Nanoscale extracellular vesicles (EVs), including exosomes, have become important players in the etiology and potential management of Sjögren disease (SjD). By transporting bioactive substances—including proteins, microRNAs (miRNAs), and autoantigens—between immune and epithelial cells, these vesicles promote intercellular communication, exacerbating glandular dysfunction and systemic autoimmunity. Remarkably, EVs extracted from SjD patients’ plasma and saliva contain unique molecular signatures, including specific proteomic profiles and miRNAs such as let-7b-5p and miR-142-3p. These EV-derived biomarkers offer promising non-invasive diagnostic and prognostic tools for early disease detection and stratification. On the other hand, the therapeutic potential of stem cell-derived EVs, particularly from mesenchymal stem cells (MSCs), presents a novel regenerative strategy. By modulating key immune responses—particularly by shifting macrophage polarization from the pro-inflammatory M1 phenotype toward the anti-inflammatory M2 phenotype (thereby reducing M1-derived cytokines like TNF-α and IL-6), suppressing pathogenic B and T follicular helper cells via pathways including AKT/GSK-3β/Slug, RORC/IL-17 A, and NF-κB, and restoring the Th17/Treg balance—these MSC-EVs reduce inflammation, restore salivary and tear secretion, and promote tissue repair. Despite these encouraging preclinical developments, challenges regarding standardization, isolation methods, and clinical translation remain. To fully realize their potential in transforming the management of Sjögren disease, future research must focus on optimizing EV-based applications for targeted therapy and validating their efficacy in human clinical trials.
Significant fibrosis is a key determinant of long-term outcomes in metabolic dysfunction-associated steatotic liver disease (MASLD). We aimed to develop explainable machine-learning models using routine clinical and laboratory variables and assess their performance in temporal external validation. This cross-sectional diagnostic prediction study used National Health and Nutrition Examination Survey (NHANES) 2017–March 2020 data for model development and internal validation, and NHANES 2021–2023 data for temporal external validation. MASLD was defined as controlled attenuation parameter ≥ 274 dB/m plus at least one cardiometabolic risk factor. Significant fibrosis was defined as liver stiffness measurement ≥ 8.0 kPa. Six machine-learning algorithms were evaluated. The fitted pipelines were applied unchanged to the external cohort. The development cohort included 3,202 adults with MASLD, including 555 (17.3
Acute Myeloid Leukemia (AML) is a type of hematologic cancer characterized by diverse molecular and clinical features that influence diagnosis, prognosis, and treatment response. Despite advances in molecular classification and therapeutic strategies, challenges remain in achieving accurate risk assessment, treatment optimization, and clinical translation of emerging biomarkers. This literature review provides a comprehensive overview of the applications of bioinformatics and machine learning (ML) approaches in AML, focusing on their contributions to diagnosis, prognostic prediction, and therapeutic decision-making. A narrative literature review was conducted using PubMed, and Google Scholar to identify relevant publications from January 2015 through 2025. Studies investigating bioinformatics and/or ML-based approaches in AML diagnosis, prognosis, biomarker discovery, and treatment-related applications were reviewed and critically evaluated. Current evidence demonstrates that computational approaches, including conventional ML algorithms, deep learning methods, and multi-omics analyses, have facilitated the identification of molecular signatures, improved disease classification, enhanced risk stratification, and supported prediction of treatment-related outcomes. However, despite promising results, the clinical implementation of AML machine-learning models remains limited due to methodological challenges, including retrospective study designs, batch effects, overfitting, insufficient external validation, data leakage, and limited assessment of clinical utility. Overall, bioinformatics and ML represent powerful tools for advancing AML research and precision medicine. Nevertheless, rigorous validation, transparent reporting, and prospective clinical evaluation are required before these approaches can be integrated into routine AML management.
Severe acute pancreatitis (SAP) is a life-threatening condition associated with systemic inflammatory response syndrome, multi-organ failure, and mortality rates of 20–40
Systemic lupus erythematosus (SLE) features uncontrolled humoral immunity. However, the alterations in innate and humoral immune profiles remain unclear. This study assessed immune cell profile (ICP) changes in patients with SLE versus healthy controls (HCs). Blood samples from 18 patients with SLE and 21 HCs were analyzed. The analysis covered granulocytes, B cells, T lymphocytes, and regulatory cells. An AI model evaluated each subset’s role in distinguishing the groups. Patients with SLE showed increased effector memory and terminal effector CD4+ T cells, marginal zone B cells, and IgMhi B cells, but decreased memory, follicular, and IgMdim B cells, indicating an active primary humoral response. The AI model perfectly distinguished patients with SLE from HCs (precision, recall, AUC = 1.00). It identified five key predictive subsets: follicular, IgMhi, and IgMdim B cells; CTLA4+PD-1+ in FoxP3+ CD4+ Treg cells; and effector memory CD8+ αβ T cells. This study identified a distinct ICP in patients with SLE, enhancing understanding of ICP dynamics in SLE pathogenesis.
