Hepatitis C virus (HCV) remains underdiagnosed in the United States despite recommendations for universal screening. A simple approach based on readily available demographic information may help target screening in settings where screening implementation continues to be incomplete. We analyzed 10 NHANES cycles (1999–2014 and 2017–2023) and defined HCV exposure as a positive HCV antibody or RNA result. Using sex, birth year, race/ethnicity, birthplace, and income-to-poverty ratio, we trained and compared logistic regression (LR) and machine learning models in training and cycle-based internal validation cohorts (48,434 and 20,762 participants, respectively). Model performance was evaluated based on sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic curve (AUROC). A web-based calculator was developed to estimate HCV-marker positivity risk and support prioritization of confirmatory HCV testing. 69,196 participants were included, with 967 showing evidence of HCV exposure. Weighted HCV prevalence remained relatively stable across cycles, ranging from 1.22

Electronic Health Records (EHRs) are pivotal in clinical practices, yet their retrieval remains a challenge mainly due to semantic gap issues. Recent advancements in dense retrieval offer promising solutions but existing models, both general-domain and biomedical-domain, fall short due to insufficient medical knowledge or mismatched training corpora. Previous EHR retrievers generally focus on a small set of queries and fail to generalize. This paper introduces DR.EHR, a series of dense retrieval models specifically tailored for EHR retrieval. We propose a two-stage training pipeline utilizing MIMIC-IV discharge summaries to address the need for extensive medical knowledge and large-scale training data. The first stage involves medical entity extraction and knowledge injection from a biomedical knowledge graph, while the second stage employs large language models to generate diverse training data. We train two variants of DR.EHR, with 110 M and 7B parameters, respectively. Evaluated on the CliniQ benchmark, our models significantly outperforms all existing dense retrievers, achieving state-of-the-art results. Detailed analyses confirm our models' superiority across various match and query types, particularly in challenging semantic matches such as implication and abbreviation. Ablation studies validate the effectiveness of each pipeline component, and supplementary experiments on EHR QA datasets demonstrate the models' generalizability on various EHR corpus and natural language questions including complex ones with multiple entities. This work significantly advances EHR retrieval, offering a robust solution for clinical applications.

Background Hepatitis C virus (HCV) remains underdiagnosed in the United States despite recommendations for universal screening. A simple approach based on readily available demographic information may help target screening in settings where screening implementation continues to be incomplete. Methods We analyzed 10 NHANES cycles (1999–2014 and 2017–2023) and defined HCV exposure as a positive HCV antibody or RNA result. Using sex, birth year, race/ethnicity, birthplace, and income-to-poverty ratio, we trained and compared logistic regression (LR) and machine learning models in training and validation cohorts (48,434 and 20,762 participants, respectively). Model performance was evaluated based on sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic curve (AUROC). A web-based calculator was developed to facilitate bedside HCV screening. Results 69,196 participants were included, with 967 showing evidence of HCV exposure. Weighted HCV prevalence remained relatively stable across cycles, ranging from 1.22% to 1.93%. The prevalence did not change significantly after the pandemic. Earlier birth year, male sex, non-Hispanic Black race, US birth, and lower income-to-poverty ratio were independently associated with HCV exposure. XGBoost performed better than LR in the validation cohort (AUROC 0.860 vs 0.762, p < 0.001). Predicted risk separated the population clearly: observed HCV prevalence increased from 0.05% in the lowest-risk decile to 7.95% in the highest, with the top decile containing 58.3% of participants with HCV exposure and the top three deciles containing 85.5%. Conclusions Five demographic variables were sufficient to build a useful HCV risk model in a nationally representative US sample. Most HCV-exposed individuals were concentrated in the highest predicted-risk groups, suggesting that this approach could help prioritize and optimize testing where universal screening uptake remains incomplete. As no laboratory data is required, it may also be practical in data-limited settings and adaptable in other health systems.

