Search

The purpose of this study was to assess whether Immune CompleX Predictive Index (iXip) improves diagnostic accuracy of multiparametric prostate MRI (mpMRI) for clinically significant prostate cancer. This study included 72 patients (mean age: 68±8 years) with suspicion of prostate cancer and available iXip score. mpMRI images were evaluated by two radiologists according to the PI-RADS v2.1. Reference standard was based on fusion biopsy and standard transperineal 12-point biopsy. Diagnostic accuracy of iXip, mpMRI and their combination were calculated. Optimal cutoff of iXip with sensitivity and specificity was identified using the Youden index. Patients with clinically significant prostate cancers had significantly higher iXip values compared to patients without clinically significant prostate cancers (median 0.411 vs 0.273; p=0.026). The AUROC for iXip was 0.795 (95% CI 0.579-1.000, p=0.026). Sensitivity and specificity were 75% and 100% respectively for mpMRI alone, and 100% and 80% respectively for mpMRI combined with iXip > 0.375. The combination of mpMRI with a cutoff value of iXip > 0.375 has a very high sensitivity for the diagnosis of prostate cancer and a moderately high specificity.

Purpose The aim of our study was to evaluate the prevalence of early complications after Transcatheter Aortic Valve Implantation (TAVI) and their correlation with the Calcium Score (CS) of the aortic valve, aorta and ilio-femoral arteries derived from pre-procedural computed tomography (CT). Materials and methods We retrospectively reviewed 226 patients (100 males, mean age 79.4 ± 6.7 years) undergoing 64-slice CT for pre-TAVI evaluation from January 2018 to April 2021. The population was divided into CS quartiles. Results Overall, 173 patients underwent TAVI procedure, of whom 61% presented paravalvular leak after the procedure, 28% presented bleeding or vascular complications, 25% presented atrioventricular block, and 8% developed acute kidney injury. The prevalence of paravalvular leak and vascular complications was higher in the upper CS quartiles for aortic valve and ilio-femoral arteries. Conclusions Aortic valve and vascular CS could help to predict post-TAVI early complications.

The volume estimation of retroperitoneal sarcoma (RPS) is often difficult due to its huge dimensions and irregular shape; thus, it often requires manual segmentation, which is time-consuming and operator-dependent. This study aimed to evaluate two fully automated deep learning networks (ENet and ERFNet) for RPS segmentation. This retrospective study included 20 patients with RPS who received an abdominal computed tomography (CT) examination. Forty-nine CT examinations, with a total of 72 lesions, were included. Manual segmentation was performed by two radiologists in consensus, and automatic segmentation was performed using ENet and ERFNet. Significant differences between manual and automatic segmentation were tested using the analysis of variance (ANOVA). A set of performance indicators for the shape comparison (namely sensitivity), positive predictive value (PPV), dice similarity coefficient (DSC), volume overlap error (VOE), and volumetric differences (VD) were calculated. There were no significant differences found between the RPS volumes obtained using manual segmentation and ENet (p-value = 0.935), manual segmentation and ERFNet (p-value = 0.544), or ENet and ERFNet (p-value = 0.119). The sensitivity, PPV, DSC, VOE, and VD for ENet and ERFNet were 91.54% and 72.21%, 89.85% and 87.00%, 90.52% and 74.85%, 16.87% and 36.85%, and 2.11% and −14.80%, respectively. By using a dedicated GPU, ENet took around 15 s for segmentation versus 13 s for ERFNet. In the case of CPU, ENet took around 2 min versus 1 min for ERFNet. The manual approach required approximately one hour per segmentation. In conclusion, fully automatic deep learning networks are reliable methods for RPS volume assessment. ENet performs better than ERFNet for automatic segmentation, though it requires more time.

Magnetic Resonance Imaging-based prostate segmentation is an essential task for adaptive radiotherapy and for radiomics studies whose purpose is to identify associations between imaging features and patient outcomes. Because manual delineation is a time-consuming task, we present three deep-learning (DL) approaches, namely UNet, efficient neural network (ENet), and efficient residual factorized convNet (ERFNet), whose aim is to tackle the fully-automated, real-time, and 3D delineation process of the prostate gland on T2-weighted MRI. While UNet is used in many biomedical image delineation applications, ENet and ERFNet are mainly applied in self-driving cars to compensate for limited hardware availability while still achieving accurate segmentation. We apply these models to a limited set of 85 manual prostate segmentations using the k-fold validation strategy and the Tversky loss function and we compare their results. We find that ENet and UNet are more accurate than ERFNet, with ENet much faster than UNet. Specifically, ENet obtains a dice similarity coefficient of 90.89% and a segmentation time of about 6 s using central processing unit (CPU) hardware to simulate real clinical conditions where graphics processing unit (GPU) is not always available. In conclusion, ENet could be efficiently applied for prostate delineation even in small image training datasets with potential benefit for patient management personalization.

Acute pancreatitis (AP) represents a pancreas inflammation of sudden onset that can present different degrees of severity. AP is a frequent cause of acute abdomen and its complications are still a cause of death. Biliary calculosis and alcohol abuse are the most frequent cause of AP. Computed tomography (CT) and magnetic resonance imaging (MRI) are not necessary for the diagnosis of AP but they are fundamental tools for the identification of the cause, degree severity and AP complications. AP severity assessment is in fact one of the most important issue in disease management. Contrast-enhanced CT is preferred in the emergency setting and is considered the gold standard in patients with AP. MRI is comparable to CT for the diagnosis of AP but requires much more time so it is not usually chosen in the emergency scenario. Complications of AP can be distinguished in localized and generalized. Among the localized complications, we can identify: acute peripancreatic fluid collections (APFC), pseudocysts, acute necrotic collections (ANC), walled off pancreatic necrosis (WOPN), venous thrombosis, pseudoaneurysms and haemorrhage. Multiple organ failure syndrome (MOFS) and sepsis are possible generalized complications of AP. In this review, we focus on CT and MRI findings in local complications of AP and when and how to perform CT and MRI. We paid also attention to recent developments in diagnostic classification of AP complications.