The clinical burden of spleen-stomach disorders is substantial. While large language models (LLMs) offer new potential for medical applications, they face three major challenges in the context of integrative Chinese and Western medicine (ICWM): a lack of high-quality data, the absence of models capable of effectively integrating the reasoning logic of traditional Chinese medicine (TCM) syndrome differentiation with that of Western medical (WM) disease diagnosis, and the shortage of a standardized evaluation benchmark. To address these interrelated challenges, we propose DongYuan, an ICWM spleen-stomach diagnostic framework. Specifically, three ICWM datasets (SSDF-Syndrome, SSDF-Dialogue, and SSDF-PD) were curated to fill the gap in high-quality data for spleen-stomach disorders. We then developed SSDF-Core, a core diagnostic LLM that acquires robust ICWM reasoning capabilities through a two-stage training regimen of supervised fine-tuning. tuning (SFT) and direct preference optimization (DPO), and complemented it with SSDF-Navigator, a pluggable consultation navigation model designed to optimize clinical inquiry strategies. Additionally, we established SSDF-Bench, a comprehensive evaluation benchmark focused on ICWM diagnosis of spleen-stomach disorders. Experimental results demonstrate that SSDF-Core significantly outperforms 12 mainstream baselines on SSDF-Bench. DongYuan lays a solid methodological foundation and provides practical technical references for the future development of intelligent ICWM diagnostic systems.
Colonoscopy plays a pivotal role in colorectal cancer (CRC) screening and reduces CRC incidence and mortality. Its effectiveness depends on colonoscopist performance, which can vary. Missed lesions during colonoscopy can lead to post-colonoscopy CRC (PCCRC), making high-quality colonoscopy essential for maximizing the preventive benefit of CRC screening. This review highlights the significance of colonoscopy quality indicators and practices for improvement. Bowel preparation, cecal intubation, and withdrawal time are key process indicators for procedure quality and are closely associated with the adenoma detection rate (ADR) and PCCRC risk. Given the role of colonoscopy in preventing CRC through the removal of precancerous lesions, the ADR serves as the core quality metric and the most reliable predictor of PCCRC. Serrated polyps have gained attention in colonoscopy quality research, as 15% to 30% of CRCs arise from serrated lesions, with an increased detection rate inversely associated with PCCRC risk. This emphasizes the critical need for continuous efforts by colonoscopists to enhance performance quality. Systemic interventions, audits and feedback during endoscopist education, basic and enhanced withdrawal and inspection techniques, and technologies such as mucosal exposure devices and computer-aided detection have demonstrated efficacy in increasing the ADR. While artificial intelligence has shown promise in increasing the ADR, inconsistent outcomes in real-world studies underscore the continued importance of the fundamental aspects of high-quality colonoscopy techniques, including complete mucosal exposure. Understanding quality indicators and ensuring high-performance quality in daily practice will ultimately lead to better CRC prevention outcomes.
Diseases of the digestive system. Gastroenterology
Metaphorical expressions are abundant in Traditional Chinese Medicine (TCM), conveying complex disease mechanisms and holistic health concepts through culturally rich and often abstract terminology. Bridging these metaphors to anatomically driven Western medical (WM) concepts poses significant challenges for both automated language processing and real-world clinical practice. To address this gap, we propose a novel multi-agent and chain-of-thought (CoT) framework designed to interpret TCM metaphors accurately and map them to WM pathophysiology. Specifically, our approach combines domain-specialized agents (TCM Expert, WM Expert) with a Coordinator Agent, leveraging stepwise chain-of-thought prompts to ensure transparent reasoning and conflict resolution. We detail a methodology for building a metaphor-rich TCM dataset, discuss strategies for effectively integrating multi-agent collaboration and CoT reasoning, and articulate the theoretical underpinnings that guide metaphor interpretation across distinct medical paradigms. We present a comprehensive system design and highlight both the potential benefits and limitations of our approach, while leaving placeholders for future experimental validation. Our work aims to support clinical decision-making, cross-system educational initiatives, and integrated healthcare research, ultimately offering a robust scaffold for reconciling TCM's symbolic language with the mechanistic focus of Western medicine.
