Hasil untuk "Diseases of the endocrine glands. Clinical endocrinology"

Kenneth Cusi, Scott Isaacs, Diana Barb et al.

Hiroya Kitsunai, Yumika Watanabe, Eisuke Nishikawa et al.

Rachel Sheskier, Parisa Verma, Alexander Kirschenbaum et al.

The presence of both an extragonadal germ cell tumor and gonadal germ cell tumor is a rare occurrence with few cases reported in the literature. We herein report a case of a young man presenting with hypophysitis due to a pituitary germinoma. After a course of high dose glucocorticoid (GC) therapy and the surgical removal of the germinoma, a testicular seminoma was discovered, an apparently second distinct primary germ cell tumor. Hypophysitis was initially attributed to lymphocytic hypophysitis due to the largely unrevealing secondary work up that included beta-human chorionic gonadotropin (b-hCG) and alpha fetoprotein (AFP), which highlights the pitfalls of relying on the tumor marker of b-hCG in both germinomas and seminomas and the important role of biopsy for definitive diagnosis of hypophysitis etiology. Furthermore, the presentation of the seminoma following a course of high dose GC indicates that immunosuppressive therapies may promote the growth of these germ cell tumors.

Jieun Lee

Dmitrii Seletkov, Sophie Starck, Ayhan Can Erdur et al.

Reliable preclinical disease risk assessment is essential to move public healthcare from reactive treatment to proactive identification and prevention. However, image-based risk prediction algorithms often consider one condition at a time and depend on hand-crafted features obtained through segmentation tools. We propose a whole-body self-supervised representation learning method for the preclinical disease risk assessment under a competing risk modeling. This approach outperforms whole-body radiomics in multiple diseases, including cardiovascular disease (CVD), type 2 diabetes (T2D), chronic obstructive pulmonary disease (COPD), and chronic kidney disease (CKD). Simulating a preclinical screening scenario and subsequently combining with cardiac MRI, it sharpens further the prediction for CVD subgroups: ischemic heart disease (IHD), hypertensive diseases (HD), and stroke. The results indicate the translational potential of whole-body representations as a standalone screening modality and as part of a multi-modal framework within clinical workflows for early personalized risk stratification. The code is available at https://github.com/yayapa/WBRLforCR/

John Wu, Adam Cross, Jimeng Sun

Rare diseases affect 1 in 10 Americans, yet standard ICD coding systems fail to capture these conditions in electronic health records (EHR), leaving crucial information buried in clinical notes. Current approaches struggle with medical abbreviations, miss implicit disease mentions, raise privacy concerns with cloud processing, and lack clinical reasoning abilities. We present Rare Disease Mining Agents (RDMA), a framework that mirrors how medical experts identify rare disease patterns in EHR. RDMA connects scattered clinical observations that together suggest specific rare conditions. By handling clinical abbreviations, recognizing implicit disease patterns, and applying contextual reasoning locally on standard hardware, RDMA reduces privacy risks while improving F1 performance by upwards of 30\% and decreasing inferences costs 10-fold. This approach helps clinicians avoid the privacy risk of using cloud services while accessing key rare disease information from EHR systems, supporting earlier diagnosis for rare disease patients. Available at https://github.com/jhnwu3/RDMA.

Jiushen Cai, Weihang Zhang, Hanruo Liu et al.

Standardization of clinical reports is crucial for improving the quality of healthcare and facilitating data integration. The lack of unified standards, including format, terminology, and style, is a great challenge in clinical fundus diagnostic reports, which increases the difficulty for large language models (LLMs) to understand the data. To address this, we construct a bilingual standard terminology, containing fundus clinical terms and commonly used descriptions in clinical diagnosis. Then, we establish two models, RetSTA-7B-Zero and RetSTA-7B. RetSTA-7B-Zero, fine-tuned on an augmented dataset simulating clinical scenarios, demonstrates powerful standardization behaviors. However, it encounters a challenge of limitation to cover a wider range of diseases. To further enhance standardization performance, we build RetSTA-7B, which integrates a substantial amount of standardized data generated by RetSTA-7B-Zero along with corresponding English data, covering diverse complex clinical scenarios and achieving report-level standardization for the first time. Experimental results demonstrate that RetSTA-7B outperforms other compared LLMs in bilingual standardization task, which validates its superior performance and generalizability. The checkpoints are available at https://github.com/AB-Story/RetSTA-7B.