Rare diseases pose a substantial clinical and public health burden, with timely and accurate diagnoses remaining a formidable challenge in many countries and settings. Large language models (LLMs) have the potential to enhance the screening and diagnosis of rare diseases. However, existing evaluations of such capacities of LLMs are insufficient due to the low sample sizes, poor coverage of different rare diseases, and weak real-world clinical relevance. To address these issues, we developed an automated data collection pipeline, RareArena, to curate a comprehensive rare disease dataset from PubMed Central case reports. Two tasks were considered: rare disease screening (RDS), which involved rare disease recognition before confirmatory diagnostic tests; and rare disease confirmation (RDC), which required diagnosis with complete information. The constructed RDS task comprised 49 760 cases of 4597 rare diseases, whereas the RDC task comprised 22 901 cases of 3522 rare diseases. Human evaluations by two physicians across three dimensions-leakage, fidelity, and complexity-validated the high quality of our dataset. We systematically benchmarked ten state-of-the-art LLMs of varying sizes (GPT-4o, Llama3·1-70B, Llama3·1-8B, Qwen2·5-72B, Qwen2·5-7B, Gemma2-9B, Phi3-7B, Llama3-70B-UltraMedical, Llama3-8B-UltraMedical, and MedLlama-7B) and found that GPT-4o achieved the best performance in both the tasks (RDS: 33·1% top-1 recall, 56·9% top-5 recall; RDC: 64·2% top-1 recall, 85·9% top-5 recall). Additionally, GPT-4o performed better on genetically inherited diseases than on non-genetically inherited diseases in both the tasks, with its highest performance observed in systemic or rheumatologic diseases (RDS: 48·5% top-1 recall, 76·5% top-5 recall; RDC: 76·0% top-1 recall, 93·5% top-5 recall). To our knowledge, our dataset represents the largest rare disease diagnostic benchmark to date, featuring broad and deep disease coverage and reflecting clinically relevant scenarios. RareArena highlights the potential of LLMs in rare disease screening and diagnosis, providing a foundation for improved care for individuals with rare diseases around the world.

Rare diseases affect millions of individuals worldwide, yet timely diagnosis remains a major public health challenge due to scarcity of specialized clinical expertise. While large language models (LLMs) show promise to support rare disease diagnosis, current models are constrained by insufficient clinical deployability, limited clinically grounded evidence, and scarcity of training data. Here we present RaDaR (Rare Disease navigatoR), an open-source, compact reasoning LLM (32B parameters) for rare disease diagnosis. RaDaR was trained with 49,170 publicly available free-text cases and 104,666 synthetic cases with reasoning-enhanced training. RaDaR showed the strongest performance among evaluated open-source models, including the 671B DeepSeek-R1, across public benchmarks and four external validation centers. In a retrospective cohort, RaDaR prioritized the final diagnosis before documented clinical suspicion in 61.06 percent of cases, corresponding to a potential lead time of 1.87 months and 50.18 percent of the within-center interval. In a randomized physician-assistance trial, RaDaR assistance improved physicians' rare-disease diagnostic accuracy by 21.44 percentage points compared with internet search alone. Synthetic-data ablations suggested that phenotype-anchored narratives provide useful training signal for long-tail rare diseases, with a monotonic scaling trend within the tested data range. Together, RaDaR and its development and validation framework provide a deployable rare-disease reasoning model and a reproducible development framework for diagnostic AI under data scarcity.

Objective: To develop a high-throughput biomedical relation extraction system that takes advantage of the large language models'(LLMs) reading comprehension ability and biomedical world knowledge in a scalable and evidential manner. Methods: We formulate the relation extraction task as binary classifications for large language models. Specifically, LLMs make the decision based on the external corpus and its world knowledge, giving the reason for the judgment for factual verification. This method is tailored for semi-structured web articles, wherein we designate the main title as the tail entity and explicitly incorporate it into the context, and the potential head entities are matched based on a biomedical thesaurus. Moreover, lengthy contents are sliced into text chunks, embedded, and retrieved with additional embedding models. Results: Using an open-source LLM, we extracted 248659 relation triplets of three distinct relation types from three reputable biomedical websites. To assess the efficacy of the basic pipeline employed for biomedical relation extraction, we curated a benchmark dataset annotated by a medical expert. Evaluation results indicate that the pipeline exhibits performance comparable to that of GPT-4. Case studies further illuminate challenges faced by contemporary LLMs in the context of biomedical relation extraction for semi-structured web articles. Conclusion: The proposed method has demonstrated its effectiveness in leveraging the strengths of LLMs for high-throughput biomedical relation extraction. Its adaptability is evident, as it can be seamlessly extended to diverse semi-structured biomedical websites, facilitating the extraction of various types of biomedical relations with ease.

