Lancet, 2016, 387(10016):341-348.
[8]
Li T,Mello-Thoms C,Brennan PC. Descriptive epidemiology of breast cancer in China: incidence, mortality, survival and prevalence [J].Breast Cancer Res Treat, 2016, 159(3): 395-406.
[9]
Brem RF,Lenihan MJ,Lieberman J,et al. Screening breast ultrasound:past,present,and future [J]. Am J Roentgenol,2015,204(2):234-240.
[10]
Lambin P,Rios-Velazquez E,Leijenaar R, et al. Radiomics: extracting more information from medical images using advanced feature analysis[J]. Eur J Cancer, 2012, 48(4): 441-6.
[11]
Moon HJ,Kim EK, Kim MJ, et al. Comparison of clinical and pathologic characteristics of ductal carcinoma in situ detected on mammography versus ultrasound only in asymptomatic patients [J].Ultrasound Med Biol, 2019, 45(1): 68-77.
[12]
秦丽莎.基于超声的影像组学结合临床参数术前预测子宫内膜癌淋巴结转移的价值[D], 2022.
[13]
Ko KH,Jung HK,Kim SJ,et al. Potential role of shear-wave ultrasound elastography for the differential diagnosis of breast non-mass lesions:preliminary report [J]. Eur Radiol, 2014, 24(2): 305-311.
[14]
Desai P,Aggarwal A. Breast cancer in women over 65 years-a review of screening and treatment options [J]. Clin Geriatr Med, 2021, 37(4):611-623.
[15]
Gunawardena DS,Burrows S,Taylor DB. Non-mass versus mass-like ultrasound patterns in ductal carcinoma in situ: is there an association with high-risk histology? [J]. Clin Radiol, 2020, 75(2): 140-147.
[16]
牛瑞兰, 付乃芹, 王知力. 非肿块型乳腺癌超声特征与组织学及生物学预后因子的相关性[J]. 中国超声医学杂志, 2022, 38(7):758-762.
[17]
薛雯, 杨柳茵, 范丽,等. 超微血管成像技术鉴别乳腺良恶性病变[J]. 中国医学影像技术, 2019, 35(1): 77-81.
[18]
Scott R,Lyburn I,Cornford E,et al. Breast calcification micromorphology classification [J]. Br J Radiol,2022,95(1139):20220485.
[19]
Yabuuchi H,Matsuo Y,Kamitani T, et al. Non-mass-like enhancement on contrast-enhanced breast MR imaging: lesion characterization using combination of dynamic contrast-enhanced and diffusion-weighted MR images [J]. Eur J Radiol, 2010, 75(1): e126-132.
[20]
Uematsu T. Non-mass-like lesions on breast ultrasonography: a systematic review [J]. Breast Cancer, 2012, 19(4): 295-301.
[21]
Ko ES,Choi HY,Kim RB, et al. Application of sonoelastography:comparison of performance between mass and non-mass lesion [J]. Eur J Radiol, 2012, 81(4): 731-736.
[22]
胡杉, 梁长虹, 刘再毅, 等. 纹理分析及影像组学在非肿瘤病变的研究应用及进展[J]. 中华放射学杂志, 2019, 53(6): 526-529.
[23]
Park KW,Park S,Shon I, et al. Non-mass lesions detected by breast US: stratification of cancer risk for clinical management [J]. Eur Radiol, 2021, 31(3): 1693-1706.
[24]
李汉森, 章强, 齐海坤, 等. 基于增强MRI 纹理分析区分非产褥期乳腺炎与非肿块样强化病灶乳腺癌[J]. 中国医学影像学杂志,2017, 25(5): 354-359.
[25]
Tan Y,Mai H,Huang Z, et al. Additive value of texture analysis based on breast MRI for distinguishing between benign and malignant nonmass enhancement in premenopausal women [J]. BMC Med Imaging,2021, 21(1): 48.
[26]
Li Y,Yang ZL,Lv WZ, et al. Non-mass enhancements on DCE-MRI:development and validation of a radiomics-based signature for breast cancer diagnoses [J]. Front Oncol, 2021, 11: 738330.
[1] 李洋, 蔡金玉, 党晓智, 常婉英, 巨艳, 高毅, 宋宏萍. 基于深度学习的乳腺超声应变弹性图像生成模型的应用研究[J/OL]. 中华医学超声杂志(电子版), 2024, 21(06): 563-570.
[2] 洪玮, 叶细容, 刘枝红, 杨银凤, 吕志红. 超声影像组学联合临床病理特征预测乳腺癌新辅助化疗完全病理缓解的价值[J/OL]. 中华医学超声杂志(电子版), 2024, 21(06): 571-579.
[3] 河北省抗癌协会乳腺癌专业委员会护理协作组. 乳腺癌中心静脉通路护理管理专家共识[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 321-329.
[4] 刘晨鹭, 刘洁, 张帆, 严彩英, 陈倩, 陈双庆. 增强MRI 影像组学特征生境分析在预测乳腺癌HER-2 表达状态中的应用[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 339-345.
[5] 张晓宇, 殷雨来, 张银旭. 阿帕替尼联合新辅助化疗对三阴性乳腺癌的疗效及预后分析[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 346-352.
[6] 程燕妮, 樊菁, 肖瑶, 舒瑞, 明昊, 党晓智, 宋宏萍. 乳腺组织定位标记夹的应用与进展[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 361-365.
[7] 涂盛楠, 胡芬, 张娟, 蔡海峰, 杨俊泉. 天然植物提取物在乳腺癌治疗中的应用[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 366-370.
[8] 朱文婷, 顾鹏, 孙星. 非酒精性脂肪性肝病对乳腺癌发生发展及治疗的影响[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 371-375.
[9] 周荷妹, 金杰, 叶建东, 夏之一, 王进进, 丁宁. 罕见成人肋骨郎格汉斯细胞组织细胞增生症被误诊为乳腺癌术后骨转移一例[J/OL]. 中华乳腺病杂志(电子版), 2024, 18(06): 380-383.
[10] 韩萌萌, 冯雪园, 马宁. 乳腺癌改良根治术后桡神经损伤1例[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(01): 117-118.
[11] 高杰红, 黎平平, 齐婧, 代引海. ETFA和CD34在乳腺癌中的表达及与临床病理参数和预后的关系研究[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(01): 64-67.
[12] 张志兆, 王睿, 郜苹苹, 王成方, 王成, 齐晓伟. DNMT3B与乳腺癌预后的关系及其生物学机制[J/OL]. 中华普外科手术学杂志(电子版), 2024, 18(06): 624-629.
[13] 王玲艳, 高春晖, 冯雪园, 崔鑫淼, 刘欢, 赵文明, 张金库. 循环肿瘤细胞在乳腺癌新辅助及术后辅助治疗中的应用[J/OL]. 中华普外科手术学杂志(电子版), 2024, 18(06): 630-633.
[14] 赵林娟, 吕婕, 王文胜, 马德茂, 侯涛. 超声引导下染色剂标记切缘的梭柱型和圆柱型保乳区段切除术的效果研究[J/OL]. 中华普外科手术学杂志(电子版), 2024, 18(06): 634-637.
[15] 陆镜明, 韩大为, 任耀星, 黄天笑, 向俊西, 张谞丰, 吕毅, 王傅民. 基于术前影像组学的肝内胆管细胞癌淋巴结转移预测的系统性分析[J/OL]. 中华肝脏外科手术学电子杂志, 2024, 13(06): 852-858.
阅读次数
全文


摘要