Qingyu Chen

Qingyu Chen, Jingcheng Du, Alexis Allot et al.

The rapid growth of biomedical literature poses a significant challenge for curation and interpretation. This has become more evident during the COVID-19 pandemic. LitCovid, a literature database of COVID-19 related papers in PubMed, has accumulated over 180,000 articles with millions of accesses. Approximately 10,000 new articles are added to LitCovid every month. A main curation task in LitCovid is topic annotation where an article is assigned with up to eight topics, e.g., Treatment and Diagnosis. The annotated topics have been widely used both in LitCovid (e.g., accounting for ~18% of total uses) and downstream studies such as network generation. However, it has been a primary curation bottleneck due to the nature of the task and the rapid literature growth. This study proposes LITMC-BERT, a transformer-based multi-label classification method in biomedical literature. It uses a shared transformer backbone for all the labels while also captures label-specific features and the correlations between label pairs. We compare LITMC-BERT with three baseline models on two datasets. Its micro-F1 and instance-based F1 are 5% and 4% higher than the current best results, respectively, and only requires ~18% of the inference time than the Binary BERT baseline. The related datasets and models are available via https://github.com/ncbi/ml-transformer.

Shubo Tian, Qiao Jin, Lana Yeganova et al. · tsinghua

ChatGPT has drawn considerable attention from both the general public and domain experts with its remarkable text generation capabilities. This has subsequently led to the emergence of diverse applications in the field of biomedicine and health. In this work, we examine the diverse applications of large language models (LLMs), such as ChatGPT, in biomedicine and health. Specifically we explore the areas of biomedical information retrieval, question answering, medical text summarization, information extraction, and medical education, and investigate whether LLMs possess the transformative power to revolutionize these tasks or whether the distinct complexities of biomedical domain presents unique challenges. Following an extensive literature survey, we find that significant advances have been made in the field of text generation tasks, surpassing the previous state-of-the-art methods. For other applications, the advances have been modest. Overall, LLMs have not yet revolutionized biomedicine, but recent rapid progress indicates that such methods hold great potential to provide valuable means for accelerating discovery and improving health. We also find that the use of LLMs, like ChatGPT, in the fields of biomedicine and health entails various risks and challenges, including fabricated information in its generated responses, as well as legal and privacy concerns associated with sensitive patient data. We believe this survey can provide a comprehensive and timely overview to biomedical researchers and healthcare practitioners on the opportunities and challenges associated with using ChatGPT and other LLMs for transforming biomedicine and health.

Ling Luo, Chih-Hsuan Wei, Po-Ting Lai et al.

The automatic assignment of species information to the corresponding genes in a research article is a critically important step in the gene normalization task, whereby a gene mention is normalized and linked to a database record or identifier by a text-mining algorithm. Existing methods typically rely on heuristic rules based on gene and species co-occurrence in the article, but their accuracy is suboptimal. We therefore developed a high-performance method, using a novel deep learning-based framework, to classify whether there is a relation between a gene and a species. Instead of the traditional binary classification framework in which all possible pairs of genes and species in the same article are evaluated, we treat the problem as a sequence-labeling task such that only a fraction of the pairs needs to be considered. Our benchmarking results show that our approach obtains significantly higher performance compared to that of the rule-based baseline method for the species assignment task (from 65.8% to 81.3% in accuracy). The source code and data for species assignment are freely available at https://github.com/ncbi/SpeciesAssignment.

Li Fang, Qingyu Chen, Chih-Hsuan Wei et al.

Pretrained language models such as Bidirectional Encoder Representations from Transformers (BERT) have achieved state-of-the-art performance in natural language processing (NLP) tasks. Recently, BERT has been adapted to the biomedical domain. Despite the effectiveness, these models have hundreds of millions of parameters and are computationally expensive when applied to large-scale NLP applications. We hypothesized that the number of parameters of the original BERT can be dramatically reduced with minor impact on performance. In this study, we present Bioformer, a compact BERT model for biomedical text mining. We pretrained two Bioformer models (named Bioformer8L and Bioformer16L) which reduced the model size by 60% compared to BERTBase. Bioformer uses a biomedical vocabulary and was pre-trained from scratch on PubMed abstracts and PubMed Central full-text articles. We thoroughly evaluated the performance of Bioformer as well as existing biomedical BERT models including BioBERT and PubMedBERT on 15 benchmark datasets of four different biomedical NLP tasks: named entity recognition, relation extraction, question answering and document classification. The results show that with 60% fewer parameters, Bioformer16L is only 0.1% less accurate than PubMedBERT while Bioformer8L is 0.9% less accurate than PubMedBERT. Both Bioformer16L and Bioformer8L outperformed BioBERTBase-v1.1. In addition, Bioformer16L and Bioformer8L are 2-3 fold as fast as PubMedBERT/BioBERTBase-v1.1. Bioformer has been successfully deployed to PubTator Central providing gene annotations over 35 million PubMed abstracts and 5 million PubMed Central full-text articles. We make Bioformer publicly available via https://github.com/WGLab/bioformer, including pre-trained models, datasets, and instructions for downstream use.

Qiao Jin, Yifan Yang, Qingyu Chen et al. · tsinghua

While large language models (LLMs) have been successfully applied to various tasks, they still face challenges with hallucinations. Augmenting LLMs with domain-specific tools such as database utilities can facilitate easier and more precise access to specialized knowledge. In this paper, we present GeneGPT, a novel method for teaching LLMs to use the Web APIs of the National Center for Biotechnology Information (NCBI) for answering genomics questions. Specifically, we prompt Codex to solve the GeneTuring tests with NCBI Web APIs by in-context learning and an augmented decoding algorithm that can detect and execute API calls. Experimental results show that GeneGPT achieves state-of-the-art performance on eight tasks in the GeneTuring benchmark with an average score of 0.83, largely surpassing retrieval-augmented LLMs such as the new Bing (0.44), biomedical LLMs such as BioMedLM (0.08) and BioGPT (0.04), as well as GPT-3 (0.16) and ChatGPT (0.12). Our further analyses suggest that: (1) API demonstrations have good cross-task generalizability and are more useful than documentations for in-context learning; (2) GeneGPT can generalize to longer chains of API calls and answer multi-hop questions in GeneHop, a novel dataset introduced in this work; (3) Different types of errors are enriched in different tasks, providing valuable insights for future improvements.

Qiao Jin, Won Kim, Qingyu Chen et al. · tsinghua

Information retrieval (IR) is essential in biomedical knowledge acquisition and clinical decision support. While recent progress has shown that language model encoders perform better semantic retrieval, training such models requires abundant query-article annotations that are difficult to obtain in biomedicine. As a result, most biomedical IR systems only conduct lexical matching. In response, we introduce MedCPT, a first-of-its-kind Contrastively Pre-trained Transformer model for zero-shot semantic IR in biomedicine. To train MedCPT, we collected an unprecedented scale of 255 million user click logs from PubMed. With such data, we use contrastive learning to train a pair of closely-integrated retriever and re-ranker. Experimental results show that MedCPT sets new state-of-the-art performance on six biomedical IR tasks, outperforming various baselines including much larger models such as GPT-3-sized cpt-text-XL. In addition, MedCPT also generates better biomedical article and sentence representations for semantic evaluations. As such, MedCPT can be readily applied to various real-world biomedical IR tasks.

Po-Ting Lai, Chih-Hsuan Wei, Ling Luo et al.

