Convolutional neural networks (CNNs) show potential for computer-aided diagnosis (CADx) by learning features directly from the image data instead of using analytically extracted features. However, CNNs are difficult to train from scratch for medical images due to small sample sizes and variations in tumor presentations. Instead, transfer learning can be used to extract tumor information from medical images via CNNs originally pretrained for nonmedical tasks, alleviating the need for large datasets. Our database includes 219 breast lesions (607 full-field digital mammographic images). We compared support vector machine classifiers based on the CNN-extracted image features and our prior computer-extracted tumor features in the task of distinguishing between benign and malignant breast lesions. Five-fold cross validation (by lesion) was conducted with the area under the receiver operating characteristic (ROC) curve as the performance metric. Results show that classifiers based on CNN-extracted features (with transfer learning) perform comparably to those using analytically extracted features [area under the ROC curve $(AUC)=0.81$]. Further, the performance of ensemble classifiers based on both types was significantly better than that of either classifier type alone ($AUC=0.86$ versus 0.81, $p=0.022$). We conclude that transfer learning can improve current CADx methods while also providing standalone classifiers without large datasets, facilitating machine-learning methods in radiomics and precision medicine.

Using quantitative radiomics, we demonstrate that computer-extracted magnetic resonance (MR) image-based tumor phenotypes can be predictive of the molecular classification of invasive breast cancers. Radiomics analysis was performed on 91 MRIs of biopsy-proven invasive breast cancers from National Cancer Institute’s multi-institutional TCGA/TCIA. Immunohistochemistry molecular classification was performed including estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2, and for 84 cases, the molecular subtype (normal-like, luminal A, luminal B, HER2-enriched, and basal-like). Computerized quantitative image analysis included: three-dimensional lesion segmentation, phenotype extraction, and leave-one-case-out cross validation involving stepwise feature selection and linear discriminant analysis. The performance of the classifier model for molecular subtyping was evaluated using receiver operating characteristic analysis. The computer-extracted tumor phenotypes were able to distinguish between molecular prognostic indicators; area under the ROC curve values of 0.89, 0.69, 0.65, and 0.67 in the tasks of distinguishing between ER+ versus ER−, PR+ versus PR−, HER2+ versus HER2−, and triple-negative versus others, respectively. Statistically significant associations between tumor phenotypes and receptor status were observed. More aggressive cancers are likely to be larger in size with more heterogeneity in their contrast enhancement. Even after controlling for tumor size, a statistically significant trend was observed within each size group (P=0.04 for lesions ⩽2 cm; P=0.02 for lesions >2 to ⩽5 cm) as with the entire data set (P-value=0.006) for the relationship between enhancement texture (entropy) and molecular subtypes (normal-like, luminal A, luminal B, HER2-enriched, basal-like). In conclusion, computer-extracted image phenotypes show promise for high-throughput discrimination of breast cancer subtypes and may yield a quantitative predictive signature for advancing precision medicine.

Genomic and radiomic imaging profiles of invasive breast carcinomas from The Cancer Genome Atlas and The Cancer Imaging Archive were integrated and a comprehensive analysis was conducted to predict clinical outcomes using the radiogenomic features. Variable selection via LASSO and logistic regression were used to select the most-predictive radiogenomic features for the clinical phenotypes, including pathological stage, lymph node metastasis, and status of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). Cross-validation with receiver operating characteristic (ROC) analysis was performed and the area under the ROC curve (AUC) was employed as the prediction metric. Higher AUCs were obtained in the prediction of pathological stage, ER, and PR status than for lymph node metastasis and HER2 status. Overall, the prediction performances by genomics alone, radiomics alone, and combined radiogenomics features showed statistically significant correlations with clinical outcomes; however, improvement on the prediction performance by combining genomics and radiomics data was not found to be statistically significant, most likely due to the small sample size of 91 cancer cases with 38 radiomic features and 144 genomic features.

Combining radiomic analysis of breast tumors with that of parenchyma may improve diagnostic accuracy for breast cancer. BackgroundPrevious studies have suggested that breast parenchymal texture features may reflect the biologic risk factors associated with breast cancer development. Therefore, combining the characte...

The texture of a medical image refers to the coarseness, consistency, and arrangement of pixels within the image. Radiologists typically make qualitative assessments regarding a patient’s condition based on the texture and spatial patterns they perceive within the image.¹ With large databases of medical images available for researchers and radiologists to analyze, high-throughput computing through the conversion of medical images into mineable data (i.e., radiomics) is possible, which may offer additional insight into a patient’s underlying pathophysiology.^{2 – 4} Since the development of several systematic and large-scale texture analysis-based computer-aided diagnosis (CAD) schemes began at the University of Chicago in the early 1980s, investigators have used texture analysis over the decades to develop various automated detection, diagnosis, and segmentation strategies. By the mid 1980s, the University of Chicago had set the foundation for CAD schemes used to detect lung nodules in digital chest radiographs as well as microcalcifications in mammography images.^{5 – 8} While these early investigations resulted in a relatively large number of false positives per patient, it was shown that radiologists’ ability to detect lesions significantly improved when these texture analysis-based CAD schemes were used.^9,10 Since these earlier investigations, programs using texture analysis have extended to examine and make clinical predictions concerning additional tissues such as the colon, liver, and brain.^{11 – 13} Because of the growing promise texture analysis and radiomics have shown and the large amounts of imaging data available, investigators have become more interested in quantifying texture to increase the amount of information that can be extracted from medical images and to limit variability among radiologists.