搜索结果：JCEM case reports

Rare diseases, including Inborn Errors of Metabolism (IEM), pose significant diagnostic challenges. Case reports serve as key but computationally underutilized resources to inform diagnosis. Clinical dense information extraction refers to organizing medical information into structured predefined categories. Large Language Models (LLMs) may enable scalable information extraction from case reports but are rarely evaluated for this task. We introduce CaseReportBench, an expert-annotated dataset for dense information extraction of case reports, focusing on IEMs. Using this dataset, we assess various models and prompting strategies, introducing novel approaches such as category-specific prompting and subheading-filtered data integration. Zero-shot chain-of-thought prompting offers little advantage over standard zero-shot prompting. Category-specific prompting improves alignment with the benchmark. The open-source model Qwen2.5-7B outperforms GPT-4o for this task. Our clinician evaluations show that LLMs can extract clinically relevant details from case reports, supporting rare disease diagnosis and management. We also highlight areas for improvement, such as LLMs' limitations in recogni

Analyzing large volumes of case law to uncover evolving legal principles, across multiple cases, on a given topic is a demanding task for legal professionals. Structured topical reports provide an effective solution by summarizing key issues, principles, and judgments, enabling comprehensive legal analysis on a particular topic. While prior works have advanced query-based individual case summarization, none have extended to automatically generating multi-case structured reports. To address this, we introduce LexGenie, an automated LLM-based pipeline designed to create structured reports using the entire body of case law on user-specified topics within the European Court of Human Rights jurisdiction. LexGenie retrieves, clusters, and organizes relevant passages by topic to generate a structured outline and cohesive content for each section. Expert evaluation confirms LexGenie's utility in producing structured reports that enhance efficient, scalable legal analysis.

Timing of clinical events is central to characterization of patient trajectories, enabling analyses such as process tracing, forecasting, and causal reasoning. However, structured electronic health records capture few data elements critical to these tasks, while clinical reports lack temporal localization of events in structured form. We present a system that transforms case reports into textual time series-structured pairs of textual events and timestamps. We contrast manual and large language model (LLM) annotations (n=320 and n=390 respectively) of ten randomly-sampled PubMed open-access (PMOA) case reports (N=152,974) and assess inter-LLM agreement (n=3,103; N=93). We find that the LLM models have moderate event recall(O1-preview: 0.80) but high temporal concordance among identified events (O1-preview: 0.95). By establishing the task, annotation, and assessment systems, and by demonstrating high concordance, this work may serve as a benchmark for leveraging the PMOA corpus for temporal analytics.

This article explores the potential of generative AI (GenAI) to support actuarial practice through four implemented case studies. It situates these case studies within the broader evolution of artificial intelligence in actuarial science, from early neural networks and machine learning to modern transformer-based GenAI systems. The first case study illustrates how large language models (LLMs) can improve claim cost prediction by extracting informative features from unstructured text for use in the underlying supervised learning task. The second case study demonstrates the automation of market comparisons using Retrieval-Augmented Generation to identify, extract, and structure relevant information from insurers' annual reports. The third case study highlights the capabilities of fine-tuned vision-enabled LLMs in classifying car damage types and extracting contextual information from images. The fourth case study presents a multi-agent system that autonomously migrates actuarial legacy code from R to Python and validates the translation against the original code's outputs. In addition to these case studies, we outline further GenAI applications in the insurance industry. Finally, we

Background: A large number of neurology case reports have been published, but it is a challenging task for human medical experts to explore all of these publications. Text mining offers a computational approach to investigate neurology literature and capture meaningful patterns. The overarching goal of this study is to provide a new perspective on case reports of neurological disease and syndrome analysis over the last six decades using text mining. Methods: We extracted diseases and syndromes (DsSs) from more than 65,000 neurology case reports from 66 journals in PubMed over the last six decades from 1955 to 2017. Text mining was applied to reports on the detected DsSs to investigate high-frequency DsSs, categorize them, and explore the linear trends over the 63-year time frame. Results: The text mining methods explored high-frequency neurologic DsSs and their trends and the relationships between them from 1955 to 2017. We detected more than 18,000 unique DsSs and found 10 categories of neurologic DsSs. While the trend analysis showed the increasing trends in the case reports for top-10 high-frequency DsSs, the categories had mixed trends. Conclusion: Our study provided new insigh