Biomedical relation extraction (RE) is the task of automatically identifying and characterizing relations between biomedical concepts from free text. RE is a central task in biomedical natural language processing (NLP) research and plays a critical role in many downstream applications, such as literature-based discovery and knowledge graph construction. State-of-the-art methods were used primarily to train machine learning models on individual RE datasets, such as protein-protein interaction and chemical-induced disease relation. Manual dataset annotation, however, is highly expensive and time-consuming, as it requires domain knowledge. Existing RE datasets are usually domain-specific or small, which limits the development of generalized and high-performing RE models. In this work, we present a novel framework for systematically addressing the data heterogeneity of individual datasets and combining them into a large dataset. Based on the framework and dataset, we report on BioREx, a data-centric approach for extracting relations. Our evaluation shows that BioREx achieves significantly higher performance than the benchmark system trained on the individual dataset, setting a new SOTA from 74.4% to 79.6% in F-1 measure on the recently released BioRED corpus. We further demonstrate that the combined dataset can improve performance for five different RE tasks. In addition, we show that on average BioREx compares favorably to current best-performing methods such as transfer learning and multi-task learning. Finally, we demonstrate BioREx's robustness and generalizability in two independent RE tasks not previously seen in training data: drug-drug N-ary combination and document-level gene-disease RE. The integrated dataset and optimized method have been packaged as a stand-alone tool available at https://github.com/ncbi/BioREx.

Rui Yang, Qingcheng Zeng, Keen You et al.

This study introduces Ascle, a pioneering natural language processing (NLP) toolkit designed for medical text generation. Ascle is tailored for biomedical researchers and healthcare professionals with an easy-to-use, all-in-one solution that requires minimal programming expertise. For the first time, Ascle evaluates and provides interfaces for the latest pre-trained language models, encompassing four advanced and challenging generative functions: question-answering, text summarization, text simplification, and machine translation. In addition, Ascle integrates 12 essential NLP functions, along with query and search capabilities for clinical databases. The toolkit, its models, and associated data are publicly available via https://github.com/Yale-LILY/MedGen.

Ehsan Misaghi, Sean T Berkowitz, Bing Yu Chen et al.

Large language models (LLMs) show potential for ophthalmic clinical reasoning, yet individual models risk introducing harm. We evaluated whether multi-agent LLM deliberative councils improve diagnostic performance and mitigate harm compared to individual LLMs. In a comparative cross-sectional study, we assessed 12 individual LLMs and three multi-agent councils on 100 ophthalmology clinical vignettes. Each council comprised four models assembled by type: proprietary flagship, proprietary fast, and open-source. Models independently answered a vignette, anonymously ranked one another's responses, and a designated chair synthesized all responses and peer reviews into a final answer. Councils consistently outperformed pooled individual models across all three tiers. Accuracy improved for proprietary flagship (95.0% vs 90.8%; risk difference [RD]: 4.25 [95% CI: 0.45, 8.05]), proprietary fast (96.0% vs 86.5%; RD: 9.50 [5.31, 13.59]), and open-source councils (91.0% vs 83.2%; RD: 7.75 [4.17, 11.33]). Harm rates declined for proprietary flagship (10.0% vs 22.5%; RD: -12.50 [-16.86, -8.14]), proprietary fast (16.0% vs 31.8%; RD: -15.75 [-21.49, -10.01]), and open-source councils (22.0% vs 38.5%; RD: -16.50 [-22.27, -10.73]). Coverage analysis revealed net positive gains for accuracy (ΔCoverage: 4.4-9.8 percentage points) and safety (ΔCoverage: 13.6-20.6), indicating councils recovered correct diagnoses and averted harm. Councils elevated correct diagnoses to higher rank positions; and produced more complete differentials and management plans (all P<.05). Harmful council responses showed reduced combined commission-and-omission errors and tended to be less severe. Structured deliberation via multi-agent LLM councils may enhance the reliability of LLM-assisted ophthalmic clinical reasoning.

Ling Luo, Chih-Hsuan Wei, Po-Ting Lai et al.

Biomedical named entity recognition (BioNER) seeks to automatically recognize biomedical entities in natural language text, serving as a necessary foundation for downstream text mining tasks and applications such as information extraction and question answering. Manually labeling training data for the BioNER task is costly, however, due to the significant domain expertise required for accurate annotation. The resulting data scarcity causes current BioNER approaches to be prone to overfitting, to suffer from limited generalizability, and to address a single entity type at a time (e.g., gene or disease). We therefore propose a novel all-in-one (AIO) scheme that uses external data from existing annotated resources to enhance the accuracy and stability of BioNER models. We further present AIONER, a general-purpose BioNER tool based on cutting-edge deep learning and our AIO schema. We evaluate AIONER on 14 BioNER benchmark tasks and show that AIONER is effective, robust, and compares favorably to other state-of-the-art approaches such as multi-task learning. We further demonstrate the practical utility of AIONER in three independent tasks to recognize entity types not previously seen in training data, as well as the advantages of AIONER over existing methods for processing biomedical text at a large scale (e.g., the entire PubMed data).

Robert Leaman, Rezarta Islamaj, Alexis Allot et al.

A significant percentage of COVID-19 survivors experience ongoing multisystemic symptoms that often affect daily living, a condition known as Long Covid or post-acute-sequelae of SARS-CoV-2 infection. However, identifying scientific articles relevant to Long Covid is challenging since there is no standardized or consensus terminology. We developed an iterative human-in-the-loop machine learning framework combining data programming with active learning into a robust ensemble model, demonstrating higher specificity and considerably higher sensitivity than other methods. Analysis of the Long Covid collection shows that (1) most Long Covid articles do not refer to Long Covid by any name (2) when the condition is named, the name used most frequently in the literature is Long Covid, and (3) Long Covid is associated with disorders in a wide variety of body systems. The Long Covid collection is updated weekly and is searchable online at the LitCovid portal: https://www.ncbi.nlm.nih.gov/research/coronavirus/docsum?filters=e_condition.LongCovid

Qingyu Chen, Alexis Allot, Robert Leaman et al.

The COVID-19 pandemic has been severely impacting global society since December 2019. Massive research has been undertaken to understand the characteristics of the virus and design vaccines and drugs. The related findings have been reported in biomedical literature at a rate of about 10,000 articles on COVID-19 per month. Such rapid growth significantly challenges manual curation and interpretation. For instance, LitCovid is a literature database of COVID-19-related articles in PubMed, which has accumulated more than 200,000 articles with millions of accesses each month by users worldwide. One primary curation task is to assign up to eight topics (e.g., Diagnosis and Treatment) to the articles in LitCovid. Despite the continuing advances in biomedical text mining methods, few have been dedicated to topic annotations in COVID-19 literature. To close the gap, we organized the BioCreative LitCovid track to call for a community effort to tackle automated topic annotation for COVID-19 literature. The BioCreative LitCovid dataset, consisting of over 30,000 articles with manually reviewed topics, was created for training and testing. It is one of the largest multilabel classification datasets in biomedical scientific literature. 19 teams worldwide participated and made 80 submissions in total. Most teams used hybrid systems based on transformers. The highest performing submissions achieved 0.8875, 0.9181, and 0.9394 for macro F1-score, micro F1-score, and instance-based F1-score, respectively. The level of participation and results demonstrate a successful track and help close the gap between dataset curation and method development. The dataset is publicly available via https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/ for benchmarking and further development.

Qingyu Chen, Tiarnan D L Keenan, Elvira Agron et al.

Timely disease diagnosis is challenging due to increasing disease burdens and limited clinician availability. AI shows promise in diagnosis accuracy but faces real-world application issues due to insufficient validation in clinical workflows and diverse populations. This study addresses gaps in medical AI downstream accountability through a case study on age-related macular degeneration (AMD) diagnosis and severity classification. We designed and implemented an AI-assisted diagnostic workflow for AMD, comparing diagnostic performance with and without AI assistance among 24 clinicians from 12 institutions with real patient data sampled from the Age-Related Eye Disease Study (AREDS). Additionally, we demonstrated continual enhancement of an existing AI model by incorporating approximately 40,000 additional medical images (named AREDS2 dataset). The improved model was then systematically evaluated using both AREDS and AREDS2 test sets, as well as an external test set from Singapore. AI assistance markedly enhanced diagnostic accuracy and classification for 23 out of 24 clinicians, with the average F1-score increasing by 20% from 37.71 (Manual) to 45.52 (Manual + AI) (P-value < 0.0001), achieving an improvement of over 50% in some cases. In terms of efficiency, AI assistance reduced diagnostic times for 17 out of the 19 clinicians tracked, with time savings of up to 40%. Furthermore, a model equipped with continual learning showed robust performance across three independent datasets, recording a 29% increase in accuracy, and elevating the F1-score from 42 to 54 in the Singapore population.

