Hasil untuk "Public aspects of medicine"

Teerachate Nantakeeratipat

BackgroundLarge Language Models (LLMs) in healthcare holds immense promise yet carries the risk of perpetuating social biases. While artificial intelligence (AI) fairness is a growing concern, a gap exists in understanding how these models perform under conditions of clinical ambiguity, a common feature in real-world practice.MethodsWe conducted a study using an ambiguity-probe methodology with a set of 42 sociodemographic personas and 15 clinical vignettes based on the 2018 classification of periodontal diseases. Ten were clear-cut scenarios with established ground truths, while five were intentionally ambiguous. OpenAI's GPT-4o and Google's Gemini 2.5 Pro were prompted to provide periodontal stage and grade assessments using 630 vignette-persona combinations per model.ResultsIn clear-cut scenarios, GPT-4o demonstrated significantly higher combined (stage and grade) accuracy (70.5%) than Gemini Pro (33.3%). However, a robust fairness analysis using cumulative link models with false discovery rate correction revealed no statistically significant sociodemographic bias in either model. This finding held true across both clear-cut and ambiguous clinical scenarios.ConclusionTo our knowledge, this is among the first study to use simulated clinical ambiguity to reveal the distinct ethical fingerprints of LLMs in a dental context. While LLM performance gaps exist, our analysis decouples accuracy from fairness, demonstrating that both models maintain sociodemographic neutrality. We identify that the observed errors are not bias, but rather diagnostic boundary instability. This highlights a critical need for future research to differentiate between these two distinct types of model failure to build genuinely reliable AI.

YunHe Su, Zhengyang Lu, Junhui Liu et al.

This paper explores the advancements and applications of large-scale models in the medical field, with a particular focus on Medical Large Models (MedLMs). These models, encompassing Large Language Models (LLMs), Vision Models, 3D Large Models, and Multimodal Models, are revolutionizing healthcare by enhancing disease prediction, diagnostic assistance, personalized treatment planning, and drug discovery. The integration of graph neural networks in medical knowledge graphs and drug discovery highlights the potential of Large Graph Models (LGMs) in understanding complex biomedical relationships. The study also emphasizes the transformative role of Vision-Language Models (VLMs) and 3D Large Models in medical image analysis, anatomical modeling, and prosthetic design. Despite the challenges, these technologies are setting new benchmarks in medical innovation, improving diagnostic accuracy, and paving the way for personalized healthcare solutions. This paper aims to provide a comprehensive overview of the current state and future directions of large models in medicine, underscoring their significance in advancing global health.

Faraz Farhadi, Shadi A. Esfahani, Fereshteh Yousefirizi et al.

The field of Clinical-Computational Nuclear Medicine is rapidly advancing, fueled by AI, tracer kinetic modeling, radiomics, and integrated informatics. These technologies improve imaging quality, automate lesion detection, and enable personalized radiopharmaceutical therapy through physiologically based pharmacokinetic (PBPK) modeling and voxel-level dosimetry. Workflow automation and Natural Language Processing (NLP) further enhance operational efficiency. However, successful implementation and adoption of these tools require clinical oversight to ensure accuracy, interpretability, and patient safety. This paper highlights key computational innovations and emphasizes the critical role of clinician-guided evaluation in shaping the future of precision imaging and therapy.

Tyler J Poore, Christopher J Pinard, Aleena Shabbir et al.

Large language models (LLMs) are increasingly used in clinical settings, yet their performance in veterinary medicine remains underexplored. We evaluated three commercially available veterinary-focused LLM summarization tools (Product 1 [Hachiko] and Products 2 and 3) on a standardized dataset of veterinary oncology records. Using a rubric-guided LLM-as-a-judge framework, summaries were scored across five domains: Factual Accuracy, Completeness, Chronological Order, Clinical Relevance, and Organization. Product 1 achieved the highest overall performance, with a median average score of 4.61 (IQR: 0.73), compared to 2.55 (IQR: 0.78) for Product 2 and 2.45 (IQR: 0.92) for Product 3. It also received perfect median scores in Factual Accuracy and Chronological Order. To assess the internal consistency of the grading framework itself, we repeated the evaluation across three independent runs. The LLM grader demonstrated high reproducibility, with Average Score standard deviations of 0.015 (Product 1), 0.088 (Product 2), and 0.034 (Product 3). These findings highlight the importance of veterinary-specific commercial LLM tools and demonstrate that LLM-as-a-judge evaluation is a scalable and reproducible method for assessing clinical NLP summarization in veterinary medicine.