Natural language processing (NLP) has been traditionally applied to medicine, and generative large language models (LLMs) have become prominent recently. However, the differences between them across different medical tasks remain underexplored. We analyzed 19,123 studies, finding that generative LLMs demonstrate advantages in open-ended tasks, while traditional NLP dominates in information extraction and analysis tasks. As these technologies advance, ethical use of them is essential to ensure their potential in medical applications.
Abstract Background Simultaneous bilateral percutaneous nephrolithotomy (PCNL) offers the advantage of treating stones in both kidneys, thereby reducing the need for multiple surgeries. Due to the limited number of cases, simultaneous PCNL has unwarranted safety and efficacy concerns. This study aimed to evaluate the complications and stone-free rates of simultaneous bilateral PCNL in the treatment of bilateral large complex stones and to compare different access methods. Method Between January 2012 and December 2022, 36 consecutive patients who underwent simultaneous bilateral PCNL for large complex renal stones were enrolled. Guy’s stone score (GSS) was used to assess the complexity of stone. The preoperative, intraoperative, and post-operative parameters were assessed. The patients were first categorized based on channel size (conventional vs. mini-PCNL), and then further sub-grouped according to specific combinations of tract size and dilation method for comparative analysis. Results Thirty-six consecutive patients (72 renal units) underwent simultaneous bilateral PCNL. The median stone burden was 602.43 mm2 (interquartile range: 225–1332.72 mm2), mean surgical duration was 70.9 ± 29.6 minutes for each renal unit (range, 30–140 minutes), and the mean hematocrit reduction was 6.8±8.4%. The mean length of stay was four days, and the stone-free rate was 81.9%. Notably, eGFR (estimated Glomerular filtration rate) values showed significant improvement at one-year follow-up (p < 0.001), with 29.4% of patients showing clinical downstaging. The overall complication rate was 16.7%, with the majority of complications being transient fever. Mini-PCNL had a shorter length of stay (p < 0.05). The complication rates for Amplatz, balloon, and mini-PCNL were 13.3%, 23.1%, and 12.5%, respectively. The post-operative radiographic stone-free rate (SFR) for 72 renal units was 81.9%, with the highest rate in the mini-PCNL group (93.7%). Conclusions There was no increase in the rate of complications compared to unilateral PCNL. This study provides valuable insights into surgical outcomes using different access methods.
Data pre-processing is a significant step in machine learning to improve the performance of the model and decreases the running time. This might include dealing with missing values, outliers detection and removing, data augmentation, dimensionality reduction, data normalization and handling the impact of confounding variables. Although it is found the steps improve the accuracy of the model, but they might hinder the explainability of the model if they are not carefully considered especially in medicine. They might block new findings when missing values and outliers removal are implemented inappropriately. In addition, they might make the model unfair against all the groups in the model when making the decision. Moreover, they turn the features into unitless and clinically meaningless and consequently not explainable. This paper discusses the common steps of the data preprocessing in machine learning and their impacts on the explainability and interpretability of the model. Finally, the paper discusses some possible solutions that improve the performance of the model while not decreasing its explainability.
Juan Miguel Lopez Alcaraz, Hjalmar Bouma, Nils Strodthoff
Background: AI-driven prediction algorithms have the potential to enhance emergency medicine by enabling rapid and accurate decision-making regarding patient status and potential deterioration. However, the integration of multimodal data, including raw waveform signals, remains underexplored in clinical decision support. Methods: We present a dataset and benchmarking protocol designed to advance multimodal decision support in emergency care. Our models utilize demographics, biometrics, vital signs, laboratory values, and electrocardiogram (ECG) waveforms as inputs to predict both discharge diagnoses and patient deterioration. Results: The diagnostic model achieves area under the receiver operating curve (AUROC) scores above 0.8 for 609 out of 1,428 conditions, covering both cardiac (e.g., myocardial infarction) and non-cardiac (e.g., renal disease, diabetes) diagnoses. The deterioration model attains AUROC scores above 0.8 for 14 out of 15 targets, accurately predicting critical events such as cardiac arrest, mechanical ventilation, ICU admission, and mortality. Conclusions: Our study highlights the positive impact of incorporating raw waveform data into decision support models, improving predictive performance. By introducing a unique, publicly available dataset and baseline models, we provide a foundation for measurable progress in AI-driven decision support for emergency care.
