RESUMEN
Background: The American College of Radiology Breast Imaging Reporting and Data System (ACR BI-RADS) used with ultrasonography cannot guide the individual management of solid breast tumors, but preoperative core biopsy categories (CBCs) can. We aimed to use machine learning to analyze clinical and ultrasonic features for predicting CBCs and to aid in the development of a new ultrasound (US) imaging reporting system for solid tumors of the breast. Methods: This retrospective study included women with solid breast tumors who underwent US-guided core needle biopsy from March 1, 2019, to December 31, 2019. All patients were randomly assigned to a training or validation cohort (7:3 ratio). CBC was predicted using 5 machine learning models: random forest (RF), support vector machine (SVM), k-nearest-neighbor (KNN), multilayer perceptron (MLP), and ridge regression (RR). In the validation cohort, the area under the curve (AUC) and accuracy were ascertained for every algorithm. Based on AUC values, the optimal algorithm was determined, and the features' importance was depicted. Results: A total of 1,082 female patients were included (age range, 12-96 years; mean age ± standard deviation, 42.22±13.37 years). The proportion of the 4 CBCs was 4% (44/1,185) for the B1 group, 60% (714/1,185) for the B2 group, 5% (57/1,185) for the B3 group, and 31% (370/1,185) for the B5 group. In the validation cohort, AUCs of the optimal algorithm constructed RF were 0.78, 0.88, 0.64, and 0.92 for B1, B2, B3, and B5, respectively, with an accuracy of 0.82. Conclusions: Machine learning could strongly predict CBC, particularly in B2 and B5 categories of solid breast tumors, with RF being the optimal machine learning model.
RESUMEN
OBJECTIVES: Perfluorobutane ultrasound contrast agent as a new type of contrast agent has a good performance in the diagnosis of hepatocellular carcinoma (HCC). This study aim to evaluate the accuracy and reliability of Perfluorobutane contrast-enhanced ultrasonography (P-CEUS) in the diagnosis of HCC with a systematic review and meta-analysis. METHODS: Web of Science, EMBASE, Cochrane, Clinical Key, Wan Fang, CBM and CNKI databases were systematically searched and checked for studies using P-CEUS in HCC, from 2007 to 2020. Data necessary to construct 2 × 2 contingency tables were extracted from included studies. The QUADAS tool was utilized to assess the methodologic quality of the studies. Meta-analysis included data pooling, subgroup analyses, meta-regression and investigation of publication bias was comprehensively performed. RESULTS: Nine studies were included in this meta-analysis and the overall diagnostic accuracy in characterization of HCC was as follows: pooled sensitivity, 0.90 (95% confidence interval: 0.82-0.95); pooled specificity, 0.97 (0.93-0.98); pooled positive likelihood ratio, 27.2 (14.1 to - 52.3); and pooled negative likelihood ratio, 0.10 (0.06-0.18). The area under the comprehensive receiving operation characteristic curve was 0.98 (0.97-0.99). CONCLUSION: The sensitivity and specificity of P-CEUS are more valuable than other imaging techniques (such as computer tomography or magnetic resonance imaging). However, due to the large differences in the data samples collected in this study, statistical heterogeneity results. P-CEUS can significantly improve the diagnostic efficiency of previous contrast-enhanced ultrasound for HCC. PROSPERO registration number: PROSPERO (CRD42020200040).