OBJECTIVE:To construct an Alzheimer's Disease Knowledge Graph (ADKG) by extracting and integrating relationships among Alzheimer's disease (AD), genes, variants, chemicals, drugs, and other diseases from biomedical literature, aiming to identify existing treatments, potential targets, and diagnostic methods for AD. METHODS:We annotated 800 PubMed abstracts (ADERC corpus) with 20,886 entities and 4935 relationships, augmented via GPT-4. A SpERT model (SciBERT-based) trained on this data extracted relations from PubMed abstracts, supported by biomedical databases and entity linking refined via abbreviation resolution/string matching. The resulting knowledge graph trained embedding models to predict novel relationships. ADKG's utility was validated by integrating it with UK Biobank data for predictive modeling. RESULTS:The ADKG contained 3,199,276 entity mentions and 633,733 triplets, linking >5K unique entities and capturing complex AD-related interactions. Its graph embedding models produced evidence-supported predictions, enabling testable hypotheses. In UK Biobank predictive modeling, ADKG-enhanced models achieved higher AUROC of 0.928 comparing to 0.903 without ADKG enhancement. CONCLUSION:By synthesizing literature-derived insights into a computable framework, ADKG bridges molecular mechanisms to clinical phenotypes, advancing precision medicine in Alzheimer's research. Its structured data and predictive utility underscore its potential to accelerate therapeutic discovery and risk stratification.

Clinical narratives in electronic health records (EHRs) contain essential diagnostic, therapeutic, and temporal information that is often missing from structured fields, leaving manual chart review as the de facto standard for high-quality labels, but slow, costly, and variable, thereby constraining accurate cohort construction for clinical trials, large-scale epidemiologic studies, and the development of robust machine-learning models. We developed CLINES, a modular agentic pipeline that extracts and structures clinical concepts: semantic chunking of long notes; extraction by reasoningcapable large language models; assignment of attributes (assertion/experiencer, numerical values with SI units); normalization to the Unified Medical Language System (UMLS); resolution of explicit and relative dates; and aggregation into an i2b2-style schema. Zero-shot evaluation was conducted on de-identified EHR: MIMIC-III notes, 4CE notes, and CORAL oncology reports (breast, pancreas). Comparators included rule/lexicon systems, transformer encoders, and single-prompt LLM baselines. Outcomes were F1 scores for entity extraction, assertion status, value&unit extraction, and date processing. Across all datasets, CLINES led every baseline. F1 scores (entity / assertion / value&unit / date) were: MIMIC-III 0.69 / 0.93 / 0.90 (date not evaluated); 4CE 0.87 / 0.88 / 0.79 / 0.79; CORAL–Breast 0.81 / 0.84 / 0.77 / 0.73; CORAL–Pancreas 0.85 / 0.87 / 0.90 / 0.78. Gains over the strongest single-prompt LLM were +0.21–0.38 across tasks, and transformer encoders trailed by +0.28–0.68 F1 on entity extraction. Performance remained stable across note-length quantiles, while transformer baselines lost recall as notes lengthened. CLINES translates narrative text from electronic health records into ontology-grounded, auditable, and schema-ready data, offering a practical route to scale chart-review-like extraction for cohort discovery and real-world evidence. CLINES is model agnostic–different open and close models can be substituted to achieve specific cost, performance, and privacy goals. Future work aims to quantify inter-annotator agreements and explore adaptive feedback and domain-specific fine-tuning.