Jasmine Chiat Ling Ong, Liyuan Jin, Kabilan Elangovan
et al.
Importance: We introduce a novel Retrieval Augmented Generation (RAG)-Large Language Model (LLM) framework as a Clinical Decision Support Systems (CDSS) to support safe medication prescription. Objective: To evaluate the efficacy of LLM-based CDSS in correctly identifying medication errors in different patient case vignettes from diverse medical and surgical sub-disciplines, against a human expert panel derived ground truth. We compared performance for under 2 different CDSS practical healthcare integration modalities: LLM-based CDSS alone (fully autonomous mode) vs junior pharmacist + LLM-based CDSS (co-pilot, assistive mode). Design, Setting, and Participants: Utilizing a RAG model with state-of-the-art medically-related LLMs (GPT-4, Gemini Pro 1.0 and Med-PaLM 2), this study used 61 prescribing error scenarios embedded into 23 complex clinical vignettes across 12 different medical and surgical specialties. A multidisciplinary expert panel assessed these cases for Drug-Related Problems (DRPs) using the PCNE classification and graded severity / potential for harm using revised NCC MERP medication error index. We compared. Results RAG-LLM performed better compared to LLM alone. When employed in a co-pilot mode, accuracy, recall, and F1 scores were optimized, indicating effectiveness in identifying moderate to severe DRPs. The accuracy of DRP detection with RAG-LLM improved in several categories but at the expense of lower precision. Conclusions This study established that a RAG-LLM based CDSS significantly boosts the accuracy of medication error identification when used alongside junior pharmacists (co-pilot), with notable improvements in detecting severe DRPs. This study also illuminates the comparative performance of current state-of-the-art LLMs in RAG-based CDSS systems.
Vadim K. Weinstein, Tamara Alshammari, Kalle G. Timperi
et al.
When designing a robot's internal system, one often makes assumptions about the structure of the intended environment of the robot. One may even assign meaning to various internal components of the robot in terms of expected environmental correlates. In this paper we want to make the distinction between robot's internal and external worlds clear-cut. Can the robot learn about its environment, relying only on internally available information, including the sensor data? Are there mathematical conditions on the internal robot system which can be internally verified and make the robot's internal system mirror the structure of the environment? We prove that sufficiency is such a mathematical principle, and mathematically describe the emergence of the robot's internal structure isomorphic or bisimulation equivalent to that of the environment. A connection to the free-energy principle is established, when sufficiency is interpreted as a limit case of surprise minimization. As such, we show that surprise minimization leads to having an internal model isomorphic to the environment. This also parallels the Good Regulator Principle which states that controlling a system sufficiently well means having a model of it. Unlike the mentioned theories, ours is discrete, and non-probabilistic.
BACKGROUND:
Owing to the adverse effects of unilateral neglect (UN) on rehabilitation outcomes, fall risk, and activities of daily living, this field has gradually got considerable interest. Notwithstanding, there is presently an absence of efficient portrayals of the entire research field; hence, the motivation behind this study was to dissect and evaluate the literature published in the field of UN following stroke and other nonprogressive brain injuries to identify hotspots and trends for future research.
MATERIALS AND METHODS:
Original articles and reviews related to UN from 1970 to 2022 were retrieved from the Science Citation Index Expanded of the Web of Science Core Collection. CiteSpace, VOSviewer, and Bibliometrix software were used to observe publication fields, countries, and authors.
RESULTS:
A total of 1,202 publications were incorporated, consisting of 92% of original articles, with an overall fluctuating upward trend in the number of publications. Italy, the United Kingdom, and the United States made critical contributions, with Neuropsychologia being the most persuasive academic journal, and Bartolomeo P. ranked first in both the quantity of publications and co-citations. Keywords were divided into four clusters, and burst keyword detection demonstrated that networks and virtual reality might additionally emerge as frontiers of future development and warrant additional attention.
CONCLUSIONS:
UN is an emerging field, and this study presents the first bibliometric analysis to provide a comprehensive overview of research in the field. The insights and guidance garnered from our research on frontiers, trends, and popular topics could prove highly valuable in facilitating the rapid development of this field while informing future research directions.
