BackgroundStrain elastography can be purely qualitative or semiquantitative using both strain score and strain ratio. The aim of this study was to establish the accuracy of semiquantitative elastography using both strain score and strain ratio in differentiating benign from malignant breast masses. The diagnostic performance of the two methods was analysed for any statistically significant difference.MethodsA prospective study was carried out from May to December 2014 in the University of Nairobi, Department of Diagnostic Imaging and Radiation Medicine. One hundred and eighteen patients referred for breast ultrasound following clinical detection of masses certified the inclusion criteria. All solid masses identified on grey scale imaging were subjected to strain elastography. Elastographic findings were represented in both strain score and strain ratio. Comparison of diagnostic performance with histological findings as the gold standard for all detected solid masses was done. Fisher’s exact test and receiver operating characteristics curves were applied for statistical analysis to look for any significant differences between the diagnostic performance of strain score and strain ratio.ResultsOut of the 118, three patients did not attend for all the examinations and three biopsy results were misplaced therefore analysis was done for 112 subjects. The sensitivity, specificity, positive predictive value and negative predictive value of elasticity strain (Ueno) score were 0.86, 0.96, 0.89 and 0.96 respectively. For the strain ratio the values were 0.93, 0.96, 0.90 and 0.96 respectively. Fisher’s exact test P values comparing the sensitivity and specificity were 0.69 and 1.00 respectively not considered significant at p 0.05 levels. The areas under the curve (AUCs) from the receiver operating characteristic (ROC) curves were 0.972 and 0.976 for strain score and ratio respectively with a strong Pearson’s correlation coefficient, r 0.79 indicating a high diagnostic accuracy for both methods but no statistically significant difference in performance.ConclusionSemiquantitative ultrasound elastography has good diagnostic accuracy in differentiating benign and malignant breast solid lesions and there is no statistically significant difference between strain score and strain ratio in sensitivity, specificity and accuracy.Electronic supplementary materialThe online version of this article (doi:10.1186/s40644-016-0070-8) contains supplementary material, which is available to authorized users.

Background and Purpose. There is a wide variation in the anatomy of the CW in different individuals and population groups. The purpose of this study was to determine the proportion of variant anatomy of the circle of Willis (CW) and associated anomalies in patients with suspected cerebrovascular disease referred for cerebral computed tomography angiography (CTA) in two tertiary hospitals in Kenya. Methodology. This was a cross-sectional descriptive study conducted on 94 patients referred for cerebral CTA at the Kenyatta and Nairobi hospitals from August 2017 to February 2018. MIP and 3D reformatted images were analyzed by two senior radiologists to determine the final configuration of the CW and presence of vascular pathology. Vessels with diameters <0.8 mm were considered to be absent or hypoplastic. Chen et al. classification was used to determine the final configuration of CW. Results. Complete CW was seen in 37.2% (37.7% vs. 36.6% in males and females, respectively, p=0.909). Type A variant was the commonest anterior variant at 78.7%. Type E variant was the dominant posterior variant at 41.5%. Fetal PCA was demonstrated in 25.5%, unilateral > bilateral fetal PCA. Aneurysms were seen in 24.5% of patients. ACoA aneurysms were commonest at 43.6%. AVMs were seen in 8.5% of patients. Azygous ACA, fenestration, and duplication of vessels and persistent TA were not demonstrated. There is no significant association between aneurysms/AVMs and CW configuration. Conclusion. The findings in this study demonstrate slight differences in the CW configuration. A higher proportion of complete anterior CW was seen in female patients. Type A anterior and type E posterior circulation variants were the commonest in both males and females. No significant association was demonstrated between CW configuration and occurrence of aneurysms/AVMs in this study. Various methods of classification of CW configuration have been proposed with no standardized method to date, thus the need for consensus building between neuroanatomists, neurologists, neurosurgeons, and neuroradiologists for ease of comparison between similar future studies on the CW.

Background Elastography is an addition to grey-scale ultrasonic examination that has gained substantial traction within the last decade. Strain ratio (SR) has been incorporated as a semiquantitative measure within strain elastography, thus a potential imaging biomarker. The World Federation for Ultrasound in Medicine and Biology (WFUMB) published guidelines in 2015 for breast elastography. These guidelines acknowledge the marked variance in SR cut-off values used in differentiating benign from malignant lesions. The objective of this review was to include more recent evidence and seek to determine the optimal strain ratio cut off value for differentiating between benign and malignant breast lesions. Methods Comprehensive search of MEDLINE and Web of Science electronic databases with additional searches via Google Scholar and handsearching set from January 2000 to May 2020 was carried out. For retrieved studies, screening for eligibility, data extraction and analysis was done as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Diagnostic Test Accuracy (PRISMA-DTA) Statement guidelines of 2018. Quality and risk of bias assessment of the studies were performed using the revised Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. Results A total of 424 articles, 412 from electronic database and 12 additional searches were retrieved and 65 studies were included in the narrative synthesis and subgroup analysis. The overall threshold effect indicated significant heterogeneity among the studies with Spearman correlation coefficient of Logit (TPR) vs Logit (FPR) at − 0.301, p-value = 0.015. A subgroup under machine model consisting seven studies with 783 patients and 844 lesions showed a favourable threshold, Spearman’s correlation coefficient,0.786 (p = 0.036). Conclusion From our review, currently the optimal breast SR cut-off point or value remains unresolved despite the WFUMB guidelines of 2015. Machine model as a possible contributor to cut-off value determination was suggested from this review which can be subjected to more industry and multi-center research determination.

BackgroundCurrent advancements in dynamic contrast imaging of the liver have enabled increased sensitivity in the diagnosis of liver lesions. Evaluation and characterisation of the enhancement pattern of liver lesions in respect to the liver parenchyma aids in making a specific diagnosis.ObjectivesThe aim of this study was to determine the liver findings on dynamic contrast computed tomography (CT) scanning and correlate them with clinicopathologic findings.MethodsThis prospective cross-sectional study included 61 patients and took place between August 2017 and February 2018. Dynamic contrast CT was performed and the images were evaluated by two experienced radiologists. Correlation of the CT findings with histology results from an ultrasound-guided biopsy was done. Data analysis was performed using SPSS version 20.0.ResultsHepatocellular carcinoma (HCC) was the most common malignant lesion seen and showed three patterns of enhancement: homogenous, abnormal internal vessels and heterogeneous enhancement. Abnormal internal vessel pattern was most specific (90.6%) and showed a high positive predictive value (PPV) of 78.6%. Rapid washout showed a specificity of 87.5% and a PPV of 72.2% in the diagnosis of HCC. Dynamic contrast CT scan had a sensitivity of 93%, specificity of 50%, PPV of 91% and diagnostic accuracy of 95.5% in differentiation of benign and malignant liver lesions. Considering only Liver Imaging Reporting and Data System (LI-RADS) category 5 as conclusive for HCC diagnosis, our study did not miss a significant number of HCCs. Liver Imaging Reporting and Data System category 5 showed specificity of 81.3% and PPV of 75%.ConclusionEnhancement patterns on a dynamic contrast CT scan of the liver are useful in the interpretation of CT images for specific diagnoses.