180 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 study presents a deep learning framework based on a vision foundation model that not only accurately predicts retinal biological age from fundus images but also disentangles physiological heterogeneity by decomposing aging signals into normative and pathological components linked to systemic health factors like inflammation and hemodynamics.
Researchers developed and validated a machine learning-based Neurological Health Score (NHS) using genetic, lifestyle, and biochemical data, demonstrating that higher scores significantly correlate with a reduced risk of various neurological disorders and improved social well-being.
This study evaluates and compares various statistical and machine learning imputation methods for handling missing data in esophageal cancer patient-reported outcomes to provide evidence-based recommendations for improving research validity.
This study reveals a critical dissociation between abstract ethical reasoning and clinical memory in large language models, finding that while most models can resist harmful commands, very few can successfully recall patient-specific contraindications like allergies during ICU simulations.
This study presents a patient-centric Markov-chain framework that utilizes eight years of de-identified specialty pharmacy data to predict medication adherence risks, identifying systemic barriers like cost and administrative delays as primary drivers of non-adherence to support more equitable patient-assistance strategies.
The paper introduces BODHI, an engineering framework that enhances clinical decision support AI by decomposing epistemic uncertainty and applying virtue-based rules, which significantly improves model humility, curiosity, and overall response quality in controlled evaluations.
This study demonstrates that an explainable XGBoost machine learning model, which integrates anthropometric data with meal timing and eating habits, accurately predicts cardiometabolic risk and identifies later dinner times, irregular meal schedules, and low meal frequency as key modifiable behavioral risk factors.
This cross-sectional study utilized natural language processing to analyze over 30,000 secure ophthalmology messages, revealing that while administrative issues dominate communication, clinical concerns vary significantly by patient demographics, highlighting opportunities to improve triage safety and equity in eye care.
This study demonstrates that uncurated AI-generated data in medical records creates a self-referential cycle that rapidly erodes pathological variability and diagnostic reliability, causing critical findings to vanish and false reassurance rates to triple, thereby rendering AI-generated documentation clinically useless without mandatory human oversight.
Kauro is an open-source, graph-based chatbot that ensures deterministic, auditable, and reproducible information transmission in high-stakes biomedical contexts like informed consent by replacing the stochastic nature of generative AI with version-controlled, scripted conversation paths.