Department of Ultrasound, Second Affiliated Hospital of Harbin Medical University, 246 Xuefu Road, Harbin City, 150086, Heilongjiang Province, People's Republic of China.
Ultrasound Research Lab, Department of Radiology, University of Pennsylvania, 168B John Morgan Building, 3620 Hamilton Walk, Philadelphia, PA, 19104, USA.
Breast Cancer Res Treat. 2019 Jan;173(2):365-373. doi: 10.1007/s10549-018-4984-7. Epub 2018 Oct 20.
Early diagnosis of triple-negative (TN) breast cancer is important due to its aggressive biological characteristics, poor clinical outcomes, and limited options for therapy. The goal of this study is to evaluate the potential of machine learning with quantitative ultrasound image features for the diagnosis of TN breast cancer.
Ultrasonic and clinical data of 140 surgically confirmed breast cancer cases were analyzed retrospectively for the diagnosis of TN and non-TN (NTN) subtypes. The subtypes were classified based on the expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). Ultrasound image features were measured from the grayscale and color Doppler images and used with logistic regression for classification by machine learning. Leave-one-out cross validation was used to train and test the differentiation. Diagnostic performance was measured by the area under receiver operating characteristic (ROC) curve, and sensitivity and specificity determined at the Youdons index.
Of the twelve grayscale and Doppler features measured, eight were found to be statistically different for the TN and NTN subtypes (p < 0.05). The area under the ROC curve (AUC) of the statistically significant grayscale (GS) and color Doppler (CD) features was 0.85 and 0.65, respectively. The AUC increased to 0.88 when the GS and CD features were used together, with sensitivity of 86.96% and specificity of 82.91%. Consideration of patient age in the analysis did not improve discrimination of TN and NTN.
The analysis of breast ultrasound images by machine learning achieves high level of differentiation between the TN and NTN subtypes, exceeding the diagnostic performance by standard visual assessments of the images.
由于三阴性（TN）乳腺癌具有侵袭性的生物学特征、较差的临床预后和有限的治疗选择，因此早期诊断非常重要。本研究旨在评估基于机器学习的定量超声图像特征在诊断 TN 乳腺癌中的应用潜力。
回顾性分析了 140 例经手术证实的乳腺癌病例的超声和临床资料，以诊断 TN 和非 TN（NTN）亚型。根据雌激素受体（ER）、孕激素受体（PR）和人表皮生长因子受体 2（HER2）的表达对亚型进行分类。从灰阶和彩色多普勒图像中测量超声图像特征，并使用逻辑回归进行机器学习分类。采用留一法交叉验证进行训练和测试。通过接收者操作特征（ROC）曲线下面积（AUC）来衡量诊断性能，并通过 Youdons 指数确定灵敏度和特异性。
在所测量的 12 个灰阶和多普勒特征中，有 8 个在 TN 和 NTN 亚型之间存在统计学差异（p<0.05）。在统计学上有意义的灰阶（GS）和彩色多普勒（CD）特征的 ROC 曲线下面积（AUC）分别为 0.85 和 0.65。当同时使用 GS 和 CD 特征时，AUC 增加至 0.88，灵敏度为 86.96%，特异性为 82.91%。在分析中考虑患者年龄并不能提高 TN 和 NTN 的鉴别能力。
通过机器学习对乳腺超声图像进行分析，可以在 TN 和 NTN 亚型之间实现高度的区分，其诊断性能优于图像的标准视觉评估。