Xinyue Hu, Lin Gu, Kazuma Kobayashi et al.

Medical visual question answering (VQA) aims to answer clinically relevant questions regarding input medical images. This technique has the potential to improve the efficiency of medical professionals while relieving the burden on the public health system, particularly in resource-poor countries. Existing medical VQA methods tend to encode medical images and learn the correspondence between visual features and questions without exploiting the spatial, semantic, or medical knowledge behind them. This is partially because of the small size of the current medical VQA dataset, which often includes simple questions. Therefore, we first collected a comprehensive and large-scale medical VQA dataset, focusing on chest X-ray images. The questions involved detailed relationships, such as disease names, locations, levels, and types in our dataset. Based on this dataset, we also propose a novel baseline method by constructing three different relationship graphs: spatial relationship, semantic relationship, and implicit relationship graphs on the image regions, questions, and semantic labels. The answer and graph reasoning paths are learned for different questions.

Aidan Gilson, Xuguang Ai, Thilaka Arunachalam et al.

Despite the potential of Large Language Models (LLMs) in medicine, they may generate responses lacking supporting evidence or based on hallucinated evidence. While Retrieval Augment Generation (RAG) is popular to address this issue, few studies implemented and evaluated RAG in downstream domain-specific applications. We developed a RAG pipeline with 70,000 ophthalmology-specific documents that retrieve relevant documents to augment LLMs during inference time. In a case study on long-form consumer health questions, we systematically evaluated the responses including over 500 references of LLMs with and without RAG on 100 questions with 10 healthcare professionals. The evaluation focuses on factuality of evidence, selection and ranking of evidence, attribution of evidence, and answer accuracy and completeness. LLMs without RAG provided 252 references in total. Of which, 45.3% hallucinated, 34.1% consisted of minor errors, and 20.6% were correct. In contrast, LLMs with RAG significantly improved accuracy (54.5% being correct) and reduced error rates (18.8% with minor hallucinations and 26.7% with errors). 62.5% of the top 10 documents retrieved by RAG were selected as the top references in the LLM response, with an average ranking of 4.9. The use of RAG also improved evidence attribution (increasing from 1.85 to 2.49 on a 5-point scale, P<0.001), albeit with slight decreases in accuracy (from 3.52 to 3.23, P=0.03) and completeness (from 3.47 to 3.27, P=0.17). The results demonstrate that LLMs frequently exhibited hallucinated and erroneous evidence in the responses, raising concerns for downstream applications in the medical domain. RAG substantially reduced the proportion of such evidence but encountered challenges.

FSVideo Team, Qingyu Chen, Zhiyuan Fang et al.

We introduce FSVideo, a fast speed transformer-based image-to-video (I2V) diffusion framework. We build our framework on the following key components: 1.) a new video autoencoder with highly-compressed latent space ($64\times64\times4$ spatial-temporal downsampling ratio), achieving competitive reconstruction quality; 2.) a diffusion transformer (DIT) architecture with a new layer memory design to enhance inter-layer information flow and context reuse within DIT, and 3.) a multi-resolution generation strategy via a few-step DIT upsampler to increase video fidelity. Our final model, which contains a 14B DIT base model and a 14B DIT upsampler, achieves competitive performance against other popular open-source models, while being an order of magnitude faster. We discuss our model design as well as training strategies in this report.

Zhenyue Qin, Younjoon Chung, Elijah Lee et al.

Vision impairment affects millions globally, and early detection is critical to preventing irreversible vision loss. Ophthalmology workflows require clinicians to integrate medical images, structured clinical data, and free-text notes to determine disease severity and management, which is time-consuming and burdensome. Recent multimodal large language models (MLLMs) show promise, but existing general and medical MLLMs perform poorly in ophthalmology, and few ophthalmology-specific MLLMs are openly available. We present VOLMO (Versatile and Open Large Models for Ophthalmology), a model-agnostic, data-open framework for developing ophthalmology-specific MLLMs. VOLMO includes three stages: ophthalmology knowledge pretraining on 86,965 image-text pairs from 26,569 articles across 82 journals; domain task fine-tuning on 26,929 annotated instances spanning 12 eye conditions for disease screening and severity classification; and multi-step clinical reasoning on 913 patient case reports for assessment, planning, and follow-up care. Using this framework, we trained a compact 2B-parameter MLLM and compared it with strong baselines, including InternVL-2B, LLaVA-Med-7B, MedGemma-4B, MedGemma-27B, and RETFound. We evaluated these models on image description generation, disease screening and staging classification, and assessment-and-management generation, with additional manual review by two healthcare professionals and external validation on three independent cohorts for age-related macular degeneration and diabetic retinopathy. Across settings, VOLMO-2B consistently outperformed baselines, achieving stronger image description performance, an average F1 of 87.4% across 12 eye conditions, and higher scores in external validation.

Rui Yang, Edison Marrese-Taylor, Yuhe Ke et al.

Large language models (LLMs) have demonstrated powerful text generation capabilities, bringing unprecedented innovation to the healthcare field. While LLMs hold immense promise for applications in healthcare, applying them to real clinical scenarios presents significant challenges, as these models may generate content that deviates from established medical facts and even exhibit potential biases. In our research, we develop an augmented LLM framework based on the Unified Medical Language System (UMLS), aiming to better serve the healthcare community. We employ LLaMa2-13b-chat and ChatGPT-3.5 as our benchmark models, and conduct automatic evaluations using the ROUGE Score and BERTScore on 104 questions from the LiveQA test set. Additionally, we establish criteria for physician-evaluation based on four dimensions: Factuality, Completeness, Readability and Relevancy. ChatGPT-3.5 is used for physician evaluation with 20 questions on the LiveQA test set. Multiple resident physicians conducted blind reviews to evaluate the generated content, and the results indicate that this framework effectively enhances the factuality, completeness, and relevance of generated content. Our research demonstrates the effectiveness of using UMLS-augmented LLMs and highlights the potential application value of LLMs in in medical question-answering.

Anran Li, Yuanyuan Chen, Wenjun Long et al.

Large language models (LLMs) have demonstrated strong performance on medical benchmarks, including question answering and diagnosis. To enable their use in clinical settings, LLMs are typically further adapted through continued pretraining or post-training using clinical data. However, most medical LLMs are trained on data from a single institution, which faces limitations in generalizability and safety in heterogeneous systems. Federated learning (FL) is a promising solution for enabling collaborative model development across healthcare institutions. Yet applying FL to LLMs in medicine remains fundamentally limited. First, conventional FL requires transmitting the full model during each communication round, which becomes impractical for multi-billion-parameter LLMs given the limited computational resources. Second, many FL algorithms implicitly assume data homogeneity, whereas real-world clinical data are highly heterogeneous across patients, diseases, and institutional practices. We introduce the model-agnostic and parameter-efficient federated learning framework for adapting LLMs to medical applications. Fed-MedLoRA transmits only low-rank adapter parameters, reducing communication and computation overhead, while Fed-MedLoRA+ further incorporates adaptive, data-aware aggregation to improve convergence under cross-site heterogeneity. We apply the framework to clinical information extraction (IE), which transforms patient narratives into structured medical entities and relations. Accuracy was assessed across five patient cohorts through comparisons with BERT models, and LLaMA-3 and DeepSeek-R1, GPT-4o models. Evaluation settings included (1) in-domain training and testing, (2) external validation on independent cohorts, and (3) a low-resource new-site adaptation scenario using real-world clinical notes from the Yale New Haven Health System.

