256 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 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.
This study demonstrates that a large language model can generate psychometrically plausible and demographically consistent synthetic data for Peruvian health surveys when conditioned on user personas, though the presence of specific deviations necessitates rigorous validation before such data can be used for policy-making.
This study demonstrates that integrating partial least squares regression, convolutional neural networks, and curve-fitting approaches on FTIR blood spectra creates a synergistic framework that simultaneously achieves high accuracy in predicting HbA1c levels and provides mechanistic interpretability of glycation-related biomarkers for scalable diabetes monitoring.
The paper introduces C-RLM, a schema-enforced recursive framework that outperforms standard RAG and long-context LLMs in synthesizing complex clinical records by achieving 100% structural consistency and full auditability through iterative knowledge compilation, despite a 2.7x computational overhead.
This study demonstrates that an on-premises, open-source multimodal framework fusing structured clinical data with free-text pathology reports using large language models significantly improves the discrimination of complication risks after implant-based breast reconstruction, offering a privacy-preserving and interpretable tool for precision surgical decision-making.