Browse Articles
Discover research articles across all indexed journals
AI-Assisted identification of sex-specific patterns in diabetic retinopathy using retinal fundus images
Diabetic retinopathy (DR) is a microvascular complication of diabetes that can lead to blindness if left untreated. Regular monitoring is crucial for detecting early signs of referable DR, and the progression to moderate to severe non-proliferative DR, proliferative DR (PDR), and macular edema (ME), the most common cause of vision loss in DR. Currently, aside from considerations during pregnancy, sex is not factored into DR diagnosis, management or treatment. Here we examine whether DR manifests differently in male and female patients, using a dataset of retinal images and leveraging convolutional neural networks (CNN) integrated with explainable artificial intelligence (AI) techniques. To minimize confounding variables, we curated 2,967 fundus images from a larger dataset of DR patients acquired from EyePACS, matching male and female groups for age, ethnicity, severity of DR, and hemoglobin A1C levels. Next, we fine-tuned two pre-trained VGG16 models—one trained on the ImageNet dataset and another on a sex classification task using healthy fundus images—achieving AUC scores of 0.72 and 0.75, respectively, both significantly above chance level. To uncover how these models distinguish between male and female retinas, we used the Guided Grad-CAM technique to generate saliency maps, highlighting critical retinal regions for correct classification. Saliency maps showed CNNs focused on different retinal regions by sex: the macula in females, and the optic disc and peripheral vasculature along the arcades in males. This pattern differed noticeably from the saliency maps generated by CNNs trained on healthy eyes. These findings raise the hypothesis that DR may manifest differently by sex, with women potentially at higher risk for developing ME, as opposed to men who may be at greater risk for PDR.
Giardia duodenalis stabilizes HIF-1α and induces glycolytic alterations in intestinal epithelial cells
Environmental politics is doomed to fail — unless we tell better stories
ITGAV as a promising diagnostic, immunological, and prognostic biomarker in pan-cancer
Abstract Integrin αV (ITGAV) plays a key role in cell adhesion, migration, and immune regulation, and is implicated in tumor progression. However, its comprehensive expression profile and functional relevance across different cancers remain poorly understood. We conducted an integrative pan-cancer analysis of ITGAV using data from TCGA, GTEx, CCLE, and other public databases. Expression, diagnostic value (via ROC analysis), and prognostic significance (via Cox and Kaplan–Meier analyses of OS, DSS, PFS, and DFS) were assessed. We further explored ITGAV’s correlation with immune cell infiltration and immune-related genes, its predictive role in immunotherapy response based on immunophenoscore (IPS), and its drug-binding potential through molecular docking. (1) ITGAV was significantly overexpressed in multiple cancer types including LIHC, COAD, and STAD. (2) ROC analysis confirmed its strong diagnostic value, particularly in HNSC, UCEC, and ESCA. (3) High ITGAV expression was associated with poorer survival outcomes in most cancers, while a protective role was observed in KIRC. (4) ITGAV expression was positively correlated with immune cell infiltration and co-expressed with immune-activating and immunosuppressive genes. (5) The expression level of ITGAV correlates with the IPS score, suggesting its predictive value for the benefit of immunotherapy. (6) Molecular docking identified strong binding affinities between ITGAV and six candidate compounds, including gemcitabine and pioglitazone. Our findings demonstrate that ITGAV is a promising biomarker for diagnosis, prognosis, and immunotherapy prediction across cancers. Its immunological associations and druggability highlight its potential as a candidate therapeutic target.
Machine learning derived development and validation of extracellular matrix related signature for predicting prognosis in adolescents and young adults glioma
Abstract The mortality rates have been increasing for glioma in adolescents and young adults (AYAs, aged 15–39 years). However, current biomarkers for clinical assessment in AYAs glioma are limited, prompting the urgent need for identifying ideal prognostic signature. Extracellular matrix is involved in the development of tumors, while their prognostic significance in AYAs glioma remains unclear. By an integrated machine learning workflow and circuit training and validation procedure, we developed a machine learning-derived prognostic signature (MLDPS) based on 1,026 extracellular matrix-related genes and 3 AYAs glioma cohorts. MLDPS exhibited robust and consistent predictive performance in overall survival and could serve as an independent prognostic factor for AYAs glioma. Simultaneously, MLDPS outperformed previous 89 published prognostic signatures and traditional clinical characteristics, confirming the robust predictive capability. Besides, MLDPS had the potential to stratify prognosis in patients with other cancer types. In addition, the tumor microenvironment between high and low MLDPS groups displayed different patterns while more tumor-infiltrating immune cells were observed in high MLDPS group. Additionally, patients in low MLDPS group had significantly prolonged survival when received immunotherapy in cancers including glioblastoma, urothelial carcinoma and melanoma. Overall, our study proposes a promising signature, which can be utilized for clinicians to evaluate prognosis and might provide individualized clinical management for AYAs glioma.