Roy Jiang, Hyunjae Kim, Zhenyue Qin et al.

Multimodal large language models (MLLMs) have demonstrated promise on publicly available dermatology benchmarks. However, benchmark performance may not generalize to real-world dermatologic decision-making. To quantify this benchmark-to-bedside gap, we evaluated four open-weight MLLMs (InternVL-Chat v1.5, LLaVA-Med v1.5, SkinGPT4 and MedGemma-4B-Instruct) and one commercial MLLM (GPT-4.1) across three publicly available dermatology datasets and a retrospective multi-site hospital-based dermatology consultation cohort comprising 5,811 cases and 46,405 clinical images. Models were evaluated on two clinically relevant tasks: differential diagnosis generation and severity-based triage. Diagnostic performance was modest on public datasets and declined substantially in the real-world cohort. On public benchmarks, top-3 diagnostic accuracy reached 26.55% for the best open-weight model and 42.25% for GPT-4.1. On real-world consultation cases using images alone, top-3 diagnostic accuracy fell to 1.50%-13.35% among open-weight models and 24.65% for GPT-4.1. Incorporating clinical context improved performance across all models, increasing top-3 diagnostic accuracy up to 28.75% among open-weight models and 38.93% for GPT-4.1. However, model outputs were highly sensitive to incomplete or erroneous consultation context. For severity-based triage, models achieved moderate sensitivity (above 60%), suggesting potential utility for screening but insufficient reliability for clinical deployment. These findings demonstrate that benchmark performance substantially overestimates the real-world clinical capability of current dermatology MLLMs.

Rui Yang, Huitao Li, Weihao Xuan et al.

Despite continuous advances in medical technology, the global distribution of health care resources remains uneven. The development of large language models (LLMs) has transformed the landscape of medicine and holds promise for improving health care quality and expanding access to medical information globally. However, existing LLMs are primarily trained on high-resource languages, limiting their applicability in global medical scenarios. To address this gap, we constructed GlobMed, a large multilingual medical dataset, containing over 500,000 entries spanning 12 languages, including four low-resource languages. Building on this, we established GlobMed-Bench, which systematically assesses 56 state-of-the-art proprietary and open-weight LLMs across multiple multilingual medical tasks, revealing significant performance disparities across languages, particularly for low-resource languages. Additionally, we introduced GlobMed-LLMs, a suite of multilingual medical LLMs trained on GlobMed, with parameters ranging from 1.7B to 8B. GlobMed-LLMs achieved an average performance improvement of over 40% relative to baseline models, with a more than threefold increase in performance on low-resource languages. Together, these resources provide an important foundation for advancing the equitable development and application of LLMs globally, enabling broader language communities to benefit from technological advances.

Tianyu Liu, Weihao Xuan, Hao Wu et al.

Advances in AI have introduced several strong models in computational pathology to usher it into the era of multi-modal diagnosis, analysis, and interpretation. However, the current pathology-specific visual language models still lack capacities in making the diagnosis with rigorous reasoning paths as well as handling divergent tasks, and thus, challenges of building AI Copilots for real scenarios still exist. Here we introduce TeamPath, an AI system powered by reinforcement learning and router-enhanced solutions based on large-scale histopathology multimodal datasets, to work as a virtual assistant for expert-level disease diagnosis, patch-level information summarization, and cross-modality generation to integrate transcriptomic information for clinical usage. We also collaborate with pathologists from Yale School of Medicine to demonstrate that TeamPath can assist them in working more efficiently by identifying and correcting expert conclusions and reasoning paths. We also discuss the human evaluation results to support the reasoning quality from TeamPath. Overall, TeamPath can flexibly choose the best settings according to the needs, and serve as an innovative and reliable system for information communication across different modalities and experts.

Yuang Fan, Lilin Xu, Millie Wu et al.

Longitudinal passive sensing enables continuous health prediction, yet models often fail under cross-dataset distribution shifts. Traditional ML overfits cohort-specific artifacts, while Large Language Models (LLMs) struggle to reason reliably over long, heterogeneous time-series. We introduce TimeSRL, a two-stage LLM framework that routes predictions through an explicit semantic bottleneck. The model first abstracts raw signals into high-level natural language, then predicts behavioral outcomes from these abstractions alone. This forces the model to reason over semantic concepts that we argue generalize better than raw numbers. We optimize this process end-to-end using Group Relative Policy Optimization (GRPO) with Reinforcement Learning from Verifiable Rewards (RLVR), learning outcome-aligned abstractions without gold intermediate annotations. Instantiated on mental-health prediction, TimeSRL achieves state-of-the-art performance on a benchmark designed to stress-test cross-cohort generalization under a rigorous leave-one-dataset-out (LOSO) protocol, reducing mean absolute error (MAE) over strong non-LLM ML and LLM baselines by 3.1--10.1% and 9.5--44.1% for anxiety, and 3.2--9.6% and 27.4--57.6% for depression (all $p$s<0.05). TimeSRL significantly outperforms prior methods in cross-benchmark transfer across different sensing pipelines, rivaling its own within-domain performance without target-domain fine-tuning. These results demonstrate that semantic abstractions are reusable and point to a new direction for generalizable behavior modeling via RL-tuned LLMs.

Yan Hu, Qingyu Chen, Jingcheng Du et al.

Objective: This study quantifies the capabilities of GPT-3.5 and GPT-4 for clinical named entity recognition (NER) tasks and proposes task-specific prompts to improve their performance. Materials and Methods: We evaluated these models on two clinical NER tasks: (1) to extract medical problems, treatments, and tests from clinical notes in the MTSamples corpus, following the 2010 i2b2 concept extraction shared task, and (2) identifying nervous system disorder-related adverse events from safety reports in the vaccine adverse event reporting system (VAERS). To improve the GPT models' performance, we developed a clinical task-specific prompt framework that includes (1) baseline prompts with task description and format specification, (2) annotation guideline-based prompts, (3) error analysis-based instructions, and (4) annotated samples for few-shot learning. We assessed each prompt's effectiveness and compared the models to BioClinicalBERT. Results: Using baseline prompts, GPT-3.5 and GPT-4 achieved relaxed F1 scores of 0.634, 0.804 for MTSamples, and 0.301, 0.593 for VAERS. Additional prompt components consistently improved model performance. When all four components were used, GPT-3.5 and GPT-4 achieved relaxed F1 socres of 0.794, 0.861 for MTSamples and 0.676, 0.736 for VAERS, demonstrating the effectiveness of our prompt framework. Although these results trail BioClinicalBERT (F1 of 0.901 for the MTSamples dataset and 0.802 for the VAERS), it is very promising considering few training samples are needed. Conclusion: While direct application of GPT models to clinical NER tasks falls short of optimal performance, our task-specific prompt framework, incorporating medical knowledge and training samples, significantly enhances GPT models' feasibility for potential clinical applications.

Yan Hu, Xu Zuo, Yujia Zhou et al.

Backgrounds: Information extraction (IE) is critical in clinical natural language processing (NLP). While large language models (LLMs) excel on generative tasks, their performance on extractive tasks remains debated. Methods: We investigated Named Entity Recognition (NER) and Relation Extraction (RE) using 1,588 clinical notes from four sources (UT Physicians, MTSamples, MIMIC-III, and i2b2). We developed an annotated corpus covering 4 clinical entities and 16 modifiers, and compared instruction-tuned LLaMA-2 and LLaMA-3 against BERT in terms of performance, generalizability, computational resources, and throughput to BERT. Results: LLaMA models outperformed BERT across datasets. With sufficient training data, LLaMA showed modest improvements (1% on NER, 1.5-3.7% on RE); improvements were larger with limited training data. On unseen i2b2 data, LLaMA-3-70B outperformed BERT by 7% (F1) on NER and 4% on RE. However, LLaMA models required more computing resources and ran up to 28 times slower. We implemented "Kiwi," a clinical IE package featuring both models, available at https://kiwi.clinicalnlp.org/. Conclusion: This study is among the first to develop and evaluate a comprehensive clinical IE system using open-source LLMs. Results indicate that LLaMA models outperform BERT for clinical NER and RE but with higher computational costs and lower throughputs. These findings highlight that choosing between LLMs and traditional deep learning methods for clinical IE applications should remain task-specific, taking into account both performance metrics and practical considerations such as available computing resources and the intended use case scenarios.