Alex Hall, Vanessa Davey, Lisa McGarrigle et al.

Abstract Background How quality of care in homecare for older people is understood is important, because it influences how quality in homecare is delivered, improved, regulated and measured. We conducted a scoping review to summarise the meanings of homecare quality for key stakeholders and identify measures of homecare quality. Methods We searched four databases (CINAHL, PsycINFO, ASSIA and Social Care Online) for peer-reviewed literature from high-income countries, and websites of major UK organisations for grey literature, published between 2016 and 2023. Articles were included if they reported views of any stakeholder on quality of care in the context of homecare for older people (aged 65 and above). Data were summarised using a qualitative content analysis approach to identify key dimensions of quality. Results Overall, 93 articles from 16 countries were included. Research focussed on understanding the views of four groups: older people, family carers, care workers and service providers. Homecare quality is understood as a multi-dimensional concept. We identified four dimensions: (1) relationships and continuity of care; (2) bespoke care; (3) organisational and structural aspects of care; and (4) understanding of quality as a measurable construct. Notable gaps in the literature include a lack of evidence on how older people form and articulate their preferences for homecare; a lack of consensus on care quality measurement; and a lack of focus on optimal models of care provision within existing budgets. Many crucial perspectives were absent, including owners of homecare organisations, inspectors and assessors involved in regulation of homecare services, and the legal or advocacy professions. Conclusions There is a wealth of evidence about how homecare quality for older people is understood. These understandings are largely consistent across different constituencies and countries. It is less clear how this shared vision of high quality homecare might be realised within existing systems.

Jingqi Zeng, Xiaobin Jia

Traditional Chinese Medicine (TCM) involves complex compatibility mechanisms characterized by multi-component and multi-target interactions, which are challenging to quantify. To address this challenge, we applied graph artificial intelligence to develop a TCM multi-dimensional knowledge graph that bridges traditional TCM theory and modern biomedical science (https://zenodo.org/records/13763953 ). Using feature engineering and embedding, we processed key TCM terminology and Chinese herbal pieces (CHP), introducing medicinal properties as virtual nodes and employing graph neural networks with attention mechanisms to model and analyze 6,080 Chinese herbal formulas (CHF). Our method quantitatively assessed the roles of CHP within CHF and was validated using 215 CHF designed for COVID-19 management. With interpretable models, open-source data, and code (https://github.com/ZENGJingqi/GraphAI-for-TCM ), this study provides robust tools for advancing TCM theory and drug discovery.

Gilad Abiri

We are increasingly subjected to the power of AI authorities. As AI decisions become inescapable, entering domains such as healthcare, education, and law, we must confront a vital question: how can we ensure AI systems have the legitimacy necessary for effective governance? This essay argues that to secure AI legitimacy, we need methods that engage the public in designing and constraining AI systems, ensuring these technologies reflect the community's shared values. Constitutional AI, proposed by Anthropic, represents a step towards this goal, offering a model for democratic control of AI. However, while Constitutional AI's commitment to hardcoding explicit principles into AI models enhances transparency and accountability, it falls short in two crucial aspects: addressing the opacity of individual AI decisions and fostering genuine democratic legitimacy. To overcome these limitations, this essay proposes "Public Constitutional AI." This approach envisions a participatory process where diverse stakeholders, including ordinary citizens, deliberate on the principles guiding AI development. The resulting "AI Constitution" would carry the legitimacy of popular authorship, grounding AI governance in the public will. Furthermore, the essay proposes "AI Courts" to develop "AI case law," providing concrete examples for operationalizing constitutional principles in AI training. This evolving combination of constitutional principles and case law aims to make AI governance more responsive to public values. By grounding AI governance in deliberative democratic processes, Public Constitutional AI offers a path to imbue automated authorities with genuine democratic legitimacy, addressing the unique challenges posed by increasingly powerful AI systems while ensuring their alignment with the public interest.

