The most common cause of long-term disability and death in young adults is a traumatic brain injury. The decision for surgical intervention for craniotomy is dependent on the injury type and the patient's neurologic exam. The potential subtypes of intracranial hemorrhage that may necessitate surgical intervention include subdural hemorrhage, epidural hemorrhage, and intraparenchymal hemorrhage. We proposed a novel automatic method for segmenting the hemorrhage subtypes on a CT scan by integrated CT scan with bone window as input of a deep learning model. Brain CT scans were collected from adult patients and annotated regions of subdural hemorrhage, epidural hemorrhage, and intraparenchymal hemorrhage by neuroradiologists. Their raw DICOM images were preprocessed by two different window settings i.e., subdural and bone windows. The collected CT scans were divided into two datasets namely training and test datasets. A deep-learning model was modified to segment regions of each hemorrhage subtype. The model is a three-dimensional convolutional neural network including four parallel pathways that process the input at different resolutions. It was trained by a training dataset. After the segmentation result was produced by the deep-learning model, it was then improved in the post-processing step. The size of the segmented lesion was considered, and a region-growing algorithm was applied. We evaluated the performance of the proposed method on the test dataset. The method reached the median Dice similarity coefficients higher than 0.37 for each hemorrhage subtype. The proposed method demonstrates higher Dice similarity coefficients and improved segmentation performance compared to previously published literature.

Hepatocellular carcinoma (HCC) is the sixth most common cancer with a high mortality rate. Early diagnosis and treatment before tumor progression into an advanced stage is ideal. The current diagnosis of HCC is mainly based on imaging modalities such as ultrasound, computed tomography, and magnetic resonance imaging. These methods have some limitations including diagnosis in the case of very small tumors with atypical imaging patterns. Extracellular vesicles (EVs) are nanosized vesicles which have been shown to act as an important vector for cell‐to‐cell communication. In the past decade, EVs have been investigated with regard to their roles in HCC formation. Since these EVs contain biomolecular cargo such as nucleic acid, lipids, and proteins, it has been proposed that they could be a potential source of tumor biomarkers and a vector for therapeutic cargo. In this review, reports on the roles of HCC‐derived EVs in tumorigenesis, and clinical investigations using circulating EVs as a biomarker for HCC and their potential diagnostic roles have been comprehensively summarized and discussed. In addition, findings from in vitro and in vivo reports investigating the potential roles of EVs as therapeutic interventions are also presented. These findings regarding the potential benefits of EVs will encourage further investigations and may allow us to devise novel strategies using EVs in the early diagnosis as well as for treatment of HCC in the future.

Tan CH, et al. Korean J Radiol. 2022 Jul;23(7):697-719. https://doi.org/10.3348/kjr.2021.0593

Background: The accurate staging of liver fibrosis is crucial for managing chronic liver disease (CLD). Although magnetic resonance elastography (MRE) is the reference standard for noninvasive fibrosis assessment, its cost, specialized hardware, and operational demands restrict accessibility. In contrast, two-dimensional shear-wave elastography (2D-SWE) is more affordable, accessible, and widely integrated into routine ultrasound systems. Objective: Our aim was to determine the optimal 2D-SWE cut-offs for detecting significant fibrosis (≥F2) and evaluate its diagnostic performance across fibrosis stages. Methods: In this prospective study, 71 patients with suspected CLD underwent same-day MRE and 2D-SWE. MRE-defined cut-offs categorized fibrosis stages (≥3.5 kPa for significant fibrosis). Sensitivity, specificity, area under the receiver operating characteristic curve (AUROC), and likelihood ratios were calculated for various 2D-SWE thresholds. Results: At a 2D-SWE cut-off of 7.0 kPa, sensitivity for detecting ≥F2 fibrosis was 100% with a specificity of 85.7% and a positive likelihood ratio (LR+) of 7.0. Increasing the threshold to 8.0 kPa improved specificity to 91.8% while maintaining a sensitivity of 86.4% and achieving an AUROC of 0.89. For cirrhosis, a cut-off of 11.0 kPa achieved 100% sensitivity and 96.9% specificity. A 5.0 kPa cut-off reliably excluded abnormal stiffness with 89.1% sensitivity. Conclusions: Two-dimensional SWE is a reliable method for staging liver fibrosis. Thresholds of 7.0 kPa for screening significant fibrosis, 8.0 kPa for confirmation, and 11.0 kPa for diagnosing cirrhosis demonstrate high diagnostic accuracy. A 5.0 kPa cut-off effectively excludes abnormal liver stiffness.

The Fontan operation improves survival in patients with single ventricle physiology but is associated with Fontan-associated liver disease (FALD), characterized by progressive fibrosis due to prolonged elevated central venous pressure. While 2D shear wave elastography (2D-SWE) can assess fibrosis, it often overestimates stiffness in congestive conditions. Doppler ultrasound, which evaluates hepatic hemodynamics, may complement 2D-SWE for fibrosis assessment. This study evaluated the diagnostic performance of Doppler ultrasound and 2D-SWE in assessing hepatic fibrosis in Fontan patients and compared the findings with biopsy-proven fibrosis severity. A retrospective study was conducted on 27 Fontan patients who underwent Doppler ultrasound, 2D-SWE, and liver biopsy between January 2020 and December 2022. ROC curves and AUC values were used to assess diagnostic performance. AST to Platelet Ratio Index (APRI) and Fibrosis-4 index (FIB-4) scores demonstrated good discriminatory performance, with AUC values of 0.79 and 0.72, respectively. Resistive Index (RI) of hepatic artery showed moderate discriminatory performance (AUC = 0.62), while 2D-SWE demonstrated poor discriminatory ability (AUC = 0.35). When RI was combined with APRI, the AUC improved from 0.79 to 0.82. APRI and FIB-4 provided the most accurate assessment of significant fibrosis, while RI of hepatic artery may serve as a useful adjunct to serum biomarkers. Incorporating Doppler ultrasound into a multi-parametric model may improve fibrosis evaluation in Fontan patients.