Lingfei Qian, Mauro Giuffre, Yan Wang et al.

Clinical decision-making increasingly relies on timely and context-aware access to patient information within Electronic Health Records (EHRs), yet most existing natural language question-answering (QA) systems are evaluated solely on benchmark datasets, limiting their practical relevance. To overcome this limitation, we introduce EHRNavigator, a multi-agent framework that harnesses AI agents to perform patient-level question answering across heterogeneous and multimodal EHR data. We assessed its performance using both public benchmark and institutional datasets under realistic hospital conditions characterized by diverse schemas, temporal reasoning demands, and multimodal evidence integration. Through quantitative evaluation and clinician-validated chart review, EHRNavigator demonstrated strong generalization, achieving 86% accuracy on real-world cases while maintaining clinically acceptable response times. Overall, these findings confirm that EHRNavigator effectively bridges the gap between benchmark evaluation and clinical deployment, offering a robust, adaptive, and efficient solution for real-world EHR question answering.

Anran Li, Yuanyuan Chen, Chao Ren et al.

For privacy-preserving graph learning tasks involving distributed graph datasets, federated learning (FL)-based GCN (FedGCN) training is required. A key challenge for FedGCN is scaling to large-scale graphs, which typically incurs high computation and communication costs when dealing with the explosively increasing number of neighbors. Existing graph sampling-enhanced FedGCN training approaches ignore graph structural information or dynamics of optimization, resulting in high variance and inaccurate node embeddings. To address this limitation, we propose the Federated Adaptive Importance-based Sampling (FedAIS) approach. It achieves substantial computational cost saving by focusing the limited resources on training important nodes, while reducing communication overhead via adaptive historical embedding synchronization. The proposed adaptive importance-based sampling method jointly considers the graph structural heterogeneity and the optimization dynamics to achieve optimal trade-off between efficiency and accuracy. Extensive evaluations against five state-of-the-art baselines on five real-world graph datasets show that FedAIS achieves comparable or up to 3.23% higher test accuracy, while saving communication and computation costs by 91.77% and 85.59%.

Qianqian Xie, Qingyu Chen, Aokun Chen et al.

Recent advancements in large language models (LLMs) like ChatGPT and LLaMA show promise in medical applications, yet challenges remain in medical language comprehension. This study presents Me-LLaMA, a new medical LLM family based on open-source LLaMA models, optimized for medical text analysis and diagnosis by leveraging large-scale, domain-specific datasets. The Me-LLaMA family, including foundation models Me-LLaMA 13/70B and their chat-enhanced versions, was developed through continued pre-training and instruction tuning with 129B tokens and 214K samples from biomedical and clinical sources. Training the 70B models required over 100,000 A100 GPU hours. Me-LLaMA's performance was evaluated across six medical text analysis tasks using 12 benchmark datasets and complex clinical case diagnosis, with automatic and human evaluations. Results indicate Me-LLaMA outperforms LLaMA and other open-source medical LLMs in zero-shot and supervised settings. Task-specific tuning further boosts performance, surpassing ChatGPT on 7 of 8 datasets and GPT-4 on 5 of 8. For complex clinical cases, Me-LLaMA achieves performance comparable to ChatGPT and GPT-4. This work underscores the importance of domain-specific data in developing medical LLMs and addresses the high computational costs involved in training, highlighting a balance between pre-training and fine-tuning strategies. Me-LLaMA models are now accessible under user agreements, providing a valuable resource for advancing medical AI.

Peisong Zhang, Manqiang Peng, Yuxuan Wu et al.

Estimating individualized treatment effects from longitudinal observational data is central to data-driven medicine, yet existing methods face a fundamental limitation: reducing confounding bias often suppresses clinically informative heterogeneity, degrading patient-specific predictions. Here, we identify this tension as a bias-precision paradox in causal representation learning and introduce sampling-based maximum mean discrepancy (sMMD), a stochastic alignment strategy that replaces global adversarial balancing with subset-level matching. We instantiate this approach in a framework for counterfactual outcome prediction with attribution-grounded interpretability. Across two large-scale ICU cohorts (n = 27,783), our framework improves accuracy under distribution shift, reducing error by up to 11.5% and substantially increasing recall in high-risk tasks. Mechanistic analyses show that sMMD selectively preserves clinically decisive variables. In human-AI evaluation, our method outperforms clinicians-in-training and large language models, and improves clinician accuracy by 14.7% while reducing decision time, enabling interpretable, real-time clinical decision support.

Kam Man Wu, Haolin Yang, Qingyu Chen et al.

Recent advances in image generation have made it easy to produce high-quality images. However, these outputs are inherently flattened, entangling foreground elements, background, and text within a fixed canvas. As a result, flexible post-generation editing remains challenging, revealing a clear last-mile gap toward practical usability. Existing approaches either rely on scarce proprietary layered assets or construct partially synthetic data from limited structural priors. However, both strategies face fundamental challenges in scalability. In this work, we investigate whether pure synthetic layered data can improve graphic design decomposition. We make the assumption that, in graphic design, effective decomposition does not require modeling inter-layer dependencies as precisely as in natural-image composition, since design elements are often intentionally arranged as modular and semantically separable components. Concretely, we conduct a data-centric study based on CLD baseline, which is a state-of-the-art layer decomposition framework. Based on the baseline, we construct our own synthetic dataset, SynLayers, generate textual supervision using vision language models, and automate inference inputs with VLM-predicted bounding boxes. Our study reveals three key findings: (1) even training with purely synthetic data can outperform non-scalable alternatives such as the widely used PrismLayersPro dataset, demonstrating its viability as a scalable and effective substitute; (2) performance consistently improves with increased training data scale, while gains begin to saturate at around 50K samples; and (3) synthetic data enables balanced control over layer-count distributions, avoiding the layer-count imbalance commonly observed in real-world datasets. We hope this data-centric study encourages broader adoption of synthetic data as a practical foundation for layered design editing systems.

Jaehoon Yun, Jiwoong Sohn, Jungwoo Park et al.

Large language models have shown promise in clinical decision making, but current approaches struggle to localize and correct errors at specific steps of the reasoning process. This limitation is critical in medicine, where identifying and addressing reasoning errors is essential for accurate diagnosis and effective patient care. We introduce Med-PRM, a process reward modeling framework that leverages retrieval-augmented generation to verify each reasoning step against established medical knowledge bases. By verifying intermediate reasoning steps with evidence retrieved from clinical guidelines and literature, our model can precisely assess the reasoning quality in a fine-grained manner. Evaluations on five medical QA benchmarks and two open-ended diagnostic tasks demonstrate that Med-PRM achieves state-of-the-art performance, with improving the performance of base models by up to 13.50% using Med-PRM. Moreover, we demonstrate the generality of Med-PRM by integrating it in a plug-and-play fashion with strong policy models such as Meerkat, achieving over 80\% accuracy on MedQA for the first time using small-scale models of 8 billion parameters. Our code and data are available at: https://med-prm.github.io/

Qingshan Hou, Yukun Zhou, Jocelyn Hui Lin Goh et al.