Medical technology, Diseases of the circulatory (Cardiovascular) system
ABSTRACT Introduction Lung adenocarcinoma (LUAD) is one of the major histopathological types of non‐small cell lung cancer (NSCLC), including solid, acinar, lepidic, papillary and micropapillary subtypes. Increasing evidence has shown that micropapillary LUAD is positively associated with a higher percentage of driver gene mutations, a higher incidence of metastasis and a poorer prognosis, while lepidic LUAD has a relatively better prognosis. However, the novel genetic change and its underlying mechanism in the progression of micropapillary LUAD have not been exactly determined. Methods A total of 181 patients with LUAD who underwent surgery at the First Affiliated Hospital of Huzhou University from January 2020 to December 2022 were enrolled. Three predominant lepidic and three predominant micropapillary LUAD tissue samples were carried out using whole‐exome sequencing. Comprehensive analysis of genomic variations and the difference between lepidic and micropapillary LUAD was performed. In addition, the TMEM229A Q200del mutation was verified using our cohort and TCGA‐LUAD datasets. The correlations between the TMEM229A Q200del mutation and the clinicopathological characteristics of patients with LUAD were further analyzed. The functions and mechanisms of TMEM229A Q200del on NSCLC cell proliferation and migration were also determined. Results The frequency of genomic changes in patients with micropapillary LUAD was higher than that in patients with lepidic LUAD. Mutations in EGFR, ATXN2, C14orf180, MUC12, NOTCH1, and PKD1L2 were concomitantly detected in three predominant micropapillary and three predominant lepidic LUAD cases. The TMEM229A Q200del mutation was only mutated in lepidic LUAD. Additionally, the TMEM229A Q200del mutation had occurred in 16 (8.8%) patients, and not found TMEM229A R76H and M346T mutations in our cohort, while TMEM229A mutations (R76H, M346T, and Q200del) occurred only in 1.0% of the TCGA‐LUAD cohort. Further correlation analysis between the TMEM229A Q200del mutation and clinicopathological characteristics suggested that a lower frequency of the Q200del mutation was significantly associated with positive lymph node metastasis, advanced TNM stage, positive cancer thrombus, and pathological features. Finally, overexpression of TMEM229A Q200del suppressed NSCLC cell proliferation and migration in vitro. Mechanistically, overexpression of TMEM229A and TMEM229A Q200del both reduced the expression level of phosphorylated (p)‐ERK and p‐AKT (Ser473), and the reduced protein level of p‐ERK in the TMEM229A Q200del group was more pronounced compared to the TMEM229A group. Conclusion Our results demonstrated that the TMEM229A Q200del mutant may play a protective role in the progression of LUAD via inactivating ERK pathway, providing a potential therapeutic target in LUAD.
Cerebrovascular dysfunction, leading to inadequate brain perfusion and oxygenation, is a major contributor to cognitive decline and dementia. Chronic hypoxia is a putative mechanism of vascular-mediated brain damage, particularly in relation to white matter lesions, as demonstrated by human neuroimaging and histopathology studies. Moreover, increasing evidence suggests that microglia, the primary immune cells of the brain parenchyma, may play a key role in modulating cerebrovascular disease outcomes. Indeed, unpublished work from our lab using a model of chronic cerebral hypoperfusion found greater vascular and white matter abnormalities concomitant with reduced microglial-vascular interactions in mice lacking the microglial immunoreceptor triggering receptor expressed on myeloid cells 2 (TREM2). However, the underlying mechanisms remain incompletely understood. Therefore, this project aims to further investigate whether microglial TREM2 signalling contributes to cerebrovascular resilience, and specifically vasculoprotection, focusing on the context of hypoxia. To address this, we are housing mice at 8%O2 to achieve chronic mild hypoxia (CMH). As corroborated by our studies, CMH induces cerebral microbleeds associated with parenchymal fibrinogen leakage in both grey and white matter regions (Figure 1), and is thus a reductionist approach well-suited for examining microglial mechanisms conferring vasculoprotection. Ongoing studies in young (5-6 months) and aged (15-18 months) cohorts are utilising histology and immunostaining to determine the impact of TREM2 deficiency and ageing on CMH-induced phenotypes, with particular focus on profiling microbleed burden, BBB integrity and interactions between microglia and other cell types within the neurovascular unit. Given that TREM2 is a key regulator of microglial metabolism and lipid processing, future work will utilise flow cytometry and spatial lipidomics to characterise microglia and brain lipid metabolism during CMH, thus providing insight into immunometabolic changes that may underpin microglial vasculoprotection in hypoxia. Findings from these studies will increase our understanding of microglia-vascular interactions, which can ultimately be exploited to promote resilience to cerebrovascular and other hypoxia- related pathologies.
