OBJECTIVETo explore the correlation between pathological findings and mammographic features of microcalcification in calcified breast carcinoma without a mass.

METHODSThe morphology and distribution of the microcalcification lesions displayed by mammography were retrospectively analyzed in 108 cases of the calcified breast carcinoma without a mass in comparison with the pathological findings of the lesions.

RESULTSThe mammographic morphology or distribution of the microcalcification lesions did not differ significantly across different pathological types of calcified breast carcinoma without a mass (P>0.05). The microcalcification lesions showed no significant morphological difference between invasive and noninvasive breast carcinomas (P>0.05). Fine pleomorphic calcifications were frequently found in both invasive and noninvasive breast carcinomas, but fine linear and fine linear branching calcifications and mixed malignant calcifications were more common in invasive breast carcinoma. The distribution of the microcalcifications showed significantly different patterns between invasive and noninvasive breast carcinoma (P=0.006), characterized by segmental and cluttered distributions of the lesions, respectively.

CONCLUSIONThere is no specific mammographic features in correlation with the pathological types of microcalcification lesions in calcified breast carcinoma without a mass, but invasive and noninvasive calcified breast carcinomas have different mammographic features in the morphology and distribution of the microcalcifications to allow their preoperative differentiation.

Breast Neoplasms ; diagnostic imaging ; pathology ; Calcinosis ; diagnostic imaging ; pathology ; Carcinoma, Ductal, Breast ; diagnostic imaging ; pathology ; Female ; Humans ; Mammography ; methods ; Retrospective Studies

Objective:To construct and evaluate a parotid mass malignancy risk model based on ultrasound image characteristics and clinical features of parotid masses.Methods:Ultrasound images and clinical features of 214 patients with parotid masses in the First People′s Hospital of Foshan were retrospectively collected from June 2018 to August 2020. The pathology results were taken as the golden standard. All the clinical features and ultrasound image features were first screened using regression analysis, and then the screened features were used to build a prediction model.Results:Malignant tumors of the parotid gland appeared on ultrasound as hypoechoic solid masses with or without abnormal cervicofacial lymph nodes with poorly defined borders and irregular morphology. Multifactorial analysis showed that facial nerve function, cervicofacial lymph node abnormalities, maximum diameter, morphology and borders of the mass were independent predictors of the risk of malignant parotid masses. A Nomogram prediction model was established using the above 5 indicators, and the results showed a concordance index(C-index) of 0.896 (95% CI=0.834-0.958) for Nomogram. The standard curve showed good agreement between the predictive effect of Nomogram and the actual situation of benign and malignant parotid swellings, with an internally validated C-index of 0.878. Conclusions:Ultrasound is of great value in identifying benign and malignant parotid tumors. The Nomogram model using ultrasound image features and clinical characteristics can assess the biocharacteristics of parotid masses, and the model shows high accuracy in predicting the risk of malignancy of parotid masses.

Objective: To assess the diagnostic performance of contrast-enhanced spectral mammography (CESM) in suspected breast lesions. Methods: A total of 97 patients with suspected breast cancer identified by clinical examination or screening underwent two-views CESM examination on the basis of digital breast tomosynthesis (DBT) combined with full-field digital mammography (FFDM), and they were finally confirmed by biopsy or pathology. Three senior radiologists analyzed images, including lesion visibility, lesion characteristics, enhancement type, degree of enhancement, BIRDS classification, etc. Finally, based on the pathology, we compared the CESM+DBT+FFDM and DBT+FFDM two models according to sensitivity, specificity and ROC for diagnostic performance. Results: There were a total of 120 lesions. Eighty-nine lesions were malignant, 31 benign; CESM was not enhanced in 2 cases, mild enhancement was performed in 22 cases, moderately intensive in 15 cases, highly intensive in 81 cases, and 2 cases were not enhanced; mass-enhanced in 96 cases, including ring-enhanced in 12 cases, 22 cases of non-mass type. The sensitivities of the combination of CESM and not combination of CESM were 91.0% and 80.9%, respectively, and the specificities were 93.5% and 87.1%, respectively. The area under the ROC curve of combination of CESM was higher than the without combination of CESM (0.923 and 0.900, P<0.05), The difference was statistically significant. Conclusion For suspicious lesions, CESM examination can improve the diagnosis accuracy of breast cancer.

Objective To evaluate the diagnostic performance of digital breast tomosynthesis (DBT) breast X-ray photography image texture characteristics based deep learning classification model on differentiating malignant masses. Methods Retrospectively collected 132 cases with simplex breast lesions (89 benign lesions and 43 malignant lesions) which were confirmed by pathology and DBT during January 2016 to December 2016 in Nanfang Hospital. DBT was performed before biopsy and surgery. Image of cranio-caudal view (CC) and medio-lateral oblique (MLO) were captured. The lesion area was segmented to acquire ROI by ITK-SNAP software. Then the processed images were input into MATLAB R2015b to establish a feature model for extracting texture features. The characteristics with high correlation was analyzed from Fisher score and one sample t test. We built up support vector machine (SVM) classification model based on extracted texture and added neural network model (CNN) for deep learning classification model. We randomly assigned collected cases into training group and validation group. The diagnosis of benign and malignant lesions were served as the reference. The efficiency was evaluated by ROC classification model. Result We extracted 82 texture characteristics from 132 images of leisure (132 images of CC and 132 images of MLO) by establishing deep learning classification model of breast lesions. We randomly chose and combined characteristics from 15 texture characteristics with statistical significance, then differentiated benign and malignant by SVM classification model. After 50 iterations on each combination of characteristics, the average diagnostic efficacy was compared to obtained the one with higher efficacy. Nine of CC and 8 of MLO was selected. The result showed that the sensitivity, specificity, accuracy and area under curve (AUC) of the model to differentiate simplex breast lesions for CC were 0.68, 0.77, 0.74 and 0.74, for MLO were 0.71, 0.71, 0.71 and 0.76. Conclusions MLO has better diagnostic performance for the diagnosis than CC. The deep learning classification model on breast lesions which was built upon DBT image texture characteristics on MLO could differentiate malignant masses effectively.