Sunan He, Yuxiang Nie, Hongmei Wang et al.

Generalist foundation models (GFMs) are renowned for their exceptional capability and flexibility in effectively generalizing across diverse tasks and modalities. In the field of medicine, while GFMs exhibit superior generalizability based on their extensive intrinsic knowledge as well as proficiency in instruction following and in-context learning, specialist models excel in precision due to their domain knowledge. In this work, for the first time, we explore the synergy between the GFM and specialist models, to enable precise medical image analysis on a broader scope. Specifically, we propose a cooperative framework, Generalist-Specialist Collaboration (GSCo), which consists of two stages, namely the construction of GFM and specialists, and collaborative inference on downstream tasks. In the construction stage, we develop MedDr, the largest open-source GFM tailored for medicine, showcasing exceptional instruction-following and in-context learning capabilities. Meanwhile, a series of lightweight specialists are crafted for downstream tasks with low computational cost. In the collaborative inference stage, we introduce two cooperative mechanisms, Mixture-of-Expert Diagnosis and Retrieval-Augmented Diagnosis, to harvest the generalist's in-context learning abilities alongside the specialists' domain expertise. For a comprehensive evaluation, we curate a large-scale benchmark featuring 28 datasets and about 250,000 images. Extensive results demonstrate that MedDr consistently outperforms state-of-the-art GFMs on downstream datasets. Furthermore, GSCo exceeds both GFMs and specialists across all out-of-domain disease diagnosis datasets. These findings indicate a significant paradigm shift in the application of GFMs, transitioning from separate models for specific tasks to a collaborative approach between GFMs and specialists, thereby advancing the frontiers of generalizable AI in medicine.

Ehsan Abadi, Aldo Badano, Predrag Bakic et al.

This submission comprises the proceedings of the 1st Virtual Imaging Trials in Medicine conference, organized by Duke University on April 22-24, 2024. The listed authors serve as the program directors for this conference. The VITM conference is a pioneering summit uniting experts from academia, industry and government in the fields of medical imaging and therapy to explore the transformative potential of in silico virtual trials and digital twins in revolutionizing healthcare. The proceedings are categorized by the respective days of the conference: Monday presentations, Tuesday presentations, Wednesday presentations, followed by the abstracts for the posters presented on Monday and Tuesday.

Keping Bi, Xiaojie Sun, Jiafeng Guo et al.

Multi-aspect dense retrieval aims to incorporate aspect information (e.g., brand and category) into dual encoders to facilitate relevance matching. As an early and representative multi-aspect dense retriever, MADRAL learns several extra aspect embeddings and fuses the explicit aspects with an implicit aspect "OTHER" for final representation. MADRAL was evaluated on proprietary data and its code was not released, making it challenging to validate its effectiveness on other datasets. We failed to reproduce its effectiveness on the public MA-Amazon data, motivating us to probe the reasons and re-examine its components. We propose several component alternatives for comparisons, including replacing "OTHER" with "CLS" and representing aspects with the first several content tokens. Through extensive experiments, we confirm that learning "OTHER" from scratch in aspect fusion is harmful. In contrast, our proposed variants can greatly enhance the retrieval performance. Our research not only sheds light on the limitations of MADRAL but also provides valuable insights for future studies on more powerful multi-aspect dense retrieval models. Code will be released at: https://github.com/sunxiaojie99/Reproducibility-for-MADRAL.

Abbi Abdel-Rehim, Oghenejokpeme Orhobor, Gareth Griffiths et al.

