Search

Abstract Focal liver lesions are frequently encountered during imaging studies, and hemangiomas represent the most common solid liver lesion. Liver hemangiomas usually show characteristic imaging features that enable characterization without the need for biopsy or follow-up. On the other hand, there are many benign and malignant liver lesions that may show one or more imaging features resembling hemangiomas that radiologists must be aware of. In this article we will review the typical imaging features of liver hemangiomas and will show a series of potential liver hemangiomas’ mimickers, giving radiologists some hints for improving differential diagnoses. Critical relevance statement The knowledge of imaging features of potential liver hemangiomas mimickers is fundamental to avoid misinterpretation. Key Points Liver hemangiomas typically show imaging features that enable avoiding a biopsy. Many benign and malignant liver lesions show imaging features resembling hemangiomas. Radiologists must know the potentially misleading imaging features of hemangiomas’ mimickers. Graphical Abstract

Abstract The standard method for identifying active Brown Adipose Tissue (BAT) is [18F]-Fluorodeoxyglucose ([18F]-FDG) PET/CT imaging, which is costly and exposes patients to radiation, making it impractical for population studies. These issues can be addressed with computational methods that predict [18F]-FDG uptake by BAT from CT; earlier population studies pave the way for developing such methods by showing some correlation between the Hounsfield Unit (HU) of BAT in CT and the corresponding [18F]-FDG uptake in PET. In this study, we propose training convolutional neural networks (CNNs) to predict [18F]-FDG uptake by BAT from unenhanced CT scans in the restricted regions that are likely to contain BAT. Using the Attention U-Net architecture, we perform experiments on datasets from four different cohorts, the largest study to date. We segment BAT regions using predicted [18F]-FDG uptake values, achieving 23% to 40% better accuracy than conventional CT thresholding. Additionally, BAT volumes computed from the segmentations distinguish the subjects with and without active BAT with an AUC of 0.8, compared to 0.6 for CT thresholding. These findings suggest CNNs can facilitate large-scale imaging studies more efficiently and cost-effectively using only CT.

Abstract Background and objectives There is an urgent need to discover blood-based biomarkers of multiple sclerosis (MS) to better define the underlying biology of relapses and monitor disease progression. The main goal of this study is to search for candidate biomarkers of MS relapses associated with circulating extracellular vesicles (EVs), an emerging tool for biomarker discovery. Methods EVs, purified from unpaired plasma and CSF samples of RRMS patients by size-exclusion chromatography (SEC), underwent proteomic analysis to discover novel biomarkers associated with MS relapses. The candidate biomarkers of disease activity were detected by comparison approach between plasma- and CSF-EV proteomes associated with relapses. Among them, a selected potential biomarker was evaluated in a cohort of MS patients, using a novel and highly reproducible flow cytometry-based approach in order to detect low abundant EV subsets in a complex body fluid such as plasma. Results The proteomic profiles of both SEC-purified plasma EVs (from 6 patients in relapse and 5 patients in remission) and SEC-purified CSF EVs (from 4 patients in relapse and 3 patients in remission) revealed a set of proteins associated with MS relapses significant enriched in the synaptic transmission pathway. Among common proteins, excitatory amino-acid transporter 2, EAAT2, responsible for the majority of the glutamate uptake in CNS, was worthy of further investigation. By screening plasma samples from 110 MS patients, we found a significant association of plasma EV-carried EAAT2 protein (EV-EAAT2) with MS relapses, regardless of disease-modifying therapies. This finding was confirmed by investigating the presence of EV-EAAT2 in plasma samples collected longitudinally from 10 RRMS patients, during relapse and remission. Moreover, plasma EV-EAAT2 levels correlated positively with Expanded Disability Status Scale (EDSS) score in remitting MS patients but showed a negative correlation with age in patients with secondary progressive (SPMS). Conclusion Our results emphaticize the usefulness of plasma EVs as a source of accessible biomarkers to remotely analyse the CNS status. Plasma EV-EAAT2 showed to be a promising biomarker for MS relapses. Further studies are required to assess the clinical relevance of this biomarker also for disability progression independent of relapse activity and transition from RRMS towards SPMS.

Abstract: The TIGIT+FOXP3+Treg subset (TIGIT+Tregs) exerts robust suppressive activity on cellular immunity and predisposes septic individuals to opportunistic infection. We hypothesized that TIGIT+Tregs could play an important role in intensifying the COVID-19 severity and hampering the defense against nosocomial infections during hospitalization. Herein we aimed to verify the association between the levels of the TIGIT+Tregs with the mechanical ventilation requirement, fatal outcome, and bacteremia during hospitalization. TIGIT+Tregs were immunophenotyped by flow cytometry from the peripheral blood of 72 unvaccinated hospitalized COVID-19 patients at admission from May 29th to August 6th, 2020. The patients were stratified during hospitalization according to their mechanical ventilation requirement and fatal outcome. COVID-19 resulted in a high prevalence of the TIGIT+Tregs at admission, which progressively increased in patients with mechanical ventilation needs and fatal outcomes. The prevalence of TIGIT+Tregs positively correlated with poor pulmonary function and higher plasma levels of LDH, HMGB1, FGL2, and TNF. The non-survivors presented higher plasma levels of IL-33, HMGB1, FGL2, IL-10, IL-6, and 5.54 times more bacteremia than survivors. Conclusions: The expansion of the TIGIT+Tregs in COVID-19 patients was associated with inflammation, lung dysfunction, bacteremia, and fatal outcome.