Specialties of internal medicine, Neurosciences. Biological psychiatry. Neuropsychiatry
In both quantum mechanics and relativity theory, the concept of the observer plays a critical role. However, there is no consensus on the definition of observer in these theories. Following Einstein's thought experiments, one could ask: What would it look like to sit inside a photon or to be a photon? And what type of observer could represent this more global perspective of the photon's interior? To address these questions, we introduce the concepts of internal and external observers with a focus on their relationship in quantum theory and relativity theory. The internal observer, associated with the internal observables super-algebra, glues the external interactions. Drawing inspiration from the advancements in abstract algebraic topology, we propose mathematical representation of the internal observer. We also outline principles for ensuring the consistency of observers in terms of information theory. It becomes evident, through the analysis of the introduced hierarchy of observers, that entanglement is a primitive of space-time causal relationships. While external observers must abide by the relativistic causality linked with the no-signaling principle in quantum mechanics, the internal observer is inherently non-local and may be acausal. However, its consistency is maintained through the formulation of the self-consistency principle. One of the goals of this paper is to construct the representation of the internal observer from the local external algebra of observables, which can be associated with external observers. Additionally, we demonstrate how the concepts of internal and external observers can be applied in the fields of quantum information theory, algebraic quantum field theory, and loop quantum gravity. The concept of internal observer seems to be also fundamental for further development of quantum gravity.
This research presents a novel Discrete Event Simulation (DES) of the Lloyd's of London specialty insurance market, exploring complex market dynamics that have not been previously studied quantitatively. The proof-of-concept model allows for the simulation of various scenarios that capture important market phenomena such as the underwriting cycle, the impact of risk syndication, and the importance of appropriate exposure management. Despite minimal calibration, our model has shown that it is a valuable tool for understanding and analysing the Lloyd's of London specialty insurance market, particularly in terms of identifying areas for further investigation for regulators and participants of the market alike. The results generate the expected behaviours that, syndicates (insurers) are less likely to go insolvent if they adopt sophisticated exposure management practices, catastrophe events lead to more defined patterns of cyclicality and cause syndicates to substantially increase their premiums offered. Lastly, syndication enhances the accuracy of actuarial price estimates and narrows the divergence among syndicates. Overall, this research offers a new perspective on the Lloyd's of London market and demonstrates the potential of individual-based modelling (IBM) for understanding complex financial systems.
Nicola Scichilone, Andrew Whittamore, Chris White
et al.
Abstract Background Chronic obstructive pulmonary disease (COPD) is a common condition that causes irreversible airway obstruction. Fatigue and exertional dyspnoea, for example, have a detrimental impact on the patient’s daily life. Current research has revealed the need to empower the patient, which can result in not only educated and effective decision-making, but also a considerable improvement in patient satisfaction and treatment compliance. The current study aimed to investigate the perspectives and requirements of people living with COPD to possibly explore new ways to manage their disease. Methods Adults with COPD from 8 European countries were interviewed by human factor experts to evaluate their disease journey through the gathering of information on the age, performance, length, and impact of diagnosis, symptoms progression, and family and friends' reactions. The assessment of present symptoms, services, and challenges was performed through a 90-min semi-structured interview. To identify possible unmet needs of participants, a generic thematic method was used to explore patterns, themes, linkages, and sequences within the data collected. Flow charts and diagrams were created to communicate the primary findings. Following analysis, the data was consolidated into cohesive insights and conversation themes relevant to determining the patient's unmet needs. Results The 62, who voluntarily accepted to be interviewed, were patients (61% females, aged 32–70 years) with a COPD diagnosis for at least 6 months with stable symptoms of different severity. The main challenges expressed by the patients were the impact on their lifestyle, reduced physical activity, and issues with their mobility. About one-fourth had challenges with their symptoms or medication including difficulty in breathing. Beyond finding a cure for COPD was the primary goal for patients, their main needs were to receive adequate information on the disease and treatments, and to have adequate support to improve physical activity and mobility, helpful both for patients and their families. Conclusions These results could aid in the creation of new ideas and concepts to improve our patient’s quality of life, encouraging a holistic approach to people living with COPD and reinforcing the commitment to understanding their needs.
