Search

Background: Chest computed tomography (CT) scan is one of the most common tools used for the diagnosis of patients with coronavirus disease 2019 (COVID-19). While segmentation of COVID-19 lung lesions by radiologists can be time-consuming, the application of advanced deep learning techniques for automated segmentation can be a promising step toward the management of this infection and similar diseases in the future. Objectives: This study aimed to evaluate the performance and generalizability of deep learning-based models for the automated segmentation of COVID-19 lung lesions. Patients and Methods: Four datasets (2 private and 2 public) were used in this study. The first and second private datasets included 297 (147 healthy and 150 COVID-19 cases) and 82 COVID-19 subjects. The public datasets included the COVID19-P20 (20 COVID-19 cases from 2 centers) and the MosMedData datasets (50 COVID-19 patients from a single center). Model comparisons were made based on the Dice similarity coefficient (DSC), receiver operating characteristic (ROC) curve, and area under the curve (AUC). The predicted CT severity scores by the model were compared with those of radiologists by measuring the Pearson's correlation coefficients (PCC). Also, DSC was used to compare the inter-rater agreement of the model and expert against that of 2 experts on an unseen dataset. Finally, the generalizability of the model was evaluated, and a simple calibration strategy was proposed. Results: The VGG16-UNet model showed the best performance across both private datasets, with a DSC of 84.23% ± 1.73% on the first private dataset and 56.61% ± 1.48% on the second private dataset. Similar results were obtained on public datasets, with a DSC of 60.10% ± 2.34% on the COVID19-P20 dataset and 66.28% ± 2.80% on a combined dataset of COVID19-P20 and MosMedData. The predicted CT severity scores of the model were compared against those of radiologists and were found to be 0.89 and 0.85 on the first private dataset and 0.77 and 0.74 on the second private dataset for the right and left lungs, respectively. Moreover, the model trained on the first private dataset was examined on the second private dataset and compared against the radiologist, which revealed a performance gap of 5.74% based on DSCs. A calibration strategy was employed to reduce this gap to 0.53%. Conclusion: The results demonstrated the potential of the proposed model in localizing COVID-19 lesions on CT scans across multiple datasets; its accuracy competed with the radiologists and could assist them in diagnostic and treatment procedures. The effect of model calibration on the performance of an unseen dataset was also reported, increasing the DSC by more than 5%. [ABSTRACT FROM AUTHOR]

Background: Breast microcalcifications are a category of lesions that can lead to malignancies. They remain a major concern in imaging of suspected cases. Vacuum-assisted biopsy (VAB) has been proposed as a safe and effective measure to evaluate microcalcifications. Objectives: The present study aimed to assess the results of VAB for breast microcalcifications. Patients and Methods: This cross-sectional study was conducted on patients with microcalcifications detected on mammograms. Patients were recruited through simple random sampling during 2019 - 2020, based on the inclusion criteria. The inclusion criteria were microcalcifications on mammography, classified as the breast imaging-reporting and data system (BI-RADS) categories 3, 4B, 4C, and 5, and undergoing mammography-guided VAB for the microcalcifications. Patients with missing data and those who could not be followed-up for at least six months were excluded from the study. They were assessed regarding different imaging characteristics of lesions, including the breast density, BI-RADS classification, microcalcification distribution and morphology, and other demographic and clinical features before biopsy. Additionally, the results of stereotactic-guided VAB were assessed for various benign and malignant pathologies of microcalcifications. The results of descriptive and analytical tests for various radiological and pathological features of lesions were reported. Results: A total of 257 patients, with a mean age of 50.3 ± 8.3 yars, were included in this study. Almost half of the patients (n = 125, 48.6%) had a C-grade breast composition. Regarding the BI-RADS classification, 206 (80.2%) patients were diagnosed with 4B lesions, followed by 44 (17.1%) patients with 4C lesions. The assessment of the pattern of microcalcification distribution in imaging showed that more than half of the patients (n = 148, 57.6%) had lesions in multiple groups. The most prevalent morphology of microcalcifications was punctate amorphous (n = 109, 42.4%). The majority of patients (n = 180, 70%) had benign findings in the pathological assessment, and only 69 (26.8%) had malignant features in pathology. The distribution of malignancies differedamong various BI-RADS categories. In the 4B category of lesions, there were 166 benign lesions versus 32 malignant lesions, while in the 4C category, there were 10 benign lesions versus 34 malignant lesions (P < 0.001). Conclusion: This study described the findings of successful stereotactic-guided VAB for breast microcalcifications. VAB can be implemented as a promising assessment tool to evaluate suspected breast microcalcifications effectively. [ABSTRACT FROM AUTHOR]

Background: Transvaginal Ultrasonography is a noninvasive and inexpensive medical imaging tool used for the diagnosis of various diseases. Objectives: To identify an effective method to identify high-risk patients for developing malignancy after molar evacuation. Patients and Methods: A prospective serial assessment of 19 patients with gestational trophoblastic disease was performed. Clinical and laboratory data, transvaginal ultrasound and Doppler findings were evaluated the day before evacuation. They were followed-up in the first week after evacuation and every two weeks during the next two months, then every month until the sixth month. Results: Ovarian theca lutein cysts (P = 0.018) (among pre-evacuation factors) and first week ultrasound (P =0.02) can help in detecting high-risk patients. Even though, when β-hCG titer is not available in a high-risk patient, post evacuation myometrial involvement (P = 0.005) is a useful sign for detecting persistency. Conclusions: Some ultrasonographic features of molar pregnancy have capability to predict malignancy in the course of disease. [ABSTRACT FROM AUTHOR]