Intelligent Diagnosis of Thyroid Ultrasound Imaging Using an Ensemble of Deep Learning Methods.

Background and Objectives: At present, thyroid disorders have a great incidence in the worldwide population, so the development of alternative methods for improving the diagnosis process is necessary. Materials and Methods: For this purpose, we developed an ensemble method that fused two deep learning models, one based on convolutional neural network and the other based on transfer learning. For the first model, called 5-CNN, we developed an efficient end-to-end trained model with five convolutional layers, while for the second model, the pre-trained VGG-19 architecture was repurposed, optimized and trained. We trained and validated our models using a dataset of ultrasound images consisting of four types of thyroidal images: autoimmune, nodular, micro-nodular, and normal. Results: Excellent results were obtained by the ensemble CNN-VGG method, which outperformed the 5-CNN and VGG-19 models: 97.35% for the overall test accuracy with an overall specificity of 98.43%, sensitivity of 95.75%, positive and negative predictive value of 95.41%, and 98.05%. The micro average areas under each receiver operating characteristic curves was 0.96. The results were also validated by two physicians: an endocrinologist and a pediatrician. Conclusions: We proposed a new deep learning study for classifying ultrasound thyroidal images to assist physicians in the diagnosis process.

New insights into IL-17/IL-23 signaling in ankylosing spondylitis (Review).

Ankylosing spondylitis (AS) is a progressive common autoimmune inflammatory disease, part of the spondylarthritis group, characterized, besides clinical spinal and peripheral joint inflammation, by enthesitis and new bone formation, that can lead to severe functional impairment. Beyond intensive and continuous research on the pathogenic process extensively performed in recent years, their impact on therapeutic management remains open to future development. Better knowledge of AS pathogenesis have shown results progressively and studies are being performed to advance our current understanding of the disease. It is well known that tumor necrosis factor (TNF) exerts a central role, along with interleukin-17 (IL-17) and interleukin-23 (IL-23), demonstrated by several clinical studies. Similar to other rheumatic inflammatory conditions, SA is associated with an early process of systemic bone loss, both trabecular and cortical, consecutive osteopenia, osteoporosis, and high fracture risk. Current personalized therapeutic options benefit from new published data, to prevent future complications and to improve quality of life.

JAK/STAT pathway in pathology of rheumatoid arthritis (Review).

Rheumatoid arthritis (RA) is classified as an inflammatory, chronic autoimmune and disabling disease based on the intricate interplay between environmental and genetic factors. With a prevalence ranging from 0.3 to 1%, RA is the most prevalent inflammatory joint disease observed in adults. Disruption of immune tolerance becomes evident when abnormal stimulation of the innate and adaptive immune system occurs. This cascade of events causes persistent joint inflammation, proliferative synovitis and, ultimately, damage of the underlying cartilage as well as the subchondral bone, leading to permanent joint destruction, deformity and subsequent loss of function. With cytokines being the key to a multitude of biological processes, including inflammation, hematopoiesis and overall immune response, one must inevitably look at the main pathways through which a significant number of those molecules exert their function. Janus kinase/signal transducers and activators of transcription (JAK/STATs) represent one such pathway and, recently, JAK inhibitors (JAKinibs) have shown promise in the treatment of several inflammatory diseases, including RA. This narrative review focuses on the intricate signaling pathways involved as well as on the clinical aspects and safety profiles of JAKinibs approved for the treatment of RA.

Host immune response in chronic hepatitis C infection: involvement of cytokines and inflammasomes.

Chronic liver disease is a major health issue worldwide and chronic hepatitis C (CHC) is associated with an increased risk of cirrhosis and hepatocellular carcinoma (HCC). There is evidence that the hepatitis C virus (HCV) infection is correlated with immune senescence by way of immune activation and chronic inflammation, which lead to increased metabolic and cardiovascular risk, as well as progressive liver damage. Both the innate and adaptive immunity are firmly tied to the prognosis of an infection with HCV and its response to antiviral therapy. HCV is therefore associated with increased pro-inflammatory status, heightened production of cytokines, prolonged systemic inflammation, as well as increased morbidity and mortality, mainly due to the progression of hepatic fibrosis and HCC, but also secondary to cardiovascular diseases. Viral hepatic pathology is increasingly considered a disease that is no longer merely limited to the liver, but one with multiple metabolic consequences. Numerous in vitro studies, using experimental models of acute or chronic inflammation of the liver, has brought new information on immunopathological mechanisms resulting from viral infections and have highlighted the importance of involving complex structures, inflammasomes complex, in these mechanisms, in addition to the involvement of numerous proinflammatory cytokines. Beyond obtaining a sustained viral response and halting the aforementioned hepatic fibrosis, the current therapeutic "treat-to-target" strategies are presently focused on immune-mediated and metabolic disorders, to improve the quality of life and long-term prognosis of CHC patients.

Predictors of early and sustained virological response of viral hepatitis C.