Alvaro Andres Alvarez Peralta, Priya Desai, Somalee Datta
This manuscript explores linking real-world patient data with external death data in the context of research Clinical Data Warehouses (r-CDWs). We specifically present the linking of Electronic Health Records (EHR) data for Stanford Health Care (SHC) patients and data from the Social Security Administration (SSA) Limited Access Death Master File (LADMF) made available by the US Department of Commerce's National Technical Information Service (NTIS). The data analysis framework presented in this manuscript extends prior approaches and is generalizable to linking any two cross-organizational real-world patient data sources. Electronic Health Record (EHR) data and NTIS LADMF are heavily used resources at other medical centers and we expect that the methods and learnings presented here will be valuable to others. Our findings suggest that strong linkages are incomplete and weak linkages are noisy i.e., there is no good linkage rule that provides coverage and accuracy. Furthermore, the best linkage rule for any two datasets is different from the best linkage rule for two other datasets i.e., there is no generalization of linkage rules. Finally, LADMF, a commonly used external death data resource for r-CDWs, has a significant gap in death data making it necessary for r-CDWs to seek out more than one external death data source. We anticipate that presentation of multiple linkages will make it hard to present the linkage outcome to the end user. This manuscript is a resource in support of Stanford Medicine STARR (STAnford medicine Research data Repository) r-CDWs. The data are stored and analyzed as PHI in our HIPAA-compliant data center and are used under research and development (R&D) activities of STARR IRB.
Arman Rahmim, Tyler J. Bradshaw, Irène Buvat
et al.
The SNMMI Artificial Intelligence (SNMMI-AI) Summit, organized by the SNMMI AI Task Force, took place in Bethesda, MD on March 21-22, 2022. It brought together various community members and stakeholders from academia, healthcare, industry, patient representatives, and government (NIH, FDA), and considered various key themes to envision and facilitate a bright future for routine, trustworthy use of AI in nuclear medicine. In what follows, essential issues, challenges, controversies and findings emphasized in the meeting are summarized.
A. L. Chilingaryan, L. G. Tunyan, K. G. Adamyan
et al.
Aim. To study the structural and functional left heart parameters in patients with severe aortic stenosis (AS) and preserved ejection fraction (EF) in order to determine the risk of atrial fibrillation (AF).Material and methods. The study included 84 patients (men, 37; mean age, 68±8 years) with severe AS and EF >55%. All patients had sinus rhythm and were asymptomatic. Echocardiography was performed to assess longitudinal strain of the left ventricle (LVLS), right ventricle, left atrium (LALS) and the left atrial stiffness (LAS) using the speckle tracking method. Left ventricular mass index (LVMI) and maximum left atrium volume index (LAVI) were also determined. Patients were followed up for 1 year.Results. AF was reported in 27 (32%) patients, of which 9 (33%) had asymptomatic AF episodes detected by 48-hour electrocardiography. Eighteen (67%) patients with AF felt palpitations. Patients with and without episodes of atrial fibrillation had non-significant differences in LVMI, LAVI, and LVLS. Patients with atrial fibrillation had a lower LALS and a higher LAS compared with patients without atrial fibrillation. Regression analysis revealed that LALS and LAS were independent predictors of AF.Conclusion. AF develops in about one third of asymptomatic patients with severe AS and normal EF. The development of AF predisposes to the onset of AS symptoms in most patients. LALS and LAS were predictors of AF in these patients. Identification of patients at risk of AF will allow for earlier aortic valve replacement.
Diseases of the circulatory (Cardiovascular) system