Anthony J. Webster

There is a growing proportion of people with several disease conditions ("multimorbidity"), placing increasing demands on healthcare systems. One hypothesis is that clusters of diseases may arise from shared underlying disease processes (shared "pathogenesis"), whereby the presence of one disease indicates the state of disease progression to several related disease types. This article explains how this hypothesis can be tested using observational data for disease incidence. Specifically, a multistage model is used to test whether two diseases can have a "shared stage" or "step", before either disease can occur, and how the unobserved rate of this step can be determined. The approach offers a simple method for studying multiple diseases and identifying shared underlying causes of multiple conditions, and is illustrated with published data and numerical examples. The fundamental mathematical model is analysed to compare key statistical properties such as the expectation and variance with those of independent diseases. The main results do not need an understanding of the underlying mathematics and can be appreciated by a non-expert. Significance: It is widely believed that there are shared underlying pathways that can lead to several disease types (shared "pathogenesis"), and this may explain observed clusters of disease types. This article shows how this hypothesis can be tested for a pair or cluster of diseases, using observational data of disease incidence.

Xiangju Hu, Xin Fang, Minxia Wu

Abstract Aims/Introduction To estimate the prevalence, awareness, treatment, control rate, and influence factors of type 2 diabetes in Fujian province and provide the scientific basic for prevention. Materials and Methods A population‐based study with the analysis of binary logistic regression was carried out to estimate the odds ratios of the influencing factor on type 2 diabetes. Data of the Patient‐Centered Evaluative Assessment of Cardiac Events (PEACE) in southeast China were used. The study sample originated from 12 counties in Fujian province and included 135,352 permanent residents aged 35–75 years in 2021. Results The prevalence of type 2 diabetes was 18.32% (24,801/135,352). Among them, 13,921 (56.13%) were aware of their condition, 11,894 (47.96%) were receiving treatment, and 4,537 (18.29%) had achieved control of blood glucose. Multivariate logistic regression analysis showed that older age, men, low‐family income, low‐education level, urban locality, no medical insurance, and histories of myocardial infarction, stroke, dyslipidemia, hypertension, alcohol consumption, and obesity were associated with a higher prevalence of type 2 diabetes. Conclusions The prevalence of type 2 diabetes among residents aged 35–75 years in southeast China is high, whereas the status of its low awareness, treatment and control is severe, warranting a broad‐based global strategy, including greater efforts in earlier screening, and more effective and affordable treatment is essential.

Siyuan Yan, Zhen Yu, Clare Primiero et al.

Diagnosing and treating skin diseases require advanced visual skills across domains and the ability to synthesize information from multiple imaging modalities. While current deep learning models excel at specific tasks like skin cancer diagnosis from dermoscopic images, they struggle to meet the complex, multimodal requirements of clinical practice. Here, we introduce PanDerm, a multimodal dermatology foundation model pretrained through self-supervised learning on over 2 million real-world skin disease images from 11 clinical institutions across 4 imaging modalities. We evaluated PanDerm on 28 diverse benchmarks, including skin cancer screening, risk stratification, differential diagnosis of common and rare skin conditions, lesion segmentation, longitudinal monitoring, and metastasis prediction and prognosis. PanDerm achieved state-of-the-art performance across all evaluated tasks, often outperforming existing models when using only 10% of labeled data. We conducted three reader studies to assess PanDerm's potential clinical utility. PanDerm outperformed clinicians by 10.2% in early-stage melanoma detection through longitudinal analysis, improved clinicians' skin cancer diagnostic accuracy by 11% on dermoscopy images, and enhanced non-dermatologist healthcare providers' differential diagnosis by 16.5% across 128 skin conditions on clinical photographs. These results demonstrate PanDerm's potential to improve patient care across diverse clinical scenarios and serve as a model for developing multimodal foundation models in other medical specialties, potentially accelerating the integration of AI support in healthcare. The code can be found at https://github.com/SiyuanYan1/PanDerm.

Zihao Zhao, Yi Jing, Fuli Feng et al.

