225 papers
Health informatics sits at the vibrant intersection of medicine, data science, and technology, transforming how we store, analyze, and utilize health information. This rapidly evolving field empowers clinicians and researchers to uncover patterns in patient data, improve diagnostic accuracy, and personalize treatment plans without getting lost in complex databases. By turning raw medical records into actionable insights, these innovations are reshaping the future of healthcare delivery and population health management.
At Gist.Science, we bridge the gap between cutting-edge research and public understanding by curating the latest preprints from medRxiv specifically within this domain. Our team processes every new submission in this category, providing both accessible plain-language explanations and detailed technical summaries to ensure the science is clear for everyone, from policymakers to curious readers. Below are the latest papers in health informatics, freshly distilled and ready for you to explore.
This paper introduces EpiKG2DAG, an automated framework that transforms unstructured biomedical text into an Epidemiological Knowledge Graph to systematically generate evidence-anchored Directed Acyclic Graphs (DAGs) for causal inference, thereby addressing the critical evidence retrieval gap in traditional expert-driven modeling.
This study introduces FHIRBench-UK to demonstrate that the optimal FHIR-to-text serialisation format for clinical LLMs is context-dependent rather than universal, revealing that task-aware routing significantly improves performance across diverse models and tasks while highlighting a critical disconnect between token-level metrics and clinical quality.
In a multi-system cohort of over 17,000 patients, the 2026 AHA/ACC pulmonary embolism framework demonstrated a reproducible mortality gradient across its five main categories but showed limited prognostic refinement within the broad intermediate-risk range and modest reclassification differences compared to the 2019 ESC system.
This study integrates an agent-based immune system model with transcriptomic analysis to identify that the cytotoxic efficacy of natural killer and CD8+ T cells in recognizing tumor-associated antigens is a critical determinant of hepatoblastoma treatment outcomes.
This withdrawn protocol-based study aimed to construct a risk model for postoperative common bile duct stone recurrence by evaluating the knowledge, attitudes, and practices of 100 patients, revealing that while patients possessed adequate knowledge, they exhibited negative attitudes and poor adherence to preventive practices.
This study analyzes over 6 million tweets from Sub-Saharan Africa to reveal that COVID-19 vaccine misinformation is driven by region-specific concerns rather than income levels, peaks around policy mandates, and necessitates tailored, long-term counter-misinformation strategies.
This paper proposes a hybrid decision-support framework that integrates survey-weighted rule-ensemble learning with Decision Model and Notation (DMN) to transform narrative clinical recommendations into auditable, executable logic, demonstrating its ability to classify diabetes status and quantify the sensitivity of patient risk to hypothetical BMI reductions in an NHANES-derived cohort.
This paper introduces a multidimensional engagement metric for the Helpilepsy seizure diary, using clustering analysis to identify distinct user groups and revealing that highly engaged patients are typically older with longer epilepsy histories and more complex medication regimens, thereby highlighting the need for targeted onboarding strategies for newer patients and the importance of utilizing diverse diary features beyond simple seizure logging.
This paper introduces the EHR Density Index (EDI), a novel metric derived from UNC Health data that quantifies documentation volume and depth independent of disease burden to serve as a confounding control in real-world EHR-based research.
This study demonstrates that while a deterministic ontology and embedding-based pipeline can partially harmonize heterogeneous clinical trial condition data in Germany to ICD-10 standards, Large Language Models (specifically GPT-4o) significantly outperform it by achieving near-perfect agreement with expert human coding, thereby offering a superior solution for cross-registry disease landscape analysis.