The concept of personalised medicine in cancer therapy is becoming increasingly important. There already exist drugs administered specifically for patients with tumours presenting well-defined mutations. However, the field is still in its infancy, and personalised treatments are far from being standard of care. Personalised medicine is often associated with the utilisation of omics data. Yet, implementation of multi-omics data has proven difficult, due to the variety and scale of the information within the data, as well as the complexity behind the myriad of interactions taking place within the cell. An alternative approach to precision medicine is to employ a function-based profile of the cell. This involves screening a range of drugs against patient derived cells. Here we demonstrate a proof-of-concept, where a collection of drug screens against a highly diverse set of patient-derived cell lines, are leveraged to identify putative treatment options for a 'new patient'. We show that this methodology is highly efficient in ranking the drugs according to their activity towards the target cells. We argue that this approach offers great potential, as activities can be efficiently imputed from various subsets of the drug treated cell lines that do not necessarily originate from the same tissue type.

Hua Pu, Jiacong Mi, Shan Lu et al.

Traditional Chinese medicine (TCM) prescription is the most critical form of TCM treatment, and uncovering the complex nonlinear relationship between symptoms and TCM is of great significance for clinical practice and assisting physicians in diagnosis and treatment. Although there have been some studies on TCM prescription generation, these studies consider a single factor and directly model the symptom-prescription generation problem mainly based on symptom descriptions, lacking guidance from TCM knowledge. To this end, we propose a RoBERTa and Knowledge Enhancement model for Prescription Generation of Traditional Chinese Medicine (RoKEPG). RoKEPG is firstly pre-trained by our constructed TCM corpus, followed by fine-tuning the pre-trained model, and the model is guided to generate TCM prescriptions by introducing four classes of knowledge of TCM through the attention mask matrix. Experimental results on the publicly available TCM prescription dataset show that RoKEPG improves the F1 metric by about 2% over the baseline model with the best results.

Joshua Durso-Finley, Jean-Pierre Falet, Raghav Mehta et al.

Image-based precision medicine aims to personalize treatment decisions based on an individual's unique imaging features so as to improve their clinical outcome. Machine learning frameworks that integrate uncertainty estimation as part of their treatment recommendations would be safer and more reliable. However, little work has been done in adapting uncertainty estimation techniques and validation metrics for precision medicine. In this paper, we use Bayesian deep learning for estimating the posterior distribution over factual and counterfactual outcomes on several treatments. This allows for estimating the uncertainty for each treatment option and for the individual treatment effects (ITE) between any two treatments. We train and evaluate this model to predict future new and enlarging T2 lesion counts on a large, multi-center dataset of MR brain images of patients with multiple sclerosis, exposed to several treatments during randomized controlled trials. We evaluate the correlation of the uncertainty estimate with the factual error, and, given the lack of ground truth counterfactual outcomes, demonstrate how uncertainty for the ITE prediction relates to bounds on the ITE error. Lastly, we demonstrate how knowledge of uncertainty could modify clinical decision-making to improve individual patient and clinical trial outcomes.

Li X, Jiang H, Liang X

Xuefeng Li,1,2 Hui Jiang,1 Xiaoyu Liang1 1School of Engineering Science, University of Chinese Academy of Sciences, Beijing, 100049, People’s Republic of China; 2School of Public Policy and Management, University of Chinese Academy of Sciences, Beijing, 100049, People’s Republic of ChinaCorrespondence: Hui Jiang, School of Engineering Science, University of Chinese Academy of Sciences, 19 Yuquan Road, Shijingshan District, Beijing, People’s Republic of China, Email huijiang@ucas.ac.cnPurpose: Based on the Chinese context, this study uses severe acute respiratory syndrome (SARS) and coronavirus disease 2019 (COVID-19) outbreaks as examples to identify the risk factors that lead to the major emerging infectious diseases outbreak, and put forward risk governance strategies to improve China’s biosecurity risk prevention and control capabilities.Material and Methods: This study combines grounded theory and WSR methodology, and utilizes the NVivo 12.0 qualitative analysis software to identify the risk factors that led to the major emerging infectious diseases outbreak. The research data was sourced from 168 publicly available official documents, which are highly authoritative and reliable.Results: This study identified 10 categories of Wuli risk factors, 6 categories of logical Shili risk factors, and 8 categories of human Renli risk factors that contributed to the outbreak of major emerging infectious diseases. These risk factors were distributed across the early stages of the outbreak, and have different mechanisms of action at the macro and micro levels.Conclusion: This study identified the risk factors that lead to the outbreak of major emerging infectious disease, and discovered the mechanism of the outbreak at the macro and micro levels. At the macro level, Wuli risk factors are the forefront antecedents that lead to the outbreak of the crisis, Renli factors are the intermediate regulatory factors, and Shili risk factors are the back-end posterior factors. At the micro level, there are risk coupling, risk superposition, and risk resonance interactions among various risk factors, leading to the outbreak of the crisis. Based on these interactive relationships, this study proposes risk governance strategies that are helpful for policymakers in dealing with similar crises in the future.Keywords: major emerging infectious diseases, risk factors, SARS, COVID-19, grounded theory, WSR methodology

