A Machine Learning Ensemble Based on Radiomics to Predict BI-RADS Category and Reduce the Biopsy Rate of Ultrasound-Detected Suspicious Breast Masses

Abstract

We developed a machine learning model based on radiomics to predict the BI-RADS category of ultrasound-detected suspicious breast lesions and support medical decision-making towards short-interval follow-up versus tissue sampling. From a retrospective 2015–2019 series of ultrasound-guided core needle biopsies performed by four board-certified breast radiologists using six ultrasound systems from three vendors, we collected 821 images of 834 suspicious breast masses from 819 patients, 404 malignant and 430 benign according to histopathology. A balanced image set of biopsy-proven benign (n = 299) and malignant (n = 299) lesions was used for training and cross-validation of ensembles of machine learning algorithms supervised during learning by histopathological diagnosis as a reference standard. Based on a majority vote (over 80% of the votes to have a valid prediction of benign lesion), an ensemble of support vector machines showed an ability to reduce the biopsy rate of benign lesions by 15% to 18%, always keeping a sensitivity over 94%, when externally tested on 236 images from two image sets: (1) 123 lesions (51 malignant and 72 benign) obtained from two ultrasound systems used for training and from a different one, resulting in a positive predictive value (PPV) of 45.9% (95% confidence interval 36.3–55.7%) versus a radiologists’ PPV of 41.5% (p < 0.005), combined with a 98.0% sensitivity (89.6–99.9%); (2) 113 lesions (54 malignant and 59 benign) obtained from two ultrasound systems from vendors different from those used for training, resulting into a 50.5% PPV (40.4–60.6%) versus a radiologists’ PPV of 47.8% (p < 0.005), combined with a 94.4% sensitivity (84.6–98.8%). Errors in BI-RADS 3 category (i.e., assigned by the model as BI-RADS 4) were 0.8% and 2.7% in the Testing set I and II, respectively. The board-certified breast radiologist accepted the BI-RADS classes assigned by the model in 114 masses (92.7%) and modified the BI-RADS classes of 9 breast masses (7.3%). In six of nine cases, the model performed better than the radiologist did, since it assigned a BI-RADS 3 classification to histopathology-confirmed benign masses that were classified as BI-RADS 4 by the radiologist.

An ultrasound-based nomogram model in the assessment of pathological complete response of neoadjuvant chemotherapy in breast cancer

“…In this study, six types of texture features including GLCM, NGTDM, GLSZM, GLRM, GLDM, and First Order features were enrolled, and it was found that NGTDM and GLCM had similar weights. NGTDM quantifies the difference between a gray value and the average gray value of its neighbors within distance δ ( 30 – 34 ). Here, when we used NGTDM to evaluate the tumor ROI region, we also verified that NGTDM could accurately reflect tumor heterogeneity, thereby accurately predicting the NAC efficacy in patients.…”

“…The breast cancer patients with a total score of > 71.742 were more likely to achieve pCR after NAC. In recent years, the Nomogram prediction model has been widely used in the clinic ( 33 , 34 ). However, the indicators included in the Nomogram model for predicting the efficacy of NAC in breast cancer are confusing, and there is no conclusion on the evaluation time point of NAC.…”

The Potential and Emerging Role of Quantitative Imaging Biomarkers for Cancer Characterization

“…For example, combining prostate-specific antigen density and MRI for prostate biopsy planning allows the optimization of biopsy planning [ 110 ]. Interlenghi et al were even able to predict the BI-RADS category for suspicious breast masses using ultrasound radiomics and training-modified random forest classifiers, support vector machines, and k-nearest neighbor classifiers [ 111 ]. Further developments could even lead to radiomics replacing biopsy and facilitate therapeutic decisions by predicting BI-RADS category by imaging.…”