Maintaining reliable UI test suites in large-scale enterprise applications is a persistent and costly challenge. We present an industrial case study of a multi-agent autonomous testing system evaluated using anonymized execution data from a production-like enterprise UI testing prototype. The application features several hundred dynamic UI elements per screen. Built on a large language model with LangGraph orchestration, Playwright execution, and a RAG knowledge base, the system evolves from human-directed testing toward High-autonomy feature discovery and test execution: given no explicit test targets, it discovers over 100 testable features across 10 UI screens, dynamically expands coverage by an additional 15--30 features through runtime DOM analysis, and iteratively repairs failing tests without human intervention. We analyzed 300 consecutive autonomous execution reports encompassing 636 individual test-case executions across 10 distinct scenario families. The system achieved a 70% repair convergence rate at the scenario-family level, with a mean of 3.4 repair iterations to convergence. However, only 10% of scenario families succeeded on first attempt, 38% of reports failed to

We present a new corpus comprising annotations of medical entities in case reports, originating from PubMed Central's open access library. In the case reports, we annotate cases, conditions, findings, factors and negation modifiers. Moreover, where applicable, we annotate relations between these entities. As such, this is the first corpus of this kind made available to the scientific community in English. It enables the initial investigation of automatic information extraction from case reports through tasks like Named Entity Recognition, Relation Extraction and (sentence/paragraph) relevance detection. Additionally, we present four strong baseline systems for the detection of medical entities made available through the annotated dataset.

Purpose: We investigated the utilization of privacy-preserving, locally-deployed, open-source Large Language Models (LLMs) to extract diagnostic information from free-text cardiovascular magnetic resonance (CMR) reports. Materials and Methods: We evaluated nine open-source LLMs on their ability to identify diagnoses and classify patients into various cardiac diagnostic categories based on descriptive findings in 109 clinical CMR reports. Performance was quantified using standard classification metrics including accuracy, precision, recall, and F1 score. We also employed confusion matrices to examine patterns of misclassification across models. Results: Most open-source LLMs demonstrated exceptional performance in classifying reports into different diagnostic categories. Google's Gemma2 model achieved the highest average F1 score of 0.98, followed by Qwen2.5:32B and DeepseekR1-32B with F1 scores of 0.96 and 0.95, respectively. All other evaluated models attained average scores above 0.93, with Mistral and DeepseekR1-7B being the only exceptions. The top four LLMs outperformed our board-certified cardiologist (F1 score of 0.94) across all evaluation metrics in analyzing CMR reports.

We report results of the CASE 2022 Shared Task 1 on Multilingual Protest Event Detection. This task is a continuation of CASE 2021 that consists of four subtasks that are i) document classification, ii) sentence classification, iii) event sentence coreference identification, and iv) event extraction. The CASE 2022 extension consists of expanding the test data with more data in previously available languages, namely, English, Hindi, Portuguese, and Spanish, and adding new test data in Mandarin, Turkish, and Urdu for Sub-task 1, document classification. The training data from CASE 2021 in English, Portuguese and Spanish were utilized. Therefore, predicting document labels in Hindi, Mandarin, Turkish, and Urdu occurs in a zero-shot setting. The CASE 2022 workshop accepts reports on systems developed for predicting test data of CASE 2021 as well. We observe that the best systems submitted by CASE 2022 participants achieve between 79.71 and 84.06 F1-macro for new languages in a zero-shot setting. The winning approaches are mainly ensembling models and merging data in multiple languages. The best two submissions on CASE 2021 data outperform submissions from last year for Subtask 1 and Su