Medication recommendation systems have gained significant attention in healthcare as a means of providing tailored and effective drug combinations based on patients' clinical information. However, existing approaches often suffer from fairness issues, as recommendations tend to be more accurate for patients with common diseases compared to those with rare conditions. In this paper, we propose a novel model called Robust and Accurate REcommendations for Medication (RAREMed), which leverages the pretrain-finetune learning paradigm to enhance accuracy for rare diseases. RAREMed employs a transformer encoder with a unified input sequence approach to capture complex relationships among disease and procedure codes. Additionally, it introduces two self-supervised pre-training tasks, namely Sequence Matching Prediction (SMP) and Self Reconstruction (SR), to learn specialized medication needs and interrelations among clinical codes. Experimental results on two real-world datasets demonstrate that RAREMed provides accurate drug sets for both rare and common disease patients, thereby mitigating unfairness in medication recommendation systems.

Lisa Adams, Felix Busch, Tianyu Han et al.

Background: Recent advancements in large language models (LLMs) offer potential benefits in healthcare, particularly in processing extensive patient records. However, existing benchmarks do not fully assess LLMs' capability in handling real-world, lengthy clinical data. Methods: We present the LongHealth benchmark, comprising 20 detailed fictional patient cases across various diseases, with each case containing 5,090 to 6,754 words. The benchmark challenges LLMs with 400 multiple-choice questions in three categories: information extraction, negation, and sorting, challenging LLMs to extract and interpret information from large clinical documents. Results: We evaluated nine open-source LLMs with a minimum of 16,000 tokens and also included OpenAI's proprietary and cost-efficient GPT-3.5 Turbo for comparison. The highest accuracy was observed for Mixtral-8x7B-Instruct-v0.1, particularly in tasks focused on information retrieval from single and multiple patient documents. However, all models struggled significantly in tasks requiring the identification of missing information, highlighting a critical area for improvement in clinical data interpretation. Conclusion: While LLMs show considerable potential for processing long clinical documents, their current accuracy levels are insufficient for reliable clinical use, especially in scenarios requiring the identification of missing information. The LongHealth benchmark provides a more realistic assessment of LLMs in a healthcare setting and highlights the need for further model refinement for safe and effective clinical application. We make the benchmark and evaluation code publicly available.

Ruyin Li, Yingjiao Wang, Zirui Zhao et al.

Background Medullary thyroid carcinoma (MTC) is a special type of thyroid carcinoma derived from the C cell of the thyroid gland. Because of the poor prognosis of MTC, a large number of studies on MTC have been conducted in the last 10 years. To better comprehend, it is necessary to clarify and define the dominant countries, organizations, core journals, important authors, and their cumulative research contributions, as well as the cooperative relationships between them. Method English publications with article type article or review about MTC from January 2012 to December 2021 was retrieved from Web of Science core collection, and VOSviewer, CiteSpace, and Microsoft Excel were applied for bibliometric study. Result A total of 1208 articles and reviews were included in this study. The 1208 papers were written by 6364 authors from 1734 organizations in 67 countries, published in 408 journals, and cited 24118 references from 3562 journals. The number of publications was essentially flat from 2012-2021, with the largest proportion of publications coming from the U.S., followed by Italy and China. Thyroid was the most productive journal, and Journal of clinical endocrinology & metabolism was the most cited journal. University of Texas MD Anderson Cancer Center was the most productive institution and Luca Giovanella, was the most productive author. Diagnostic tools, surgical treatment, non-surgical treatment, genetics and relationship with other endocrine diseases were the main research interests in this field. Prognosis has been a cutting-edge topic since 2017. Conclusion As a thyroid cancer with poor prognosis, MTC has received continuous attention in recent years. Current MTC studies mainly focused on disease intervention, mechanism research and prognosis. The main point of this study is to provide an overview of the development process and hot spots of MTC in the last decade. These might provide ideas for further research in the MTC field.

Le Xu, Zhe Xu Cao, Xin Weng et al.

