A Multi-View Deep Learning Model for Thyroid Nodules Detection and Characterization in Ultrasound Imaging.

Thyroid Ultrasound (US) is the primary method to evaluate thyroid nodules. Deep learning (DL) has been playing a significant role in evaluating thyroid cancer. We propose a DL-based pipeline to detect and classify thyroid nodules into benign or malignant groups relying on two views of US imaging. Transverse and longitudinal US images of thyroid nodules from 983 patients were collected retrospectively. Eighty-one cases were held out as a testing set, and the rest of the data were used in five-fold cross-validation (CV). Two You Look Only Once (YOLO) v5 models were trained to detect nodules and classify them. For each view, five models were developed during the CV, which was ensembled by using non-max suppression (NMS) to boost their collective generalizability. An extreme gradient boosting (XGBoost) model was trained on the outputs of the ensembled models for both views to yield a final prediction of malignancy for each nodule. The test set was evaluated by an expert radiologist using the American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS). The ensemble models for each view achieved a mAP0.5 of 0.797 (transverse) and 0.716 (longitudinal). The whole pipeline reached an AUROC of 0.84 (CI 95%: 0.75-0.91) with sensitivity and specificity of 84% and 63%, respectively, while the ACR-TIRADS evaluation of the same set had a sensitivity of 76% and specificity of 34% (p-value = 0.003). Our proposed work demonstrated the potential possibility of a deep learning model to achieve diagnostic performance for thyroid nodule evaluation.

An Image-Rich Educational Review of Breast Pain.

Breast pain is extremely common, occurring in 70% to 80% of women. Most cases of breast pain are from physiologic or benign causes, and patients should be reassured and offered treatment strategies to alleviate symptoms, often without diagnostic imaging. A complete clinical history and physical examination is key for distinguishing intrinsic breast pain from extramammary pain. Breast pain without other suspicious symptoms and with a negative history and physical examination result is rarely associated with malignancy, although it is a common reason for women to undergo diagnostic imaging. When breast imaging is indicated, guidelines according to the American College of Radiology Appropriateness Criteria should be followed as to whether mammography, US, or both are recommended. This review article summarizes the initial clinical evaluation of breast pain and evidence-based guidelines for imaging. Additionally, the article reviews cyclical and noncyclical breast pain and provides an image-rich discussion of the imaging presentation and management of benign and malignant breast pain etiologies.

Quantitative Biomarkers Derived from a Novel Contrast-Free Ultrasound High-Definition Microvessel Imaging for Distinguishing Thyroid Nodules.

Low specificity in current ultrasound modalities for thyroid cancer detection necessitates the development of new imaging modalities for optimal characterization of thyroid nodules. Herein, the quantitative biomarkers of a new high-definition microvessel imaging (HDMI) were evaluated for discrimination of benign from malignant thyroid nodules. Without the help of contrast agents, this new ultrasound-based quantitative technique utilizes processing methods including clutter filtering, denoising, vessel enhancement filtering, morphological filtering, and vessel segmentation to resolve tumor microvessels at size scales of a few hundred microns and enables the extraction of vessel morphological features as new tumor biomarkers. We evaluated quantitative HDMI on 92 patients with 92 thyroid nodules identified in ultrasound. A total of 12 biomarkers derived from vessel morphological parameters were associated with pathology results. Using the Wilcoxon rank-sum test, six of the twelve biomarkers were significantly different in distribution between the malignant and benign nodules (all p < 0.01). A support vector machine (SVM)-based classification model was trained on these six biomarkers, and the receiver operating characteristic curve (ROC) showed an area under the curve (AUC) of 0.9005 (95% CI: [0.8279,0.9732]) with sensitivity, specificity, and accuracy of 0.7778, 0.9474, and 0.8929, respectively. When additional clinical data, namely TI-RADS, age, and nodule size were added to the features, model performance reached an AUC of 0.9044 (95% CI: [0.8331,0.9757]) with sensitivity, specificity, and accuracy of 0.8750, 0.8235, and 0.8400, respectively. Our findings suggest that tumor vessel morphological features may improve the characterization of thyroid nodules.

Spectrum of imaging findings in hyperplastic cholecystosis and potential mimics.

Adenomyomatosis and cholesterolosis of the gallbladder, collectively termed hyperplastic cholecystosis, are commonly encountered incidental findings on imaging studies performed for a variety of indications including biliary colic or nonspecific abdominal pain. These pathologies are rarely the source of symptoms, generally considered benign and do not require further work-up. However, their imaging characteristics can overlap with more sinister conditions that should not be missed. In this review, the imaging findings of adenomyomatosis and cholesterolosis will be reviewed followed by other gallbladder pathologies that might mimic these conditions radiologically. Important differentiating factors will be discussed that can aid the radiologist in making a more confident imaging diagnosis.

Bleeding Complications After Percutaneous Liver Biopsy: Do Subcapsular Lesions Pose a Higher Risk?

OBJECTIVE: The purpose of this study is to determine the incidence of moderate and severe bleeding complications after subcapsular liver mass biopsy and compare it with the incidences of such complications after nonsubcapsular liver mass biopsy and after random biopsy. MATERIALS AND METHODS: This retrospective review of ultrasound-guided liver biopsies was performed between January 2006 and February 2015. Biopsy type, needle size, the number of needle passes, and mass size and depth were recorded. A mass was categorized as subcapsular if it abutted the capsule and if the needle entered the mass directly without first traversing normal liver. Patients were observed postprocedurally, and the electronic medical record and findings from imaging performed within 1 month of the procedure were reviewed to assess for bleeding. Bleeding complications were categorized by size (with mild bleeding denoted by perihepatic blood ≤ 2 cm thick; moderate bleeding denoted by perihepatic blood > 2 cm thick, observation of blood elsewhere in the upper abdomen, or a combination of both findings; and severe bleeding denoted by blood extending into the lower abdomen, pelvis, or both). Bleeding rates were compared between groups using the Fisher exact test. RESULTS: Of a total of 1876 biopsies, 347 (18.5%) involved subcapsular masses, 760 (40.5%) involved nonsubcapsular masses, and 769 (41.0%) were random biopsies. There were no significant differences in biopsy type (p = 1.00), needle size (p = 0.12), or the number of needle passes (p = 0.10) when subcapsular and nonsubcapsular masses were compared. Thirteen moderate or severe bleeding complications (0.69%) occurred overall. The bleeding rate after subcapsular mass biopsy (3/347; 0.86%) was not statistically different from that noted after nonsubcapsular mass biopsy (5/760; 0.66%) (p = 0.71) or after random biopsy (5/769; 0.65%) (p = 0.71). CONCLUSION: Moderate and severe bleeding complications are rare after subcapsular liver mass biopsy, and their incidence is not significantly higher than that noted after nonsubcapsular mass biopsy or random biopsy. Biopsy of subcapsular lesions should no longer be considered contraindicated.