Due to complex interplay between host and viral factors, pathogenesis of chronic hepatitis C (CHC) is considered a challenging issue. Infection with hepatitis C virus (HCV) is not confined only to liver but can induce disturbances in many other organs and systems. Our primary aim for this study was to evaluate biological response rates and sustained virological response (SVR) in patients diagnosed with CHC, treated with Interferon-alpha (IFN-α), Pegylated (PEG)-IFN-α2a or -α2b plus Ribavirin. The second aim of the study was the identification of predictive factors for a favorable response to antiviral therapy in patients diagnosed with CHC. We enrolled in this study 210 patients diagnosed with CHC who have accomplished all inclusion and exclusion criteria, treated with PEG-IFN plus Ribavirin. Patients' recovery progress has been evaluated by determining: age, gender; biochemical tests: alanine aminotransferase (ALT), aspartate aminotransferase (AST); serological assays - detect anti-HCV antibody and molecular assays - detect, quantify and/or characterize hepatitis C viral load (ribonucleic acid) (HCV-RNA); liver histopathological (HP) examination. According to their response to treatment, they were classified into responders (n=145) and non-responders (n=65). Liver biopsies were histopathologically evaluated for necroinflammatory grade and fibrosis stage according to the modified Ishak and Metavir scoring systems for chronic hepatitis. Demographic, laboratory, and HP results were introduced in statistical analysis. These parameters were included in area under curve (AUC) analysis in order to estimate their degree of influence on getting early virological response (EVR) and SVR. Our study demonstrates that factors connected to treatment failure in CHC are linked to older age, high hepatitis C viral load, and impaired glucose tolerance at beginning of treatment [high fasting glucose and insulin, high homeostatic model assessment of insulin resistance (HOMA-IR) index] and also to liver histology features (high fibrosis score, liver steatosis, iron infiltration, and more or less high necroinflammatory activity). Analyzing results of our study shows that HOMA-IR index, serum insulin levels, baseline HCV-RNA, baseline mean blood glucose and HP score like Ishak fibrosis score, steatosis score and liver iron score may have a predictive value for obtaining an EVR in patients diagnosed with CHC.

The impact of liver steatosis on early and sustained treatment response in chronic hepatitis C patients.

Steatosis is a frequent feature of hepatitis C virus (HCV) infection. Steatosis may be an important cofactor in both accelerating fibrosis and increasing liver necroinflammatory activity in chronic hepatitis C. The main objective of this study was the evaluation of biological response rates, early viral response, sustained viral response in patients with chronic hepatitis C treated with Interferon-alpha (IFN-α), Pegylated (PEG)-IFN-α2a or -α2b plus Ribavirin and to relate it to the presence of hepatic steatosis. There were selected to take part to the research 210 patients with chronic hepatitis C who have fulfilled all inclusion and exclusion criteria and were treated with PEG-IFN plus Ribavirin. Patients' progress has been monitored by determining next parameters: age, gender; biochemical tests - alanine aminotransferase (ALT), aspartate aminotransferase (AST); serological assays - detect antibody to hepatitis C virus (anti-HCV); molecular assays - detect, quantify and/or characterize HCV-RNA; liver histopathological examination. Steatosis was graded using the Brunt system. These parameters were included in an area under curve (AUC) analysis. Purpose is to estimate their degree of influence on getting early viral response (EVR) and sustained viral response (SVR). Based on the obtained results, it appears that initial value of HCV-RNA, dVL parameter value (low relative percentage of viral load during the first 12 weeks of treatment), histological scores steatosis may be predictive in the viral response in chronic hepatitis C. Our research demonstrates that a high degree of liver steatosis impairs both EVR and SVR in chronic hepatitis C treated with standard PEG-IFN and Ribavirin for 48 weeks and that a steatosis score of ≤3 predicts EVR with a sensibility of 91.03% with specificity of 21.54%.

The cadherin switch assessment in the epithelial-mesenchymal transition of urothelial bladder carcinomas.

INTRODUCTION: The epithelial-mesenchymal transition (EMT) process is a complex molecular mechanism that is involved in the acquisition of an aggressive, invasive and metastatic phenotype by carcinomas. The cadherin switch consists in the alteration of E-cadherin and N-cadherin expression and is specific for the EMT process. MATERIALS AND METHODS: This study included 35 cases of primitive urothelial carcinomas investigated in relation with clinicopathological prognostic parameters and expression of E- and N-cadherins in the advancing edge and intratumoral compartments. RESULTS: In both compartments, the immunoexpression of E-cadherin decreased, while that of N-cadherin increased in high grade, deeply invasive, or those cases with lymph node metastases and advanced stages carcinomas, with a negative linear correlation observed between their expression percentage values. In this study, it was observed the presence of cadherin switch in urothelial carcinomas, the variation of the two proteins' immunostaining patterns being higher at the advancing edge. The presence of N-cadherin in intratumoral compartment designated it as actively involved in EMT process. CONCLUSIONS: The analysis of cadherins switch can be used to identify superficial urothelial carcinoma with invasion and metastasis potential.