To save lives in heatwaves, focus on how human bodies work
Intelligent text analysis for effective evaluation of english Language teaching based on deep learning
We need a new ethics for a world of AI agents
Exploring the effectiveness of virtual reality-based training for sustainable health and occupational safety in industry 4.0
This company claimed to ‘de-extinct’ dire wolves. Then the fighting started
Spontaneously assembled cellulose nanocrystal structural color films with tunable properties
Top 10 drugs most frequently associated with adverse events of myocarditis and pericarditis
Quantity and morphology of microplastics in the Tehran and Nowshahr MSW incinerators ashes
Geometric principles of wobble board design for balance training and rehabilitation
Agent-based modelling of the early stages of actin polymerisation required to drive endocytosis in Saccharomyces cerevisiae
Abstract Endocytosis is critical. Its complexity means that many aspects remain poorly understood. We have developed an agent-based model covering key components of actin filament generation in endocytosis in Saccharomyces cerevisiae. The model incorporates realistic values for rates, affinities, concentrations, and mobilities, and reproduces essential features of endocytosis, from the arrival of WASp/Las17 and its inhibitor Sla1 at the membrane up to the burst of actin polymerisation. The model yields relative rates and affinities for interactions that cannot be measured experimentally, and places limitations on plausible scenarios. Specifically, it reveals three novel findings. First, Las17 must form multimeric complexes. Second, de novo F-actin nucleation occurs in two stages, involving the slow formation of linear trimers, followed by rapid polymerisation once an additional actin monomer is positioned at the side of the aligned monomers. Third, competition between SH3 domains and other factors, including actin, is critical to ensure on/off switching. This requires: (1) tandem domains binding to adjacent polyproline sites outcompeting single domains; (2) these tandem domains being weakened in overall affinity through a reduction in avidity by competition with single SH3 domains. We conclude with a pathway that proposes how controlled actin polymerisation occurs, and raises implications for further testing.
Ambulatory physiological measures obtained under naturalistic urban mobility conditions have acceptable reliability
Abstract Ambulatory assessment methods in psychology and clinical neuroscience are powerful research tools for collecting data outside of the laboratory. These methods encompass physiological, behavioral, and self-report measures obtained while individuals navigate in real-world environments, thereby increasing the ecological validity of experimental approaches. Despite the recent increase in applications of ambulatory physiology, data on the reliability of these measures is still limited. To address this issue, twenty-six healthy participants ( N = 15 female, 18–34 years) completed an urban walking route (distance M = 2.2 km, ±SD = 0.11; duration M = 30.8 min, ±SD = 1.34; temperature M = 18.34° degree Celsius, ±SD = 1.19, Range = 16°-21° degrees Celsius) on two separate testing days, while assessing the effect of metabolic state (sated vs. fasted). GPS-location and ambulatory physiological measures (cardiovascular and electrodermal activity) were continuously recorded. The results showed no significant differences in single physiological measures between fasted and sated states. Bootstrapped test-retest reliabilities of single measures and aggregate scores derived via principal component analysis (PCA) were computed. The first principal component (PC#1) accounted for 39–45% of variance across measures. PC#1 scores demonstrated an acceptable test-retest reliability ( r = .60) across testing days, exceeding the reliabilities of most individual measures (heart rate: r = .53, heart rate variability: r = .50, skin conductance level: r = .53, no. of skin conductance responses: r = .28, skin conductance response amplitude: r = .60). Results confirm that ambulatory physiological measures recorded during naturalistic navigation in urban environments exhibit acceptable test-retest reliability, in particular when compound scores across physiological measures are analyzed, a prerequisite for applications in (clinical) psychology and digital health.
Efficacy and safety of taxane plus ramucirumab for advanced gastric cancer after chemotherapy plus nivolumab
Intricate origins of ice mummy’s ink revealed
Analysis of effect of thickness and surface treatment on sound transmission loss characteristics of natural fibres
Abstract Noise pollution, driven by rapid urbanisation and city expansion, has created a growing demand for innovative and bio-degradable sound absorption materials. Traditional solutions such as synthetic acoustic foams are widely used due to their efficiency and low cost but raise environmental concerns because of their non-biodegradable nature. This study explored the use of natural fibres—coir and sponge gourd—as environmentally friendly alternatives for sound insulation. The research focused on the effect of fibre thickness and surface treatment (using sodium hydroxide (NaOH)) on their acoustic performance. The Fourier Transform Infrared Spectroscopy (FTIR) technique was used to understand the effects of the treatment on the functional groups of the fibre. The surface modification of the fibre surface was studied using an optical microscope, Brumauer–Emmet–Teller (BET) analysis and by analysis of Scanning Electron Microscope (SEM) images. An impedance tube setup was used to measure the sound transmission loss for both the untreated and treated fibres. The results showed that increasing the thickness of both coir and sponge gourd fibres improves transmission loss at lower frequencies but reduces effectiveness at higher frequencies. Surface treatment had a material-dependent effect: sponge gourd fibres showed improved transmission loss due to increased surface roughness and airflow resistivity, whereas coir fibres exhibited a decline in performance after treatment. These findings contribute to a better understanding of how natural materials can be optimised for acoustic applications through structural modifications.