With the growth of global maritime transportation, energy optimization has become crucial for reducing costs and ensuring operational efficiency. Shaft power is the mechanical power transmitted from the engine to the shaft and directly impacts fuel consumption, making its accurate prediction a paramount step in optimizing vessel performance. Power consumption is highly correlated with ship parameters such as speed and shaft rotation per minute, as well as weather and sea conditions. Frequent access to this operational data can improve prediction accuracy. However, obtaining high-quality sensor data is often infeasible and costly, making alternative sources such as noon reports a viable option. In this paper, we propose a transfer learning-based approach for predicting vessels shaft power, where a model is initially trained on high-frequency data from a vessel and then fine-tuned with low-frequency daily noon reports from other vessels. We tested our approach on sister vessels (identical dimensions and configurations), a similar vessel (slightly larger with a different engine), and a different vessel (distinct dimensions and configurations). The experiments showed that the mean abso

Timely identification of issue reports reflecting software vulnerabilities is crucial, particularly for Internet-of-Things (IoT) where analysis is slower than non-IoT systems. While Machine Learning (ML) and Large Language Models (LLMs) detect vulnerability-indicating issues in non-IoT systems, their IoT use remains unexplored. We are the first to tackle this problem by proposing two approaches: (1) combining ML and LLMs with Natural Language Processing (NLP) techniques to detect vulnerability-indicating issues of 21 Eclipse IoT projects and (2) fine-tuning a pre-trained BERT Masked Language Model (MLM) on 11,000 GitHub issues for classifying \vul. Our best performance belongs to a Support Vector Machine (SVM) trained on BERT NLP features, achieving an Area Under the receiver operator characteristic Curve (AUC) of 0.65. The fine-tuned BERT achieves 0.26 accuracy, emphasizing the importance of exposing all data during training. Our contributions set the stage for accurately detecting IoT vulnerabilities from issue reports, similar to non-IoT systems.

