Purpose The nonalcoholic fatty liver disease is the most common liver abnormality. Up to date, liver biopsy is the reference standard for direct liver steatosis quantification in hepatic tissue samples. In this paper we propose a neural network-based approach for nonalcoholic fatty liver disease assessment in ultrasound. Methods We used the Inception-ResNet-v2 deep convolutional neural network pre-trained on the ImageNet dataset to extract high-level features in liver B-mode ultrasound image sequences. The steatosis level of each liver was graded by wedge biopsy. The proposed approach was compared with the hepatorenal index technique and the gray-level co-occurrence matrix algorithm. After the feature extraction, we applied the support vector machine algorithm to classify images containing fatty liver. Based on liver biopsy, the fatty liver was defined to have more than 5% of hepatocytes with steatosis. Next, we used the features and the Lasso regression method to assess the steatosis level. Results The area under the receiver operating characteristics curve obtained using the proposed approach was equal to 0.977, being higher than the one obtained with the hepatorenal index method, 0.959, and much higher than in the case of the gray-level co-occurrence matrix algorithm, 0.893. For regression the Spearman correlation coefficients between the steatosis level and the proposed approach, the hepatorenal index and the gray-level co-occurrence matrix algorithm were equal to 0.78, 0.80 and 0.39, respectively. Conclusions The proposed approach may help the sonographers automatically diagnose the amount of fat in the liver. The presented approach is efficient and in comparison with other methods does not require the sonographers to select the region of interest.

Aims Despite growing interest in biomarkers application for risk evaluation in acute pulmonary embolism (APE), no decision-making levels have been defined. Methods and results We developed a biomarker-based risk stratification in 100 consecutive, normotensive on admission, APE patients (35 males, 65 females, 62±18 years). On admission serum NT-proBNP and cardiac troponin T (cTnT) levels were assessed and echocardiography was performed. All-cause 40-day mortality was 15% and APE mortality was 8%. In univariable analysis, cTnT>0.07 µg/L predicted all-cause mortality, hazard ratio (HR) 9.2 (95% CI: 3.3–26.1, P7600 ng/L predicted all-cause and APE mortalities [HR 6.7 (95% CI: 2.4–19.0, P=0.0003) and 7.3 (95% CI: 1.7–30.6, P=0.007)]. NT-proBNP0.07 µg/L was the most significant independent predictor, whereas NT-proBNP and systemic systolic blood pressure measured on admission and echocardiographic parameters were non-significant. APE mortality in patients with NT-proBNP≥600 ng/L and cTnT≥0.07 µg/L reached 33%. NT-proBNP<600 ng/L indicated group without deaths. APE mortality for patients with NT-proBNP≥600 ng/L and cTnT<0.07 µg/L was 3.7%. Incorporation of echocardiographic data did not improve group selection. Conclusion Simultaneous measurement of serum cTnT and NT-proBNP allows for precise APE prognosis. Normotensive patients on admission with cTnT≥0.07 µg/L and NT-proBNP≥600 ng/L are at high risk of APE mortality, whereas NTproBNP<600 ng/L indicates excellent prognosis.

Plasma brain natriuretic peptide (BNP), released from myocytes of ventricles upon stretch, has been reported to differentiate pulmonary from cardiac dyspnoea. Limited data have shown elevated plasma BNP levels in acute pulmonary embolism (APE), frequently accompanied by dyspnoea and right ventricular (RV) dysfunction. The aim of this study was to assess plasma N‐terminal proBNP (NT‐proBNP) in APE, and to establish whether it reflects the severity of RV overload and if it can be used to predict adverse clinical outcome.On admission, NT‐proBNP and echocardiography for RV overload were performed in 79 APE patients (29 males), aged 63±16 yrs.Plasma NT‐proBNP was elevated in 66 patients (83.5%) and was higher in patients with (median 4,650 pg·mL⁻¹(range 61–60,958)) than without RV strain (363 pg·mL⁻¹(16–16,329)). RV‐to-left ventricular ratio and inferior vena cava dimension correlated with NT‐proBNP. All 15 in-hospital deaths and 24 serious adverse events occurred in the group with elevated NT‐proBNP, while all 13 (16.5%) patients with normal values had an uncomplicated clinical course. Plasma NT‐proBNP predicted in-hospital mortality.Plasma N‐terminal pro-brain natriuretic peptide is elevated in the majority of cases of pulmonary embolism resulting in right ventricular overload. Plasma levels reflect the degree of right ventricular overload and may help to predict short-term outcome. Acute pulmonary embolism should be considered in the differential diagnosis of patients with dyspnoea and abnormal levels of brain natriuretic peptide.

In hemodialysis patients, a native arteriovenous fistula (AVF) is the preferred form of permanent vascular access. Despite recent improvements, vascular access dysfunction remains an important cause of morbidity in these patients. In this prospective observational cohort study, we evaluated potential risk factors for native AVF dysfunction. We included 68 patients with chronic renal disease stage 5 eligible for AVF construction at the Department of General and Vascular Surgery, Central Clinical Hospital Ministry of Internal Affairs, Warsaw, Poland. Patient characteristics and biochemical parameters associated with increased risk for AVF failure were identified using Cox proportional hazards models. Vessel biopsies were analyzed for inflammatory cells and potential associations with biochemical parameters. In multivariable analysis, independent predictors of AVF dysfunction were the number of white blood cells (hazard ratio [HR] 1.67; 95% confidence interval [CI] 1.24 to 2.25; p<0.001), monocyte number (HR 0.02; 95% CI 0.00 to 0.21; p = 0.001), and red blood cell distribution width (RDW) (HR 1.44; 95% CI 1.17 to 1.78; p16.2% was associated with a significantly reduced AVF patency frequency 24 months after surgery. Immunohistochemical analysis revealed CD45-positive cells in the artery/vein of 39% of patients and CD68-positive cells in 37%. Patients with CD68-positive cells in the vessels had significantly higher white blood cell count. We conclude that RDW, a readily available laboratory value, is a novel prognostic marker for AVF failure. Further studies are warranted to establish the mechanistic link between high RDW and AVF failure.