Machine learning for diagnostic ultrasound of triple-negative breast cancer

• Published in: Breast cancer research and treatment Vol. 173; no. 2; pp. 365 - 373
• Main Authors: Wu, Tong, Sultan, Laith R., Tian, Jiawei, Cary, Theodore W., Sehgal, Chandra M.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.01.2019; Springer; Springer Nature B.V
• Subjects: Adult; Artificial intelligence; Breast - diagnostic imaging; Breast - pathology; Breast cancer; Cancer research; Clinical Trial; Diagnosis; Discrimination; Early Detection of Cancer - methods; Epidermal growth factor; Epidermal growth factors; ErbB-2 protein; Estrogen receptors; Estrogens; Female; Humans; Image processing; Image Processing, Computer-Assisted - methods; Learning algorithms; Machine Learning; Medicine; Medicine & Public Health; Middle Aged; Oncology; Phenols (Class of compounds); Progesterone; Retrospective Studies; ROC Curve; Sex hormones; Statistical analysis; Triple Negative Breast Neoplasms - diagnostic imaging; Triple Negative Breast Neoplasms - pathology; Ultrasonic imaging; Ultrasonography, Doppler, Color - methods; Ultrasonography, Mammary - methods; Ultrasound; Visual discrimination; Quantitative breast ultrasound; Breast cancer; Triple negative breast cancer; Computer-aided diagnosis; Machine learning
• ISSN: 0167-6806; 1573-7217; 1573-7217
• DOI: 10.1007/s10549-018-4984-7

Purpose 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. Methods 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. Results 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. Conclusions 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.