The demographic change in submissions to JCEM over the last decade is very apparent and largely driven by the emergence of China as a scientific superpower. Propelled by increases to budgets for research and development and national infrastructure platforms, China has overtaken the United States and Europe in terms of the size of its research workforce and its published scientific outputs. The Nature Index 2024 Research Leaders lists the leading institutions and countries/territories in the natural and health sciences according to their output in Nature Index journals (1) and shows that in 2023 China surpassed the United States for the first time in biomedical and health disciplines. Perhaps not surprisingly, as the most-cited journal in our discipline and one that is very popular in China, JCEM has statistics that mirror those of Nature journals. Ten years ago, manuscripts from China accounted for 11% of all submissions; by 2019 this had increased to 21%. The data for 2024 show that this figure has now increased dramatically, to more than 50% of submissions. JCEM is totally committed to attracting and publishing the best research in our discipline from around the world. We embrace geographical diversity in our governance (we have editors from 12 countries and editorial board members from 25 countries, with China well represented). We publish outputs from submitting authors based in 42 countries (2023 data). So what is the problem? In a nutshell, quality. Acceptance rates for submissions from China through 2014-2019 were ∼10%, but despite the rapid increase in submissions, the numbers of accepted papers have only modestly increased, such that their acceptance rate is now 6%. We are processing thousands of manuscripts per year from China, the vast majority of them either out of scope or lacking a mechanistic or experimental basis. We commonly see submissions based on single association studies, drawn from analysis of large published datasets such as the UK Biobank or the National Health and Nutrition Examination Survey, exploring relationships between unexplained and biologically irrelevant variables. Consequently the “reject without review rate” is unacceptably high, bringing with it significant staff time implications as well as diverting editors to “fire-fighting” rather than focusing on content strategy and enhancement of the journal for readers. These problems aside, I want to reassure our community that we are maintaining the quality of our peer review process and what we publish; preliminary data indicate that citation rates for manuscripts published from China are comparable to those published from the United States and western Europe. A pressing global concern is the authenticity of what is submitted. The pressure on emerging researchers to publish, particularly in clinical medicine, is intense in some countries, including China. This has contributed to the growth of “paper mills”—commercial organizations that charge for researchers to be listed as authors on wholly or partly fabricated, ghost-written papers. It is a huge problem and one that is not specific to endocrinology and metabolism. Studies show that some 2% of all scientific papers published in 2022 resembled paper-mill productions (2, 3). China, Thailand, and Saudi Arabia have all come under the spotlight (4), but fraudulent commercial activities are not confined to any single country and solutions must be global. In January 2025 the Chinese government announced a crackdown on paper mills and we await with interest to see the impact that this may have. Identity theft and misuse of identification data are further issues that are eroding one of the bedrock principles of publishing—trust in an individual's identity as a researcher and in their claimed affiliation and publication record. I cannot say with confidence that JCEM has published no fraudulent papers or never been a victim of the paper-mill industry. There are well-validated pointers that raise alarm bells in screening for such malpractice, and we have rejected several manuscripts without review in recent years on the basis of such suspicions. Further stringency will follow as we work with our publishing partner, Oxford University Press, on software screening tools. I do want to reassure all authors and readers of JCEM that we are committed to action. We have tightened our guidance on what research is within scope for JCEM and added the requirement for source documentation on ethics approvals. We have given additional guidance on Mendelian randomization (MR) studies, requiring that authors adhere to the STROBE-MR (STrengthening the Reporting of OBservational studies in Epidemiology) guidelines. We also encourage authors to follow STROBE guidelines for observational studies generally (5). A key underpinning issue that has been identified is the prolific use of noninstitutional email accounts, which allows false self-identification and facilitates identity manipulation by bad actors (6). Our experience from returning to authors literally hundreds of submitted manuscripts that list no author with an institutional email account is that few of them are subsequently returned. More than a year ago, we instigated the requirement that submissions should list at least 1 author with an institutional email account, and we will be tightening this requirement further during 2025. We accept that this will be unpopular in some countries where many junior researchers and PhD students do not currently have access to institutional accounts. This is especially the case for emerging clinical researchers employed by hospitals rather than universities. Of course, we will be mindful of exceptions, but with the understanding that change must be a two-way process. Academic centers of excellence across the world with hospital/healthcare organization partnerships must take greater ownership and accountability for what is submitted to JCEM. From February, either an institutional email address or signoff by all authors with credible identity validation will be required for all authors lacking such validation, so they collectively confirm the identity of those 1 or 2 authors of a paper who unavoidably lack an institutional email address. As always, the submitting author must attest to the authenticity of the research and should ensure that it falls within the scope of JCEM. In our societal move to a digitally driven, “user-friendly” economy, electronic submission systems, researchers, and institutions will have to adapt. The current publishing environment has far too many gaps that have allowed this unprecedented surge of poor-quality manuscripts, some of which are undoubtedly fraudulent. Because no significant original research in endocrinology can be produced without the support and infrastructure of an institution or organization, it is time for all to recognize that employers must take responsibility for the work of their employees—by providing them with the means to identify themselves as legitimate researchers employed by their institution or organization. In short, if the employer is not willing to trust a researcher with an identifier linking them to the organization, why should we at JCEM trust their research? Raising the bar on research quality, reproducibility, and accountability is a strong commitment that goes hand in hand with JCEM being a global platform for knowledge transfer and innovation. The quote “Houston we have a problem” was popularized by the 1995 film Apollo 13. Those of you with an ear for detail will know that this was actually a misquote. The phrase used in the 1970 real life episode by command module pilot John Swigert was “Houston we’ve had a problem here.” Hopefully, concerted action here from all of us will enable our immediate problem to become an issue of the past. The author wishes to acknowledge the support of Endocrine Society Publications Department staff for providing invaluable input to this editorial. The author is Editor-in-Chief of The Journal of Clinical Endocrinology & Metabolism.