Electronic Health Records (EHRs) hold immense potential for advancing healthcare, offering rich, longitudinal data that combines structured information with valuable insights from unstructured clinical notes. However, the unstructured nature of clinical text poses significant challenges for secondary applications. Traditional methods for structuring EHR free-text data, such as rule-based systems and multi-stage pipelines, are often limited by their time-consuming configurations and inability to adapt across clinical notes from diverse healthcare settings. Few systems provide a comprehensive attribute extraction for terminologies. While giant large language models (LLMs) like GPT-4 and LLaMA 405B excel at structuring tasks, they are slow, costly, and impractical for large-scale use. To overcome these limitations, we introduce GENIE, a Generative Note Information Extraction system that leverages LLMs to streamline the structuring of unstructured clinical text into usable data with standardized format. GENIE processes entire paragraphs in a single pass, extracting entities, assertion statuses, locations, modifiers, values, and purposes with high accuracy. Its unified, end-to-end approach simplifies workflows, reduces errors, and eliminates the need for extensive manual intervention. Using a robust data preparation pipeline and fine-tuned small scale LLMs, GENIE achieves competitive performance across multiple information extraction tasks, outperforming traditional tools like cTAKES and MetaMap and can handle extra attributes to be extracted. GENIE strongly enhances real-world applicability and scalability in healthcare systems. By open-sourcing the model and test data, we aim to encourage collaboration and drive further advancements in EHR structurization.

Electronic Health Record (EHR) retrieval plays a pivotal role in various clinical tasks, but its development has been severely impeded by the lack of publicly available benchmarks. In this paper, we introduce a novel public EHR retrieval benchmark, CliniQ, to address this gap. We consider two retrieval settings: Single-Patient Retrieval and Multi-Patient Retrieval, reflecting various real-world scenarios. Single-Patient Retrieval focuses on finding relevant parts within a patient note, while Multi-Patient Retrieval involves retrieving EHRs from multiple patients. We build our benchmark upon 1,000 discharge summary notes along with the ICD codes and prescription labels from MIMIC-III, and collect 1,246 unique queries with 77,206 relevance judgments by further leveraging powerful LLMs as annotators. Additionally, we include a novel assessment of the semantic gap issue in EHR retrieval by categorizing matching types into string match and four types of semantic matches. On our proposed benchmark, we conduct a comprehensive evaluation of various retrieval methods, ranging from conventional exact match to popular dense retrievers. Our experiments find that BM25 sets a strong baseline and performs competitively to the dense retrievers, and general domain dense retrievers surprisingly outperform those designed for the medical domain. In-depth analyses on various matching types reveal the strengths and drawbacks of different methods, enlightening the potential for targeted improvement. We believe that our benchmark will stimulate the research communities to advance EHR retrieval systems.

Mental health is a major global concern, yet findings remain fragmented across studies and databases, hindering integrative understanding and clinical translation. To address this gap, we present the Mental Disorders Knowledge Graph (MDKG)-a large-scale, contextualized knowledge graph built using large language models to unify evidence from biomedical literature and curated databases. MDKG comprises over 10 million relations, including nearly 1 million novel associations absent from existing resources. By structurally encoding contextual features such as conditionality, demographic factors, and co-occurring clinical attributes, the graph enables more nuanced interpretation and rapid expert validation, reducing evaluation time by up to 70%. Applied to predictive modeling in the UK Biobank, MDKG-enhanced representations yielded significant gains in predictive performance across multiple mental disorders. As a scalable and semantically enriched resource, MDKG offers a powerful foundation for accelerating psychiatric research and enabling interpretable, data-driven clinical insights.

Electronic health records (EHR) contain a wealth of biomedical information, serving as valuable resources for the development of precision medicine systems. However, privacy concerns have resulted in limited access to high-quality and large-scale EHR data for researchers, impeding progress in methodological development. Recent research has delved into synthesizing realistic EHR data through generative modeling techniques, where a majority of proposed methods relied on generative adversarial networks (GAN) and their variants for EHR synthesis. Despite GAN-based methods attaining state-of-the-art performance in generating EHR data, these approaches are difficult to train and prone to mode collapse. Recently introduced in generative modeling, diffusion models have established cutting-edge performance in image generation, but their efficacy in EHR data synthesis remains largely unexplored. In this study, we investigate the potential of diffusion models for EHR data synthesis and introduce a novel method, EHRDiff. Through extensive experiments, EHRDiff establishes new state-of-the-art quality for synthetic EHR data, protecting private information in the meanwhile.