COVID-19 is a disease of dysfunctional immune responses, but the mechanisms triggering immunopathogenesis are not established. The functional plasticity of macrophages allows this cell type to promote pathogen elimination and inflammation or suppress inflammation and promote tissue remodeling and injury repair. During an infection, the clearance of dead and dying cells, a process named efferocytosis, can modulate the interplay between these contrasting functions. Here, we show that engulfment of SARS-CoV-2-infected apoptotic cells exacerbates inflammatory cytokine production, inhibits the expression of efferocytic receptors, and impairs continual efferocytosis by macrophages. We also provide evidence supporting that lung monocytes and macrophages from severe COVID-19 patients have compromised efferocytic capacity. Our findings reveal that dysfunctional efferocytosis of SARS-CoV-2-infected cell corpses suppresses macrophage anti-inflammation and efficient tissue repair programs and provides mechanistic insights for the excessive production of pro-inflammatory cytokines and accumulation of tissue damage associated with COVID-19 immunopathogenesis.

Artificial intelligence (AI) is revolutionizing plant sciences by enabling precise plant species identification, early disease diagnosis, crop yield prediction, and precision agriculture optimization. AI uses machine learning and image recognition to aid ecological research and biodiversity conservation. It plays a crucial role in plant breeding by accelerating the development of resilient, high-yielding crops with desirable traits. AI models using climate and soil data contribute to sustainable agriculture and food security. In plant phenotyping, AI automates the measurement and analysis of plant characteristics, enhancing our understanding of plant growth. Ongoing research aims to improve AI models’ robustness and interpretability while addressing data privacy and algorithmic biases. Interdisciplinary collaboration is essential to fully harness AI’s potential in plant sciences for a sustainable, food-secure future.

Abstract TB/HIV coinfection poses a complex public health challenge. Accurate forecasting of future trends is essential for efficient resource allocation and intervention strategy development. This study compares classical statistical and machine learning models to predict TB/HIV coinfection cases stratified by gender and the general populations. We analyzed time series data using exponential smoothing and ARIMA to establish the baseline trend and seasonality. Subsequently, machine learning models (SVR, XGBoost, LSTM, CNN, GRU, CNN-GRU, and CNN-LSTM) were employed to capture the complex dynamics and inherent non-linearities of TB/HIV coinfection data. Performance metrics (MSE, MAE, sMAPE) and the Diebold-Mariano test were used to evaluate the model performance. Results revealed that Deep Learning models, particularly Bidirectional LSTM and CNN-LSTM, significantly outperformed classical methods. This demonstrates the effectiveness of Deep Learning for modeling TB/HIV coinfection time series and generating more accurate forecasts.

Abstract Objective To assess the evolution of the COVID-19 pandemic in Brazil and its macro-regions, considering disease incidence and mortality rates, as well as identifying territories with still rising disease indices and evaluating vaccine coverage and population adherence to COVID-19 immunization. Methods An ecological study conducted in Brazil with COVID-19 cases and deaths reported between February 2020 and April 2024, obtained through the Coronavirus Panel. Historical series were constructed from incidence and mortality rates to assess the pandemic’s evolution, and temporal trends were estimated using the Seasonal Trend Decomposition using Loess (STL) method. The Spatial Variation in Temporal Trends (SVTT) technique was employed to identify clusters with significant variations in temporal trends. Vaccination was analyzed considering the percentage of vaccinated and unvaccinated population in each municipality of the country. Results Brazil recorded a total of 38,795,966 cases and 712,038 deaths from COVID-19 during the study period. Incidence and mortality rates showed three waves of the disease, with a fourth wave of smaller amplitude. Four clusters with significant case growth and two with increased deaths were identified. Vaccine coverage varied among municipalities, with some regions showing low vaccination rates and others with high immunization adherence. Conclusion The study provided a comprehensive overview of coronavirus behavior in Brazil, and its results highlight the ongoing importance of vaccination and the need to direct efforts and resources to areas of higher risk.

Abstract Soybean is the main oilseed cultivated worldwide. Even though Brazil is the world’s largest producer and exporter of soybean, its production is severely limited by biotic factors. Soil borne diseases are the most damaging biotic stressors since they significantly reduce yield and are challenging to manage. In this context, the present study aimed to evaluate the potential of a bacterial strain (Ag109) as a biocontrol agent for different soil pathogens (nematodes and fungi) of soybean. In addition, the genome of Ag109 was wholly sequenced and genes related to secondary metabolite production and plant growth promotion were mined. Ag109 showed nematode control in soybean and controlled 69 and 45% of the populations of Meloidogyne javanica and Pratylenchus brachyurus, respectively. Regarding antifungal activity, these strains showed activity against Macrophomia phaseolina, Rhizoctonia solani, and Sclerotinia sclerotiorum. For S. sclerotiorum, this strain increased the number of healthy plants and root dry mass compared to the control (with inoculation). Based on the average nucleotide identity and digital DNA-DNA hybridization, this strain was identified as Bacillus velezensis. Diverse clusters of specific genes related to secondary metabolite biosynthesis and root growth promotion were identified, highlighting the potential of this strain to be used as a multifunctional microbial inoculant that acts as a biological control agent while promoting plant growth in soybean.