193 papers
This study reveals that a Finnish healthcare AI assistant exhibits significant gendered biases, including stereotypical framing of female patients around reproductive health and inconsistent clinical reasoning, stemming from flaws in both its retrieval and generation stages.
This study presents a comprehensive, literature-informed synthetic phase II oncology clinical trial framework that successfully integrates radiographic, molecular (ctDNA), safety, and survival data to generate a biologically plausible, analytically coherent efficacy-safety signal, thereby serving as a decision-oriented prototype for translational clinical data science.
This study demonstrates that replacing raw chest radiographs with expert-written reports in multimodal mortality prediction models leads to significant information loss and altered risk prioritization, suggesting that text summaries are imperfect proxies for visual prognostic cues.
This systematic review and scientometric analysis of 928 chest X-ray AI studies reveals that research leadership and training data are overwhelmingly dominated by high-income countries, particularly the US and China, creating significant disparities that risk developing AI systems with inconsistent performance across diverse global populations and exacerbating healthcare inequities.
This multimethod study reveals that while perceived safety significantly influences patient and care partner facility selection in behavioral health crisis units, there are notable discrepancies between human and AI-identified environmental risks, underscoring the value of integrating AI tools to support safer decision-making while acknowledging their current limitations in capturing nuanced human perceptions.
This large-scale international survey reveals that while medical journal peer reviewers are highly familiar with AI chatbots, their actual use in peer review remains limited due to significant ethical concerns and a lack of institutional training, despite a strong expressed interest in receiving guidance for future implementation.
This paper presents a high-throughput workflow utilizing linked electronic health record and claims data to generate standardized, comprehensive evidence packages across 40 clinical domains, thereby shifting comparative effectiveness research from fragmented studies toward a cohesive evidence base that supports precision medicine and reduces redundant stakeholder-specific investigations.
This study presents a prevalence-adaptive Bayesian ensemble model that utilizes Variational Autoencoder-based data augmentation and entropy-driven uncertainty quantification to achieve perfect separation on a validation cohort and clinically meaningful sensitivity with zero false positives in predicting perioperative mortality within resource-limited surgical settings.
The paper introduces TELF, a lightweight, end-to-end temporal encoder with late fusion that bypasses resource-intensive pre-training to achieve superior predictive performance and interpretable clinical trajectory mining for disease risk prediction using longitudinal real-world data.
This study demonstrates that voice-based models trained on 7,081 adults can accurately predict biological age and identify accelerated aging linked to cardiometabolic risks, establishing voice as a scalable, non-invasive biomarker for health monitoring.
This paper presents "Digital Registrar," a schema-first framework that leverages local large language models and a CAP-aligned clinical ontology to accurately transform free-text surgical pathology reports into structured, privacy-preserving registry data across multiple cancer types, achieving high accuracy and generalizability while decoupling clinical logic from specific AI models.
This study demonstrates that while CD276 is significantly associated with meningioma grade, it functions as a biologically relevant target-of-interest rather than a dominant standalone predictor, with the model's robust performance and stability relying on a broader multigene transcriptomic structure that requires conservative, calibration-aware interpretation.
This study evaluates the reproducibility and robustness of three distinct large language models in extracting mobility functional status from clinical narratives, demonstrating that while prompt variations and higher temperatures can significantly degrade stability, self-consistency via majority voting offers an effective mitigation strategy to enhance reliability without sacrificing predictive performance.
This study demonstrates that machine learning models trained on electronic health record data can accurately estimate Kansas City Cardiomyopathy Questionnaire scores and classify heart failure severity, offering a viable solution to overcome the limitations of incomplete patient-reported outcome data collection.
This study demonstrates that large language models, particularly advanced "mini" variants, outperform rule-based systems in extracting social determinants of health from unstructured clinical notes, with the highest accuracy achieved by a late-fusion ensemble combining both approaches.
This study demonstrates that supervised continued pre-training of a 4B-parameter LLM on 500,000 de-identified clinical notes significantly enhances its performance on real-world medical decision-making tasks and specific clinical benchmarks, surpassing larger untrained models while successfully retaining general-domain knowledge.
Med-ICE is an autonomous multi-agent framework that enhances the factual accuracy and reliability of medical AI by employing an iterative peer-review consensus mechanism to eliminate hallucinations and outperform existing single-model and self-refinement approaches.
This paper introduces MedResearchBench to evaluate six medical AI research systems, revealing that citation hallucination is the decisive factor in their success and demonstrating that multi-agent verification pipelines significantly outperform subjective single-model evaluations in ensuring trustworthy scientific writing.
This study demonstrates that composite-only multi-task learning effectively profiles multidimensional psychosocial burdens in cancer peer-support text, while soft-label supervision derived from large language models underperforms compared to hard-label baselines for emotion classification.
This study demonstrates that machine learning models trained on acoustic features of vowel production, particularly fundamental frequency, can effectively distinguish autistic adults from neurotypical individuals with up to 89% accuracy, suggesting a promising, scalable, and interpretable tool for autism spectrum disorder screening.