Electronic medical records (EMRs) contain the majority of patients' healthcare details. It is an abundant resource for developing an automatic healthcare system. Most of the natural language processing (NLP) studies on EMR processing, such as concept extraction, are adversely affected by the inaccurate segmentation of EMR sections. At the same time, not enough attention has been given to the accurate sectioning of EMRs. The information that may occur in section structures is unvalued. This work focuses on the segmentation of EMRs and proposes a black-box segmentation method using a simple sentence embedding model and neural network, along with a proper training method. To achieve universal adaptivity, we train our model on the dataset with different section headings formats. We compare several advanced deep learning-based NLP methods, and our method achieves the best segmentation accuracies (above 98 on various test data with a proper training corpus.

Electronic health records (EHRs) hold significant value for research and applications. As a new way of information extraction, question answering (QA) can extract more flexible information than conventional methods and is more accessible to clinical researchers, but its progress is impeded by the scarcity of annotated data. In this paper, we propose a novel approach that automatically generates training data for transfer learning of QA models. Our pipeline incorporates a preprocessing module to handle challenges posed by extraction types that are not readily compatible with extractive QA frameworks, including cases with discontinuous answers and many-to-one relationships. The obtained QA model exhibits excellent performance on subtasks of information extraction in EHRs, and it can effectively handle few-shot or zero-shot settings involving yes-no questions. Case studies and ablation studies demonstrate the necessity of each component in our design, and the resulting model is deemed suitable for practical use.

Objective: Biomedical Knowledge Graphs play a pivotal role in various biomedical research domains. Concurrently, term clustering emerges as a crucial step in constructing these knowledge graphs, aiming to identify synonymous terms. Due to a lack of knowledge, previous contrastive learning models trained with Unified Medical Language System (UMLS) synonyms struggle at clustering difficult terms and do not generalize well beyond UMLS terms. In this work, we leverage the world knowledge from Large Language Models (LLMs) and propose Contrastive Learning for Representing Terms via Explanations (CoRTEx) to enhance term representation and significantly improves term clustering. Materials and Methods: The model training involves generating explanations for a cleaned subset of UMLS terms using ChatGPT. We employ contrastive learning, considering term and explanation embeddings simultaneously, and progressively introduce hard negative samples. Additionally, a ChatGPT-assisted BIRCH algorithm is designed for efficient clustering of a new ontology. Results: We established a clustering test set and a hard negative test set, where our model consistently achieves the highest F1 score. With CoRTEx embeddings and the modified BIRCH algorithm, we grouped 35,580,932 terms from the Biomedical Informatics Ontology System (BIOS) into 22,104,559 clusters with O(N) queries to ChatGPT. Case studies highlight the model's efficacy in handling challenging samples, aided by information from explanations. Conclusion: By aligning terms to their explanations, CoRTEx demonstrates superior accuracy over benchmark models and robustness beyond its training set, and it is suitable for clustering terms for large-scale biomedical ontologies.

Medical automatic diagnosis aims to organize real-world diagnostic processes similar to those from human doctors and to achieve accurate diagnoses by interacting with patients. The task is formulated as a sequential decision-making problem with a series of information inquiry steps (asking about symptoms and ordering examinations) and the final diagnosis. Recent research has studied incorporating reinforcement learning for information inquiry and classification techniques for disease diagnosis, respectively. However, studies on efficiently and effectively combining the two procedures are still lacking. To address this issue, we devised an adaptive mechanism to align reinforcement learning and classification methods using distribution entropy as the medium. Additionally, we created a new dataset for patient simulation to address the lack of large-scale evaluation benchmarks. The dataset is extracted from the MedlinePlus knowledge base and contains significantly more diseases and more comprehensive symptom and examination information than existing datasets. Experimental evaluation shows that our method outperforms three current state-of-the-art methods on different datasets by achieving higher medical diagnostic accuracy with fewer inquiring turns.