BackgroundIn view of the rapid increase in the incidence of thyroid cancer (TC) and the spread of overdiagnosis around the world, the quantitative evaluation of the effect of age, period and birth cohort on the incidence of TC, and the analysis of the role of different factors in the incidence trend can provide scientific basis and data support for the national health departments to formulate reasonable prevention and treatment policies.MethodsThe study collated the global burden disease study data of TC incidence from 1990 to 2019, and used APC model to analyze the contribution of age, period and birth cohort to the incidence trend of TC.ResultsThere was an obvious unfavorable upward trend in terms of age and cohort effect all over the world. Since 2007, the growth rate of risk slowed down and the risk in female even decreased since 2012, which mainly contributed to the developed countries. In all SDI countries, 2002 is the dividing point of risk between male and female. In 2019, The global age-standardized incidence rate (ASIR) of TC in the 5 SDI countries all showed a significant upward trend, with the largest upward trend in the middle SDI countries.ConclusionThe trend of rapid increase in the incidence of TC has begun to slow down, but the global incidence of TC has obvious gender and regional/national heterogeneity. Policy makers should tailor specific local strategies to the risk factors of each country to further reduce the burden of TC.

Alexander Y. Xu, Stephen H. Wang, Tim Q. Duong

Introduction Patients with prediabetes who contract SARS-CoV-2 infection (COVID-19) could be at higher risk of developing frank diabetes compared those who do not. This study aims to investigate the incidence of new-onset diabetes in patients with prediabetes after COVID-19 and if it differs from those not infected.Research design and methods Using electronic medical record data, 42 877 patients with COVID-19, 3102 were identified as having a history of prediabetes in the Montefiore Health System, Bronx, New York. During the same time period, 34 786 individuals without COVID-19 with history of prediabetes were identified and 9306 were propensity matched as controls. SARS-CoV-2 infection status was determined by a real-time PCR test between March 11, 2020 and August 17, 2022. The primary outcomes were new-onset in-hospital diabetes mellitus (I-DM) and new-onset persistent diabetes mellitus (P-DM) at 5 months after SARS-CoV-2 infection.Results Compared with hospitalized patients without COVID-19 with history of prediabetes, hospitalized patients with COVID-19 with history of prediabetes had a higher incidence of I-DM (21.9% vs 6.02%, p<0.001) and of P-DM 5 months postinfection (14.75% vs 7.51%, p<0.001). Non-hospitalized patients with and without COVID-19 with history of prediabetes had similar incidence of P-DM (4.15% and 4.1%, p>0.05). Critical illness (HR 4.6 (95% CI 3.5 to 6.1), p<0.005), in-hospital steroid treatment (HR 2.88 (95% CI 2.2 to 3.8), p<0.005), SARS-CoV-2 infection status (HR 1.8 (95% CI 1.4 to 2.3), p<0.005), and hemoglobin A1c (HbA1c) (HR 1.7 (95% CI 1.6 to 1.8), p<0.005) were significant predictors of I-DM. I-DM (HR 23.2 (95% CI 16.1 to 33.4), p<0.005), critical illness (HR 2.4 (95% CI 1.6 to 3.8), p<0.005), and HbA1c (HR 1.3 (95% CI 1.1 to 1.4), p<0.005) were significant predictors of P-DM at follow-up.Conclusions SARS-CoV-2 infection confers a higher risk for developing persistent diabetes 5 months post-COVID-19 in patients with prediabetes who were hospitalized for COVID-19 compared with COVID-19-negative counterparts with prediabetes. In-hospital diabetes, critical illness, and elevated HbA1c are risk factors for developing persistent diabetes. Patients with prediabetes with severe COVID-19 disease may need more diligent monitoring for developing P-DM postacute SARS-CoV-2 infection.

Georgios Boutzios, Sofia Chatzi, Andreas V. Goules et al.

