clinical cohort / trial

This paper introduces a hierarchical temporal transformer for cancer grade prediction from pathology reports, showing improved performance across cancer types and using visit timing when identifiable.

This study identifies clinical, biochemical, and radiographic features of ECHS1 deficiency through a multi-center retrospective analysis of 113 patients, highlighting differences in onset and outcomes across age groups.

BioResearchBench evaluates AI models on 50 biology research tasks, revealing low success rates and highlighting challenges in data analysis, interpretation, and experimental design.

This study shows that better adherence to the DASH diet in midlife and early elderhood improves late-life verbal memory and brain health through structural and neurovascular mechanisms.

This study identifies NEPC as a malignant state converging on a fetal pulmonary neuroendocrine-like program and identifies ASCL1-linked regulatory targets.

The paper describes a severely affected MRL/Lpr mouse with divergent clinical-immune features and profound metabolic alterations, highlighting metabolic dysregulation as a key factor in advanced disease.

This study demonstrates a rapid AST method using spiked urine samples achieving 98% categorical agreement with gold standard culture results for UTI pathogens.

The paper tests for collateral sensitivity SCCs in clinical data and finds the association is confounded by clonal lineage, not true biological interaction.

This study identifies CETP activity and early SAA levels as prognostic biomarkers in sepsis-associated ARDS, showing they predict improved survival and reduced disease severity.

BioMCP-TS provides an open-source server enabling AI agents to access and analyze biomedical databases and run analyses without complex dependencies.

This study examines how APOE isoforms affect biological outcomes in industrialized and non-industrial environments, finding lifestyle-dependent effects and weak reproductive benefits for APOE ε4.

This study reanalyzes koala chlamydia vaccine trial data, identifies biases and errors, and finds no significant protective efficacy of the vaccine.

This study shows that blocking the Neurokinin 1 receptor with antagonists improves survival in C. difficile infection patients and reduces vascular inflammation.

This study integrates clinical and T-cell immune profiling to identify early biomarkers of response to CD3xBCMA bispecific antibody therapy in multiple myeloma.

The paper benchmarks 17 synthetic data generators for clinical lab data, finding no single model excels across all metrics, and highlights ForestDiffusion's best privacy-utility balance.

This study shows that early suppression of the pro-inflammatory senescent secretome (SASP) with ruxolitinib reduces chronic bone damage after radiotherapy.

CancerStop.dev is a web platform that integrates prognosis, clinical trials, and genomic data to help patients and clinicians make informed decisions about cancer care.

This paper develops a multi-omics machine learning model to predict breast cancer survival, finding that epigenetic features are the strongest predictors.

The paper identifies DPP9 as a genetic susceptibility gene for checkpoint inhibitor pneumonitis and shows IL-18 blockade could be a therapeutic strategy.

The paper shows that Peyer's patch LysoDCs generate microbiota-specific Th17 responses distinct from those in mesenteric lymph nodes, with functional consequences for mucosal immunity.

This paper develops and evaluates model ensembling and machine learning approaches to predict the first dose of amoxicillin in intensive care, finding that ensembling and ML outperform standard dosing.

The paper introduces the Immunogenicity Database (IDC) to standardize and integrate clinical immunogenicity data, highlighting trends in anti-drug antibody incidence and risk factors.

The paper models in vivo functions of five anti-HIV bnAbs using clinical trial data to quantify neutralization, Fc-mediated killing, and resistance.

The paper analyzes CGM data to understand hypoglycemia kinetics in T1D patients, leading to the development of new glucagon delivery systems for improved therapy.

This study compares phantomless and inline phantom-based densitometric calibration of vertebral properties from CT scans, finding a strong correlation between the two methods using air and aorta as calibration regions.

This study examines how mutualism affects legume niches globally, finding that mutualist species occupy broader niches than non-mutualists, with effects varying by mutualism type and latitude.

This study characterizes a leptospirosis outbreak in dogs in Los Angeles County, finding that all isolates were Leptospira interrogans serovar Canicola, with urine PCR as the most sensitive test and ICFs as a potential source.

This study shows HS3ST1 deficiency worsens lung injury and increases mortality after trauma and hemorrhagic shock, particularly with fresh frozen plasma resuscitation.