193 papers
This qualitative study reveals that cybersecurity professionals view health AI as a fragile, augmented clinical infrastructure where institutional trust is contingent upon visible governance and accountability rather than technical performance alone.
This study demonstrates that Large Language Models exhibit a High-Income Country bias in respiratory diagnoses, offering fewer relevant diagnostic suggestions for Low- and Middle-Income Country settings compared to local clinicians, even when prompted with specific location information.
The paper introduces GEN-KnowRD, a framework that leverages large language models to generate schema-guided rare disease profiles and construct a computable knowledge base, thereby decoupling knowledge construction from patient-level inference to significantly improve rare disease recognition and early diagnosis compared to existing state-of-the-art methods.
This study demonstrates that machine learning algorithms leveraging contextual electronic health record features can effectively detect absolute unwarranted clinical variation, specifically inappropriate antibiotic prescriptions for pediatric acute viral pharyngitis, offering a scalable and interpretable alternative to traditional statistical methods.
This paper presents a scalable framework that reconstructs metastatic breast cancer treatment lines from real-world EHR data using radiology-anchored progression evidence to generate individualized, interpretable progression-free survival estimates that enable robust risk stratification across diverse patient subgroups and successive lines of therapy.
PhenoSS is a novel, statistically principled framework that leverages Gaussian copulas and batch-effect correction to model hierarchical phenotype dependencies, thereby improving rare disease prediction accuracy and enabling effective patient clustering from clinical data.
The paper introduces RareDAI, an interpretable fine-tuned large language model framework that utilizes chain-of-thought reasoning and self-distillation to accurately guide clinicians in selecting appropriate genetic tests for rare diseases by effectively applying ACMG guidelines to heterogeneous clinical data.
This paper proposes a Deep Image Prior-based framework that integrates anatomical proton MRI and metabolic sodium MRI guidance through fused directional total variation regularization to achieve high-quality, accelerated sodium MRI reconstruction with improved structural fidelity and signal preservation.
This study demonstrates that current large language models can accurately and reliably debunk common vaccine myths with high scientific accuracy, though their effectiveness as public health tools is currently limited by linguistic complexity and framing-dependent communication styles that may reduce accessibility for lay audiences.
This study demonstrates that few-shot prompted large language models can effectively categorize specific types of clinician edits to ambient AI draft notes, achieving strong performance for medication and symptom modifications while identifying complex, context-dependent edits as better suited for human review triage.
This study evaluating the implementation of Epic's GenAI tool for patient messaging in a Dutch academic hospital found that while initial adoption was high and drafts were well-structured, the tool ultimately failed to improve clinical efficiency or well-being, leading to declining usage due to barriers such as limited time savings, content inaccuracies, and misaligned expectations.
This paper proposes a novel clinical theory-driven deep learning model that operationalizes established autism constructs into a structured architecture with cross-modal attention and theory-specific weighting to achieve state-of-the-art, interpretable prediction of autism severity while providing empirical support for its multidimensional nature.
The 2025 review of healthcare AI research reveals a near-doubling of publication volume and a significant maturation of the field, characterized by a shift from classical machine learning and text-only models toward the rapid adoption of multimodal foundation models that better reflect the complexity of real-world clinical practice.
This study demonstrates that automated feature engineering via Deep Feature Synthesis significantly enhances the discrimination, calibration, and clinical utility of gradient-boosted tree models for predicting heart failure readmissions from longitudinal electronic health records, outperforming traditional clinician-curated features.
This cohort study found that the type of editorial response (retraction, expression of concern, or editorial notice) issued for integrity concerns in randomized controlled trials did not significantly influence subsequent citation rates or accelerate citation decline compared to control trials.
This study presents a governance-driven health informatics framework that leverages real-world longitudinal data to model patient treatment sequencing, persistence, and provider adoption, thereby significantly improving the accuracy of healthcare utilization forecasts for oncology and complex chronic conditions compared to traditional static market-share approaches.
This paper introduces SO_SCPLOWTRATC_SCPLOWCP, a stratified conformal framework that enables the safe clinical deployment of medical foundation models by controlling false discovery rates for immediate actions and providing calibrated, coverage-guaranteed prediction sets for deferred cases, thereby optimizing healthcare resources and reducing costs in ophthalmology and neuro-oncology.
This study evaluates the representational robustness of 15 locally deployable large language models on neurological terminology, revealing that while performance generally scales with model size, neither size nor medical fine-tuning guarantees clinical reliability due to significant variations based on terminological complexity and subdomain.
This study finds that while the Sleep Regularity Questionnaire shows modest agreement with subjective sleep diary measures, it demonstrates only weak correspondence with objective smart ring data, suggesting it is best used as a complementary tool rather than a replacement for objective monitoring in healthy adults.
This randomized controlled trial demonstrates that using large language models to simplify radiology reports significantly improves patient readability and comprehension with minimal hallucination rates, thereby enhancing patient-clinician communication and meeting strong patient demand for accessible medical information.