The advent of foundation models (FMs) is transforming medical domain. In ophthalmology, RETFound, a retina-specific FM pre-trained sequentially on 1.4 million natural images and 1.6 million retinal images, has demonstrated high adaptability across clinical applications. Conversely, DINOv2, a general-purpose vision FM pre-trained on 142 million natural images, has shown promise in non-medical domains. However, its applicability to clinical tasks remains underexplored. To address this, we conducted head-to-head evaluations by fine-tuning RETFound and three DINOv2 models (large, base, small) for ocular disease detection and systemic disease prediction tasks, across eight standardized open-source ocular datasets, as well as the Moorfields AlzEye and the UK Biobank datasets. DINOv2-large model outperformed RETFound in detecting diabetic retinopathy (AUROC=0.850-0.952 vs 0.823-0.944, across three datasets, all P<=0.007) and multi-class eye diseases (AUROC=0.892 vs. 0.846, P<0.001). In glaucoma, DINOv2-base model outperformed RETFound (AUROC=0.958 vs 0.940, P<0.001). Conversely, RETFound achieved superior performance over all DINOv2 models in predicting heart failure, myocardial infarction, and ischaemic stroke (AUROC=0.732-0.796 vs 0.663-0.771, all P<0.001). These trends persisted even with 10% of the fine-tuning data. These findings showcase the distinct scenarios where general-purpose and domain-specific FMs excel, highlighting the importance of aligning FM selection with task-specific requirements to optimise clinical performance.

Zihan Guan, Qiao Jin, Guangzhi Xiong et al.

The existing methods for evaluating the medical knowledge of Large Language Models (LLMs) are largely based on atemporal examination-style benchmarks, while in reality, medical knowledge is inherently dynamic and continuously evolves as new evidence emerges and treatments are approved. Consequently, evaluating medical knowledge without a temporal context may provide an incomplete assessment of whether LLMs can accurately reason about time-specific medical knowledge. Moreover, most medical data are historical, requiring the models not only to recall the correct knowledge, but also to know when that knowledge is correct. To bridge the gap, we built TempoMed-Bench, the first-of-its-kind benchmark for evaluating the temporal awareness of the LLMs in the medical domain through evolving guideline knowledge. Based on the TempoMed-Bench, our evaluation analysis first reveals that LLMs lack temporal awareness in medical knowledge through the key findings: (1) model performance on up-to-date medical knowledge exhibits a gradual linear decline over time rather than a sharp knowledge-cutoff behavior, suggesting that parametric medical knowledge is not strictly bounded by knowledge cutoffs; (2) LLMs consistently struggle more with recalling outdated historical medical knowledge than with up-to-date recommendations: accuracy of historical knowledge is only 25.37%-53.89% of up-to-date knowledge, indicating potential knowledge forgetting effects during training; and (3) LLMs often exhibit temporally inconsistent behaviors, where predictions fluctuate irregularly across neighboring years. We also show that the temporal awareness problem is a challenge that cannot be easily solved when integrated with agentic search tools (-3.15%-14.14%). This work highlights an important yet underexplored challenge and motivates future research on developing LLMs that can better encode time-specific medical knowledge.

Hyunjae Kim, Jiwoong Sohn, Aidan Gilson et al.

Large language models (LLMs) are transforming the landscape of medicine, yet two fundamental challenges persist: keeping up with rapidly evolving medical knowledge and providing verifiable, evidence-grounded reasoning. Retrieval-augmented generation (RAG) has been widely adopted to address these limitations by supplementing model outputs with retrieved evidence. However, whether RAG reliably achieves these goals remains unclear. Here, we present the most comprehensive expert evaluation of RAG in medicine to date. Eighteen medical experts contributed a total of 80,502 annotations, assessing 800 model outputs generated by GPT-4o and Llama-3.1-8B across 200 real-world patient and USMLE-style queries. We systematically decomposed the RAG pipeline into three components: (i) evidence retrieval (relevance of retrieved passages), (ii) evidence selection (accuracy of evidence usage), and (iii) response generation (factuality and completeness of outputs). Contrary to expectation, standard RAG often degraded performance: only 22% of top-16 passages were relevant, evidence selection remained weak (precision 41-43%, recall 27-49%), and factuality and completeness dropped by up to 6% and 5%, respectively, compared with non-RAG variants. Retrieval and evidence selection remain key failure points for the model, contributing to the overall performance drop. We further show that simple yet effective strategies, including evidence filtering and query reformulation, substantially mitigate these issues, improving performance on MedMCQA and MedXpertQA by up to 12% and 8.2%, respectively. These findings call for re-examining RAG's role in medicine and highlight the importance of stage-aware evaluation and deliberate system design for reliable medical LLM applications.

Fan Gao, Sherry T. Tong, Jiwoong Sohn et al.

While reasoning-enhanced large language models perform strongly on English medical tasks, a persistent multilingual gap remains, with substantially weaker reasoning in local languages, limiting equitable global medical deployment. To bridge this gap, we introduce Med-CoReasoner, a language-informed co-reasoning framework that elicits parallel English and local-language reasoning, abstracts them into structured concepts, and integrates local clinical knowledge into an English logical scaffold via concept-level alignment and retrieval. This design combines the structural robustness of English reasoning with the practice-grounded expertise encoded in local languages. To evaluate multilingual medical reasoning beyond multiple-choice settings, we construct MultiMed-X, a benchmark covering seven languages with expert-annotated long-form question answering and natural language inference tasks, comprising 350 instances per language. Experiments across three benchmarks show that Med-CoReasoner improves multilingual reasoning performance by an average of 5%, with particularly substantial gains in low-resource languages. Moreover, model distillation and expert evaluation analysis further confirm that Med-CoReasoner produces clinically sound and culturally grounded reasoning traces.

Dain Kim, Jiwoo Lee, Jaehoon Yun et al.

Large Vision-Language Models (LVLMs) hold significant promise for medical applications, yet their deployment is often constrained by insufficient alignment and reliability. While Direct Preference Optimization (DPO) has emerged as a potent framework for refining model responses, its efficacy in high-stakes medical contexts remains underexplored, lacking the rigorous empirical groundwork necessary to guide future methodological advances. To bridge this gap, we present the first comprehensive examination of diverse DPO variants within the medical domain, evaluating nine distinct formulations across two medical LVLMs: LLaVA-Med and HuatuoGPT-Vision. Our results reveal several critical limitations: current DPO approaches often yield inconsistent gains over supervised fine-tuning, with their efficacy varying significantly across different tasks and backbones. Furthermore, they frequently fail to resolve fundamental visual misinterpretation errors. Building on these insights, we present a targeted preference construction strategy as a proof-of-concept that explicitly addresses visual misinterpretation errors frequently observed in existing DPO models. This design yields a 3.6% improvement over the strongest existing DPO baseline on visual question-answering tasks. To support future research, we release our complete framework, including all training data, model checkpoints, and our codebase at https://github.com/dmis-lab/med-vlm-dpo.

Nicholas Wan, Qiao Jin, Joey Chan et al.

Although large language models (LLMs) have been assessed for general medical knowledge using licensing exams, their ability to support clinical decision-making, such as selecting medical calculators, remains uncertain. We assessed nine LLMs, including open-source, proprietary, and domain-specific models, with 1,009 multiple-choice question-answer pairs across 35 clinical calculators and compared LLMs to humans on a subset of questions. While the highest-performing LLM, OpenAI o1, provided an answer accuracy of 66.0% (CI: 56.7-75.3%) on the subset of 100 questions, two human annotators nominally outperformed LLMs with an average answer accuracy of 79.5% (CI: 73.5-85.0%). Ultimately, we evaluated medical trainees and LLMs in recommending medical calculators across clinical scenarios like risk stratification and diagnosis. With error analysis showing that the highest-performing LLMs continue to make mistakes in comprehension (49.3% of errors) and calculator knowledge (7.1% of errors), our findings highlight that LLMs are not superior to humans in calculator recommendation.

Qingyu Chen, Yan Hu, Xueqing Peng et al.