Screening mammography is high volume, time sensitive, and documentation heavy. Radiologists must translate subtle visual findings into consistent BI-RADS assessments, breast density categories, and structured narrative reports. While recent Vision Language Models (VLMs) enable image-to-text reporting, many rely on closed cloud systems or tightly coupled architectures that limit privacy, reproducibility, and adaptability. We present MammoWise, a local multi-model pipeline that transforms open source VLMs into mammogram report generators and multi-task classifiers. MammoWise supports any Ollama-hosted VLM and mammography dataset, and enables zero-shot, few-shot, and Chain-of-Thought prompting, with optional multimodal Retrieval Augmented Generation (RAG) using a vector database for case-specific context. We evaluate MedGemma, LLaVA-Med, and Qwen2.5-VL on VinDr-Mammo and DMID datasets, assessing report quality (BERTScore, ROUGE-L), BI-RADS classification, breast density, and key findings. Report generation is consistently strong and improves with few-shot prompting and RAG. Classification is feasible but sensitive to model and dataset choice. Parameter-efficient fine-tuning (QLoRA) of

In 1984 Edward Witten proposed that an extremely dense form of matter composed of up, down, and strange quarks may be stable at zero pressure (Witten, 1984). Massive nuggets of such dense matter, if they exist, may pass through the Earth and be detectable by the seismic signals they generate (de Rujula and Glashow, 1984). With this motivation we investigated over 1 million seismic data reports to the U.S. Geological Survey for the years 1990-1993 not associated with epicentral sources. We report two results: (1) with an average of about 0.16 unassociated reports per minute after data cuts, we found a significant excess over statistical expectation for sets with ten or more reports in ten minutes; and (2) in spite of a very small a priori probability from random reports, we found one set of reports with arrival times and other features appropriate to signals from an epilinear source. This event has the properties predicted for the passage of a nugget of strange quark matter (SQM) through the earth, although there is no direct confirmation from other phenomenologies.

Infectious disease forecasting is of great interest to the public health community and policymakers, since forecasts can provide insight into disease dynamics in the near future and inform interventions. Due to delays in case reporting, however, forecasting models may often underestimate the current and future disease burden. In this paper, we propose a general framework for addressing reporting delay in disease forecasting efforts with the goal of improving forecasts. We propose strategies for leveraging either historical data on case reporting or external internet-based data to estimate the amount of reporting error. We then describe several approaches for adapting general forecasting pipelines to account for under- or over-reporting of cases. We apply these methods to address reporting delay in data on dengue fever cases in Puerto Rico from 1990 to 2009 and to reports of influenza-like illness (ILI) in the United States between 2010 and 2019. Through a simulation study, we compare method performance and evaluate robustness to assumption violations. Our results show that forecasting accuracy and prediction coverage almost always increase when correction methods are implemented to

The recently introduced Genetic Column Generation (GenCol) algorithm has been numerically observed to efficiently and accurately compute high-dimensional optimal transport plans for general multi-marginal problems, but theoretical results on the algorithm have hitherto been lacking. The algorithm solves the OT linear program on a dynamically updated low-dimensional submanifold consisting of sparse plans. The submanifold dimension exceeds the sparse support of optimal plans only by a fixed factor $β$. Here we prove that for $β\geq 2$ and in the two-marginal case, GenCol always converges to an exact solution, for arbitrary costs and marginals. The proof relies on the concept of c-cyclical monotonicity. As an offshoot, GenCol rigorously reduces the data complexity of numerically solving two-marginal OT problems from $O(\ell^2)$ to $O(\ell)$ without any loss in accuracy, where $\ell$ is the number of discretization points for a single marginal. At the end of the paper we also present some insights into the convergence behavior in the multi-marginal case.