BackgroundGraves’ orbitopathy (GO) is an autoimmune disorder affecting the orbital fat and muscles. A significant role of IL-6 in the pathogenesis of GO has been described and tocilizumab (TCZ), an IL-6 inhibitor targeting IL-6R has been given in some patients. The aim of our case study was to evaluate the therapeutic outcome of TCZ in non-responders to first line treatments with corticosteroids.MethodsWe conducted an observational study of patients with moderate to severe GO. Twelve patients received TCZ in intravenous infusions at a dose of 8mg/kg every 28 days for 4 months and followed up for additionally 6 weeks. The primary outcome was improvement in CAS by at least 2 points, 6 weeks after the last dose of TCZ. Secondary outcomes included CAS <3 (inactive disease) 6 weeks after TCZ last dose, reduced TSI levels, proptosis reduction by > 2mm and diplopia response.ResultsThe primary outcome, was achieved in all patients 6 weeks after treatment course. Furthermore all patients had inactive disease 6 weeks after treatment cessation. Treatment with TCZ reduced significantly median CAS by 3 units (p=0.002), TSI levels by 11.02 IU/L (p=0.006), Hertel score on the right eye by 2.3 mm (p=0.003), Hertel score on the left eye by 1.6 mm (p=0.002), while diplopia persisted in fewer patients (25%) after treatment with TCZ (not statistically significant, p=0.250). After treatment with TCZ, there was a radiological improvement in 75% of patients, while 16.7% showed no response, and in 8.3% of patients deterioration was established.ConclusionTocilizumab appears to be a safe and cost effective therapeutic option for patients with active, corticosteroid-resistant, moderate to severe Graves’ orbitopathy.

Ghizal fatima, Risala H. Allami, Maitham G. Yousif

Precision medicine, tailored to individual patients based on their genetics, environment, and lifestyle, shows promise in managing complex diseases like infections. Integrating artificial intelligence (AI) into precision medicine can revolutionize disease management. This paper introduces a novel approach using AI to advance precision medicine in infectious diseases and beyond. It integrates diverse fields, analyzing patients' profiles using genomics, proteomics, microbiomics, and clinical data. AI algorithms process vast data, providing insights for precise diagnosis, treatment, and prognosis. AI-driven predictive modeling empowers healthcare providers to make personalized and effective interventions. Collaboration among experts from different domains refines AI models and ensures ethical and robust applications. Beyond infections, this AI-driven approach can benefit other complex diseases. Precision medicine powered by AI has the potential to transform healthcare into a proactive, patient-centric model. Research is needed to address privacy, regulations, and AI integration into clinical workflows. Collaboration among researchers, healthcare institutions, and policymakers is crucial in harnessing AI-driven strategies for advancing precision medicine and improving patient outcomes.

Chaoqing Xu, Zhentao Zheng, Yiting Fu et al.

Carotid artery plaques can cause arterial vascular diseases such as stroke and myocardial infarction, posing a severe threat to human life. However, the current clinical examination mainly relies on a direct assessment by physicians of patients' clinical indicators and medical images, lacking an integrated visualization tool for analyzing the influencing factors and composition of carotid artery plaques. We have designed an intelligent carotid artery plaque visual analysis system for vascular surgery experts to comprehensively analyze the clinical physiological and imaging indicators of carotid artery diseases. The system mainly includes two functions: First, it displays the correlation between carotid artery plaque and various factors through a series of information visualization methods and integrates the analysis of patient physiological indicator data. Second, it enhances the interface guidance analysis of the inherent correlation between the components of carotid artery plaque through machine learning and displays the spatial distribution of the plaque on medical images. Additionally, we conducted two case studies on carotid artery plaques using real data obtained from a hospital, and the results indicate that our designed carotid analysis system can effectively provide clinical diagnosis and treatment guidance for vascular surgeons.

S M Atikur Rahman, Sifat Ibtisum, Ehsan Bazgir et al.

The global need for effective disease diagnosis remains substantial, given the complexities of various disease mechanisms and diverse patient symptoms. To tackle these challenges, researchers, physicians, and patients are turning to machine learning (ML), an artificial intelligence (AI) discipline, to develop solutions. By leveraging sophisticated ML and AI methods, healthcare stakeholders gain enhanced diagnostic and treatment capabilities. However, there is a scarcity of research focused on ML algorithms for enhancing the accuracy and computational efficiency. This research investigates the capacity of machine learning algorithms to improve the transmission of heart rate data in time series healthcare metrics, concentrating particularly on optimizing accuracy and efficiency. By exploring various ML algorithms used in healthcare applications, the review presents the latest trends and approaches in ML-based disease diagnosis (MLBDD). The factors under consideration include the algorithm utilized, the types of diseases targeted, the data types employed, the applications, and the evaluation metrics. This review aims to shed light on the prospects of ML in healthcare, particularly in disease diagnosis. By analyzing the current literature, the study provides insights into state-of-the-art methodologies and their performance metrics.

Colin Bendell