The rapid growth of biomedical literature poses challenges for manual knowledge curation and synthesis. Biomedical Natural Language Processing (BioNLP) automates the process. While Large Language Models (LLMs) have shown promise in general domains, their effectiveness in BioNLP tasks remains unclear due to limited benchmarks and practical guidelines. We perform a systematic evaluation of four LLMs, GPT and LLaMA representatives on 12 BioNLP benchmarks across six applications. We compare their zero-shot, few-shot, and fine-tuning performance with traditional fine-tuning of BERT or BART models. We examine inconsistencies, missing information, hallucinations, and perform cost analysis. Here we show that traditional fine-tuning outperforms zero or few shot LLMs in most tasks. However, closed-source LLMs like GPT-4 excel in reasoning-related tasks such as medical question answering. Open source LLMs still require fine-tuning to close performance gaps. We find issues like missing information and hallucinations in LLM outputs. These results offer practical insights for applying LLMs in BioNLP.

Yifan Peng, Qingyu Chen, Zhiyong Lu

Multi-task learning (MTL) has achieved remarkable success in natural language processing applications. In this work, we study a multi-task learning model with multiple decoders on varieties of biomedical and clinical natural language processing tasks such as text similarity, relation extraction, named entity recognition, and text inference. Our empirical results demonstrate that the MTL fine-tuned models outperform state-of-the-art transformer models (e.g., BERT and its variants) by 2.0% and 1.3% in biomedical and clinical domains, respectively. Pairwise MTL further demonstrates more details about which tasks can improve or decrease others. This is particularly helpful in the context that researchers are in the hassle of choosing a suitable model for new problems. The code and models are publicly available at https://github.com/ncbi-nlp/bluebert

Qingyu Chen, Kyubum Lee, Shankai Yan et al.

Capturing the semantics of related biological concepts, such as genes and mutations, is of significant importance to many research tasks in computational biology such as protein-protein interaction detection, gene-drug association prediction, and biomedical literature-based discovery. Here, we propose to leverage state-of-the-art text mining tools and machine learning models to learn the semantics via vector representations (aka. embeddings) of over 400,000 biological concepts mentioned in the entire PubMed abstracts. Our learned embeddings, namely BioConceptVec, can capture related concepts based on their surrounding contextual information in the literature, which is beyond exact term match or co-occurrence-based methods. BioConceptVec has been thoroughly evaluated in multiple bioinformatics tasks consisting of over 25 million instances from nine different biological datasets. The evaluation results demonstrate that BioConceptVec has better performance than existing methods in all tasks. Finally, BioConceptVec is made freely available to the research community and general public via https://github.com/ncbi-nlp/BioConceptVec.

Tiarnan D. Keenan, Shazia Dharssi, Yifan Peng et al.

Purpose: To assess the utility of deep learning in the detection of geographic atrophy (GA) from color fundus photographs; secondary aim to explore potential utility in detecting central GA (CGA). Design: A deep learning model was developed to detect the presence of GA in color fundus photographs, and two additional models to detect CGA in different scenarios. Participants: 59,812 color fundus photographs from longitudinal follow up of 4,582 participants in the AREDS dataset. Gold standard labels were from human expert reading center graders using a standardized protocol. Methods: A deep learning model was trained to use color fundus photographs to predict GA presence from a population of eyes with no AMD to advanced AMD. A second model was trained to predict CGA presence from the same population. A third model was trained to predict CGA presence from the subset of eyes with GA. For training and testing, 5-fold cross-validation was employed. For comparison with human clinician performance, model performance was compared with that of 88 retinal specialists. Results: The deep learning models (GA detection, CGA detection from all eyes, and centrality detection from GA eyes) had AUC of 0.933-0.976, 0.939-0.976, and 0.827-0.888, respectively. The GA detection model had accuracy, sensitivity, specificity, and precision of 0.965, 0.692, 0.978, and 0.584, respectively. The CGA detection model had equivalent values of 0.966, 0.763, 0.971, and 0.394. The centrality detection model had equivalent values of 0.762, 0.782, 0.729, and 0.799. Conclusions: A deep learning model demonstrated high accuracy for the automated detection of GA. The AUC was non-inferior to that of human retinal specialists. Deep learning approaches may also be applied to the identification of CGA. The code and pretrained models are publicly available at https://github.com/ncbi-nlp/DeepSeeNet.

Yifan Peng, Shazia Dharssi, Qingyu Chen et al.

In assessing the severity of age-related macular degeneration (AMD), the Age-Related Eye Disease Study (AREDS) Simplified Severity Scale predicts the risk of progression to late AMD. However, its manual use requires the time-consuming participation of expert practitioners. Although several automated deep learning systems have been developed for classifying color fundus photographs (CFP) of individual eyes by AREDS severity score, none to date has used a patient-based scoring system that uses images from both eyes to assign a severity score. DeepSeeNet, a deep learning model, was developed to classify patients automatically by the AREDS Simplified Severity Scale (score 0-5) using bilateral CFP. DeepSeeNet was trained on 58,402 and tested on 900 images from the longitudinal follow-up of 4549 participants from AREDS. Gold standard labels were obtained using reading center grades. DeepSeeNet simulates the human grading process by first detecting individual AMD risk factors (drusen size, pigmentary abnormalities) for each eye and then calculating a patient-based AMD severity score using the AREDS Simplified Severity Scale. DeepSeeNet performed better on patient-based classification (accuracy = 0.671; kappa = 0.558) than retinal specialists (accuracy = 0.599; kappa = 0.467) with high AUC in the detection of large drusen (0.94), pigmentary abnormalities (0.93), and late AMD (0.97). DeepSeeNet demonstrated high accuracy with increased transparency in the automated assignment of individual patients to AMD risk categories based on the AREDS Simplified Severity Scale. These results highlight the potential of deep learning to assist and enhance clinical decision-making in patients with AMD, such as early AMD detection and risk prediction for developing late AMD. DeepSeeNet is publicly available on https://github.com/ncbi-nlp/DeepSeeNet.

Qingyu Chen, Yifan Peng, Zhiyong Lu

Sentence embeddings have become an essential part of today's natural language processing (NLP) systems, especially together advanced deep learning methods. Although pre-trained sentence encoders are available in the general domain, none exists for biomedical texts to date. In this work, we introduce BioSentVec: the first open set of sentence embeddings trained with over 30 million documents from both scholarly articles in PubMed and clinical notes in the MIMIC-III Clinical Database. We evaluate BioSentVec embeddings in two sentence pair similarity tasks in different text genres. Our benchmarking results demonstrate that the BioSentVec embeddings can better capture sentence semantics compared to the other competitive alternatives and achieve state-of-the-art performance in both tasks. We expect BioSentVec to facilitate the research and development in biomedical text mining and to complement the existing resources in biomedical word embeddings. BioSentVec is publicly available at https://github.com/ncbi-nlp/BioSentVec

Rui Yang, Haoran Liu, Edison Marrese-Taylor et al.

Large language models (LLMs) have demonstrated impressive generative capabilities with the potential to innovate in medicine. However, the application of LLMs in real clinical settings remains challenging due to the lack of factual consistency in the generated content. In this work, we develop an augmented LLM framework, KG-Rank, which leverages a medical knowledge graph (KG) along with ranking and re-ranking techniques, to improve the factuality of long-form question answering (QA) in the medical domain. Specifically, when receiving a question, KG-Rank automatically identifies medical entities within the question and retrieves the related triples from the medical KG to gather factual information. Subsequently, KG-Rank innovatively applies multiple ranking techniques to refine the ordering of these triples, providing more relevant and precise information for LLM inference. To the best of our knowledge, KG-Rank is the first application of KG combined with ranking models in medical QA specifically for generating long answers. Evaluation on four selected medical QA datasets demonstrates that KG-Rank achieves an improvement of over 18% in ROUGE-L score. Additionally, we extend KG-Rank to open domains, including law, business, music, and history, where it realizes a 14% improvement in ROUGE-L score, indicating the effectiveness and great potential of KG-Rank.

Tienyu Chang, Zhen Chen, Renjie Liang et al.