Rachel E. Bosserman, Christopher W. Farnsworth, Caroline A. O’Neil et al.

Abstract Objective: To determine the prevalence of severe acute respiratory coronavirus virus 2 (SARS-CoV-2) IgG nucleocapsid (N) antibodies among healthcare personnel (HCP) with no prior history of COVID-19 and to identify factors associated with seropositivity. Design: Prospective cohort study. Setting: An academic, tertiary-care hospital in St. Louis, Missouri. Participants: The study included 400 HCP aged ≥18 years who potentially worked with coronavirus disease 2019 (COVID-19) patients and had no known history of COVID-19; 309 of these HCP also completed a follow-up visit 70–160 days after enrollment. Enrollment visits took place between September and December 2020. Follow-up visits took place between December 2020 and April 2021. Methods: At each study visit, participants underwent SARS-CoV-2 IgG N-antibody testing using the Abbott SARS-CoV-2 IgG assay and completed a survey providing information about demographics, job characteristics, comorbidities, symptoms, and potential SARS-CoV-2 exposures. Results: Participants were predominately women (64%) and white (79%), with median age of 34.5 years (interquartile range [IQR], 30–45). Among the 400 HCP, 18 (4.5%) were seropositive for IgG N-antibodies at enrollment. Also, 34 (11.0%) of 309 were seropositive at follow-up. HCP who reported having a household contact with COVID-19 had greater likelihood of seropositivity at both enrollment and at follow-up. Conclusions: In this cohort of HCP during the first wave of the COVID-19 pandemic, ∼1 in 20 had serological evidence of prior, undocumented SARS-CoV-2 infection at enrollment. Having a household contact with COVID-19 was associated with seropositivity.

Guozhen Yuan, Jingjing Shi, Qiulei Jia et al.

Cardiovascular disease (CVD) is a serious threat to global public health due to its high prevalence and disability rate. Meanwhile, cardiac rehabilitation (CR) has attracted increasing attention for its positive effects on the cardiovascular system. There is overwhelming evidence that CR for patients with CVD is effective in reducing cardiovascular morbidity and mortality. To learn more about the development of CR, 5,567 papers about CR and related research were retrieved in the Web of Science Core Collection from 2001 to 2020. Then, these publications were scientometrically analyzed based on CiteSpace in terms of spatiotemporal distribution, author distribution, subject categories, topic distribution, and references. The results can be elaborated from three aspects. Firstly, the number of annual publications related to CR has increased year by year in general over the past two decades. Secondly, a co-occurrence analysis of the output countries and authors shows that a few developed countries such as the United States, Canada, and the UK are the most active in carrying out CR and where regional academic communities represented by Sherry Grace and Ross Arena were formed. Thirdly, an analysis of the subject categories and topic distribution of the papers reveals that CR is a typical interdiscipline with a wide range of disciplines involved, including clinical medicine, basic medicine, public health management, and sports science. The research topics cover the participants and implementers, components, and the objectives and requirements of CR. The current research hotspots are the three core modalities of CR, namely patient education, exercise training and mental support, as well as mobile health (mHealth) dependent on computer science. In conclusion, this work has provided some useful information for acquiring knowledge about CR, including identifying potential collaborators for researchers interested in CR, and discovering research trends and hot topics in CR, which can offer some guidance for more extensive and in-depth CR-related studies in the future.