Retrieval-based Clinical Decision Support (ReCDS) can aid clinical workflow by providing relevant literature and similar patients for a given patient. However, the development of ReCDS systems has been severely obstructed by the lack of diverse patient collections and publicly available large-scale patient-level annotation datasets. In this paper, we collect a novel dataset of patient summaries and relations called PMC-Patients to benchmark two ReCDS tasks: Patient-to-Article Retrieval (ReCDS-PAR) and Patient-to-Patient Retrieval (ReCDS-PPR). Specifically, we extract patient summaries from PubMed Central articles using simple heuristics and utilize the PubMed citation graph to define patient-article relevance and patient-patient similarity. PMC-Patients contains 167k patient summaries with 3.1 M patient-article relevance annotations and 293k patient-patient similarity annotations, which is the largest-scale resource for ReCDS and also one of the largest patient collections. Human evaluation and analysis show that PMC-Patients is a diverse dataset with high-quality annotations. We also implement and evaluate several ReCDS systems on the PMC-Patients benchmarks to show its challenges and conduct several case studies to show the clinical utility of PMC-Patients.

Liver cancer is a common malignant tumor, and its clinical stage is closely related to the clinical treatment and prognosis of patients. Currently, the BCLC staging system revised by the BCLC group of University of Barcelona is the globally recognized staging system for liver cancer. However, with the deepening of related research, the current staging system can no longer fully meet the clinical needs. In this work, we propose a novel machine learning method for constructing an automatic hepatocellular carcinoma staging model that incorporates far more clinical variables than any existing staging system. Our model is based on random survival forests, which generates a unique hazard function for each patient. B-splines are used to embed hazard functions into vectors in low-dimensional space and hierarchical clustering method groups similar patients to form staging cohorts. The resulting staging system significantly outperforms the BCLC system in terms of distinctiveness between patients in different stages.

Disease knowledge graphs have emerged as a powerful tool for artificial intelligence to connect, organize, and access diverse information about diseases. Relations between disease concepts are often distributed across multiple datasets, including unstructured plain text datasets and incomplete disease knowledge graphs. Extracting disease relations from multimodal data sources is thus crucial for constructing accurate and comprehensive disease knowledge graphs. We introduce REMAP, a multimodal approach for disease relation extraction. The REMAP machine learning approach jointly embeds a partial, incomplete knowledge graph and a medical language dataset into a compact latent vector space, aligning the multimodal embeddings for optimal disease relation extraction. Additionally, REMAP utilizes a decoupled model structure to enable inference in single-modal data, which can be applied under missing modality scenarios. We apply the REMAP approach to a disease knowledge graph with 96,913 relations and a text dataset of 1.24 million sentences. On a dataset annotated by human experts, REMAP improves language-based disease relation extraction by 10.0% (accuracy) and 17.2% (F1-score) by fusing disease knowledge graphs with language information. Furthermore, REMAP leverages text information to recommend new relationships in the knowledge graph, outperforming graph-based methods by 8.4% (accuracy) and 10.4% (F1-score). REMAP is a flexible multimodal approach for extracting disease relations by fusing structured knowledge and language information. This approach provides a powerful model to easily find, access, and evaluate relations between disease concepts.

BackgroundDifferentiating between Crohn's disease (CD) and intestinal tuberculosis (ITB) with endoscopy is challenging. We aim to perform more accurate endoscopic diagnosis between CD and ITB by building a trustworthy AI differential diagnosis application.MethodsA total of 1271 electronic health record (EHR) patients who had undergone colonoscopies at Peking Union Medical College Hospital (PUMCH) and were clinically diagnosed with CD (n = 875) or ITB (n = 396) were used in this study. We build a workflow to make diagnoses with EHRs and mine differential diagnosis features; this involves finetuning the pretrained language models, distilling them into a light and efficient TextCNN model, interpreting the neural network and selecting differential attribution features, and then adopting manual feature checking and carrying out debias training.ResultsThe accuracy of debiased TextCNN on differential diagnosis between CD and ITB is 0.83 (CR F1: 0.87, ITB F1: 0.77), which is the best among the baselines. On the noisy validation set, its accuracy was 0.70 (CR F1: 0.87, ITB: 0.69), which was significantly higher than that of models without debias. We also find that the debiased model more easily mines the diagnostically significant features. The debiased TextCNN unearthed 39 diagnostic features in the form of phrases, 17 of which were key diagnostic features recognized by the guidelines.ConclusionWe build a trustworthy AI differential diagnosis application for differentiating between CD and ITB focusing on accuracy, interpretability and robustness. The classifiers perform well, and the features which had statistical significance were in agreement with clinical guidelines.