The advent of foundation models has heralded a new era in medical artificial intelligence (AI), enabling the extraction of generalizable representations from large-scale unlabeled datasets. However, current ophthalmic AI paradigms are predominantly constrained to single-modality inference, thereby creating a dissonance with clinical practice where diagnosis relies on the synthesis of complementary imaging modalities. Furthermore, the deployment of high-performance AI in resource-limited settings is frequently impeded by the unavailability of advanced three-dimensional imaging hardware. Here, we present the Ophthalmic multimodal Masked Autoencoder (OphMAE), a multi-imaging foundation model engineered to synergize the volumetric depth of 3D Optical Coherence Tomography (OCT) with the planar context of 2D en face OCT. By implementing a novel cross-modal fusion architecture and a unique adaptive inference mechanism, OphMAE was pre-trained on a massive dataset with of 183,875 paired OCT images derived from 32,765 patients. In a rigorous benchmark encompassing 17 diverse diagnostic tasks with 48,340 paired OCT images from 8,191 patients, the model demonstrated state-of-the-art performance, achieving an Area Under the Curve (AUC) of 96.9% for Age-related Macular Degeneration (AMD) and 97.2% for Diabetic Macular Edema (DME), consistently surpassing existing single-modal and multimodal foundation models. Crucially, OphMAE exhibits robust engineering adaptability: it maintains high diagnostic accuracy, such as 93.7\% AUC for AMD, even when restricted to single-modality 2D inputs, and demonstrates exceptional data efficiency by retaining 95.7% AUC with as few as 500 labeled samples. This work establishes a scalable and adaptable framework for ophthalmic AI, ensuring robust performance across different tasks.

Fan Ma, Yuntian Liu, Xiang Lan et al.

Evidence derived from large-scale real-world data (RWD) is increasingly informing regulatory evaluation and healthcare decision-making. Administrative claims provide population-scale, longitudinal records of healthcare utilization, expenditure, and detailed coding of diagnoses, procedures, and medications, yet their potential as a substrate for healthcare foundation models remains largely unexplored. Here we present ReClaim, a generative transformer trained from scratch on 43.8 billion medical events from more than 200 million enrollees in the MarketScan claims data spanning 2008-2022. ReClaim models longitudinal trajectories across diagnoses, procedures, medications, and expenditure, and was scaled to 140 million, 700 million, and 1.7 billion parameters. Across over 1,000 disease-onset prediction tasks, ReClaim achieved a mean AUC of 75.6%, substantially outperforming disease-specific LightGBM (66.3%) and the transformer-based Delphi model (69.4%), with the largest gains for rare diseases. These advantages held across retrospective and prospective evaluations and in external validation on two independent datasets. Performance improved monotonically with scale, and post-training added 13.8 percentage points over pre-training alone. Beyond disease prediction, ReClaim captured financial outcomes and improved real-world evidence (RWE) analyses: for healthcare expenditure forecasting it increased explained variance from 0.28 to 0.37 relative to LightGBM, and in a target trial emulation it reduced systematic bias by 72% on average relative to Delphi. Together, these results establish administrative claims as a scalable substrate for healthcare foundation models and show that learned representations generalize across time periods and data sources, supporting disease surveillance, expenditure forecasting, and RWE generation.

Weihao Xuan, Rui Yang, Heli Qi et al.

Existing large language model (LLM) evaluation benchmarks primarily focus on English, while current multilingual tasks lack parallel questions that specifically assess cross-linguistic reasoning abilities. This dual limitation makes it challenging to comprehensively assess LLMs' performance in the multilingual setting. To fill this gap, we introduce MMLU-ProX, a comprehensive benchmark covering 29 languages, built on an English benchmark. Each language version consists of 11,829 identical questions, enabling direct cross-linguistic comparisons. Additionally, to meet efficient evaluation needs, we provide a lite version containing 658 questions per language. To ensure the high quality of MMLU-ProX, we employ a rigorous development process that involves multiple powerful LLMs for translation, followed by expert review to ensure accurate expression, consistent terminology, and cultural relevance. Building on this, we systematically evaluate 36 state-of-the-art LLMs, including reasoning-enhanced and multilingual-optimized LLMs. The results reveal significant disparities in the multilingual capabilities of LLMs: While they perform well in high-resource languages, their performance declines markedly in low-resource languages, with gaps of up to 24.3%. Through MMLU-ProX, we aim to advance the development of more inclusive AI systems and promote equitable access to technology across global contexts.

Qiao Jin, Zhizheng Wang, Yifan Yang et al.

Clinical calculators play a vital role in healthcare by offering accurate evidence-based predictions for various purposes such as prognosis. Nevertheless, their widespread utilization is frequently hindered by usability challenges, poor dissemination, and restricted functionality. Augmenting large language models with extensive collections of clinical calculators presents an opportunity to overcome these obstacles and improve workflow efficiency, but the scalability of the manual curation process poses a significant challenge. In response, we introduce AgentMD, a novel language agent capable of curating and applying clinical calculators across various clinical contexts. Using the published literature, AgentMD has automatically curated a collection of 2,164 diverse clinical calculators with executable functions and structured documentation, collectively named RiskCalcs. Manual evaluations show that RiskCalcs tools achieve an accuracy of over 80% on three quality metrics. At inference time, AgentMD can automatically select and apply the relevant RiskCalcs tools given any patient description. On the newly established RiskQA benchmark, AgentMD significantly outperforms chain-of-thought prompting with GPT-4 (87.7% vs. 40.9% in accuracy). Additionally, we also applied AgentMD to real-world clinical notes for analyzing both population-level and risk-level patient characteristics. In summary, our study illustrates the utility of language agents augmented with clinical calculators for healthcare analytics and patient care.

Zhiyuan Wang, Jinhao Duan, Chenxi Yuan et al.

Uncertainty estimation is crucial for the reliability of safety-critical human and artificial intelligence (AI) interaction systems, particularly in the domain of healthcare engineering. However, a robust and general uncertainty measure for free-form answers has not been well-established in open-ended medical question-answering (QA) tasks, where generative inequality introduces a large number of irrelevant words and sequences within the generated set for uncertainty quantification (UQ), which can lead to biases. This paper introduces Word-Sequence Entropy (WSE), a method that calibrates uncertainty at both the word and sequence levels, considering semantic relevance. WSE quantifies uncertainty in a way that is more closely aligned with the reliability of LLMs during uncertainty quantification (UQ). We compare WSE with six baseline methods on five free-form medical QA datasets, utilizing seven popular large language models (LLMs). Experimental results demonstrate that WSE exhibits superior performance in UQ under two standard criteria for correctness evaluation. Additionally, in terms of real-world medical QA applications, the performance of LLMs is significantly enhanced (e.g., a 6.36% improvement in model accuracy on the COVID-QA dataset) by employing responses with lower uncertainty that are identified by WSE as final answers, without any additional task-specific fine-tuning or architectural modifications.

Pagnarasmey Pit, Xingjun Ma, Mike Conway et al.

Large Language Models (LLMs) have gained significant popularity for their application in various everyday tasks such as text generation, summarization, and information retrieval. As the widespread adoption of LLMs continues to surge, it becomes increasingly crucial to ensure that these models yield responses that are politically impartial, with the aim of preventing information bubbles, upholding fairness in representation, and mitigating confirmation bias. In this paper, we propose a quantitative framework and pipeline designed to systematically investigate the political orientation of LLMs. Our investigation delves into the political alignment of LLMs across a spectrum of eight polarizing topics, spanning from abortion to LGBTQ issues. Across topics, the results indicate that LLMs exhibit a tendency to provide responses that closely align with liberal or left-leaning perspectives rather than conservative or right-leaning ones when user queries include details pertaining to occupation, race, or political affiliation. The findings presented in this study not only reaffirm earlier observations regarding the left-leaning characteristics of LLMs but also surface particular attributes, such as occupation, that are particularly susceptible to such inclinations even when directly steered towards conservatism. As a recommendation to avoid these models providing politicised responses, users should be mindful when crafting queries, and exercise caution in selecting neutral prompt language.