Mallika Srasri, Prayook Srivilai, Panida Loutchanwoot

Pueraria candollei var. mirifica (Fabaceae) root (PMR) has recently been developed as a potential selective estrogen receptor modulator (SERM) in menopausal women. Nowadays, many premenopausal women also take dietary PMR supplements, however, the exact biological effects of PMR have not been evaluated. This study included the application of the OECD guideline 407 for the assessment of 28-day oral exposure to PMR on pituitary-ovarian (PO) axis function and metabolic parameters in the premenopausal rat model. Ovary-intact adult rats were orally administrated with 10, 100, 750, 1000, and 1500 mg/kg body weight (BW)/day of PMR powder. The positive estrogenic group was given 2 mg 17β-estradiol (E2)/kg BW/day. Serum levels of reproductive hormones, lipid and thyroid parameters, estrous cycle determination, and histomorphometric and histopathological evaluations of the anterior pituitary, ovary, uterus, vagina, mammary gland, and liver were investigated. PMR displayed neutral effects on uterine, vaginal, and body weights, and circulating E2 and prolactin levels. PMR exerted E2-like effects by i) reducing ovarian and increasing hepatic weights, ii) decreasing serum gonadotropins, iii) lowering serum lipids without altering thyroid parameters, iv) increasing the prevalence of abnormal estrous cycles with prolonged estrus, v) increasing nuclear diameter of anterior pituitary cells, vi) decreasing ovarian size and follicular numbers and increasing follicular degeneration, vii) thickening of uterine myometrium and luminal epithelium, and vaginal epithelium, and viii) induction of mammary alveolar hyperplasia and ductal secretion. Unlike E2, the appearance of very small numbers of focal microvesicular steatosis in hepatocytes demonstrated mild toxicity at high PMR doses. This is the first report that high-dose PMR exerted actions exactly like E2 on gonadotrope-ovarian axis function and histology, lipid, and thyroid parameters without affecting uterine and vaginal growth in ovary-intact rats according to OECD guidelines.

Shreesh Raj Sammi, Shreesh Raj Sammi, Laura E. Jameson et al.

Neurological hazard assessment of industrial and pesticidal chemicals demands a substantial amount of time and resources. Caenorhabditis elegans is an established model organism in developmental biology and neuroscience. It presents an ideal test system with relatively fewer neurons (302 in hermaphrodites) versus higher-order species, a transparent body, short lifespan, making it easier to perform neurotoxic assessment in a time and cost-effective manner. Yet, no regulatory testing guidelines have been developed for C. elegans in the field of developmental and adult neurotoxicity. Here, we describe a set of morphological and behavioral assessment protocols to examine neurotoxicity in C. elegans with relevance to cholinergic and dopaminergic systems. We discuss the homology of human genes and associated proteins in these two signaling pathways and evaluate the morphological and behavioral endpoints of C. elegans in the context of published adverse outcome pathways of neurodegenerative diseases. We conclude that C. elegans neurotoxicity testing will not only be instrumental to eliminating mammalian testing in neurological hazard assessment but also lead to new knowledge and mechanistic validation in the adverse outcome pathway framework.

Sameera Ezzat, Ruzica Biga, Zisis Kozlakidis

Biobanking facilities are well established in high-income settings, where substantial funding has been invested in infrastructure. In contrast, such facilities are much less developed in resource-restricted settings. However, low-and middle-income countries (LMICs) still face a disproportionately high infectious diseases burden. Thus, the further development of infrastructure facilities, including biobanks is warranted as an important component of this unfolding clinical research environment. This perspective manuscript summarises the challenges and enablers for biobanking in LMICs, with a particular focus on infectious diseases, incorporating some of the lessons learned from the recent coronavirus disease 2019 (COVID-19) pandemic.

Sopiko Kvaratskhelia, Mamuka Gogiberidze, Mariam Orjonikidze et al.

No Abstract