Hasil untuk "Medical emergencies. Critical care. Intensive care. First aid"

Farzana Shariq Mujtaba, Biwei Cai, Nurhafizah Binte Mohamed Alifi et al.

Abstract Background With the increasing prevalence and complexity of disasters globally, it is essential to have well-prepared Emergency Medical Teams (EMTs) in place. Nurses play a crucial role within these teams, a role made more demanding in disaster settings due to the unique clinical, operational and psychological challenges. This study aimed to explore the experiences of Singapore’s Emergency Medical Team (SGEMT) nurses before, during and after their first deployment to Mandalay, Myanmar in April 2025. Methods This study adopted a descriptive qualitative research design to collect data. A purposive sampling method was used, and semi-structured virtual interviews were conducted within six weeks post-deployment. All ten nurses involved in SGEMT’s maiden deployment were interviewed. Inductive thematic analysis was conducted, which included analysing transcripts, extracting key sentences, deriving keywords, and identifying codes and sub-themes. A collaborative discussion was conducted to finalise the main themes. Results The analysis revealed four key categories: (a) preparedness and training, (b) clinical role adaptation and flexibility, (c) team dynamics and peer support and (d) attitudes towards humanitarian work. Findings indicated that nurses’ preparedness extended beyond clinical training to include logistical coordination, teamwork, and emotional readiness. Team-based simulation and cross-training were associated with increased confidence and adaptability, while peer support strengthened resilience. However, gaps were identified in environmental awareness, role clarity, and ethical preparedness. Conclusion This study found that while experiential training strengthened nurses’ confidence and teamwork during EMT deployment, gaps remained in environmental preparedness, role clarity, and ethical support. Nurses relied on adaptability, peer support, and trust in leadership to manage clinical, logistical, and emotional challenges, indicating the need for broader preparation beyond clinical skills in future EMT deployments.

Kuan-Yu Chi, MD, Michael G. Nanna, MD, MHS

Nebojsa Brezic, Strahinja Gligorevic, Tatjana Atanasijevic et al.

Fat embolism syndrome (FES) is a rare but serious complication most commonly associated with trauma, particularly long bone fractures. However, symptomatic FES remains a significant diagnostic and therapeutic challenge. We present the case of a 20-year-old man who, after sustaining multiple long bone fractures in a motorcycle accident and initially appearing stable, experienced a rapid and fatal progression of FES. This case underscores the unpredictable course of FES even in young, previously healthy individuals and highlights the critical need for early recognition and intervention. It also emphasizes the importance of identifying risk factors that may predict severe outcomes and mortality.

Tushara Surapaneni, Anna Patrikakou, Antigoni Faka et al.

Yansong Liu, Ronnie Stafford, Pramit Khetrapal et al.

For patients undergoing systemic cancer therapy, the time between clinic visits is full of uncertainties and risks of unmonitored side effects. To bridge this gap in care, we developed and prospectively trialed a multi-modal AI framework for remote patient monitoring (RPM). This system integrates multi-modal data from the HALO-X platform, such as demographics, wearable sensors, daily surveys, and clinical events. Our observational trial is one of the largest of its kind and has collected over 2.1 million data points (6,080 patient-days) of monitoring from 84 patients. We developed and adapted a multi-modal AI model to handle the asynchronous and incomplete nature of real-world RPM data, forecasting a continuous risk of future adverse events. The model achieved an accuracy of 83.9% (AUROC=0.70). Notably, the model identified previous treatments, wellness check-ins, and daily maximum heart rate as key predictive features. A case study demonstrated the model's ability to provide early warnings by outputting escalating risk profiles prior to the event. This work establishes the feasibility of multi-modal AI RPM for cancer care and offers a path toward more proactive patient support.(Accepted at Europe NeurIPS 2025 Multimodal Representation Learning for Healthcare Workshop. Best Paper Poster Award.)

Daniel Anadria, Roel Dobbe, Anastasia Giachanou et al.

In this work, we reflect on the data-driven modeling paradigm that is gaining ground in AI-driven automation of patient care. We argue that the repurposing of existing real-world patient datasets for machine learning may not always represent an optimal approach to model development as it could lead to undesirable outcomes in patient care. We reflect on the history of data analysis to explain how the data-driven paradigm rose to popularity, and we envision ways in which systems thinking and clinical domain theory could complement the existing model development approaches in reaching human-centric outcomes. We call for a purpose-driven machine learning paradigm that is grounded in clinical theory and the sociotechnical realities of real-world operational contexts. We argue that understanding the utility of existing patient datasets requires looking in two directions: upstream towards the data generation, and downstream towards the automation objectives. This purpose-driven perspective to AI system development opens up new methodological opportunities and holds promise for AI automation of patient care.

Yutong Song, Chenhan Lyu, Pengfei Zhang et al.

Mild-stage dementia patients primarily experience two critical symptoms: severe memory loss and emotional instability. To address these challenges, we propose DEMENTIA-PLAN, an innovative retrieval-augmented generation framework that leverages large language models to enhance conversational support. Our model employs a multiple knowledge graph architecture, integrating various dimensional knowledge representations including daily routine graphs and life memory graphs. Through this multi-graph architecture, DEMENTIA-PLAN comprehensively addresses both immediate care needs and facilitates deeper emotional resonance through personal memories, helping stabilize patient mood while providing reliable memory support. Our notable innovation is the self-reflection planning agent, which systematically coordinates knowledge retrieval and semantic integration across multiple knowledge graphs, while scoring retrieved content from daily routine and life memory graphs to dynamically adjust their retrieval weights for optimized response generation. DEMENTIA-PLAN represents a significant advancement in the clinical application of large language models for dementia care, bridging the gap between AI tools and caregivers interventions.

Orhun Demir, Omer Zuhtu Yondem, Mehmet Doganay

Conclusion: In the presence of amyloidosis with intervenient infections, the risk of chronic renal failure in HS cases can be kept in mind even if the patient is receiving TNF inhibitors.

Mohammed Alshalhoub, Faisal Alhusain, Feras Alsulaiman et al.

Abstract Introduction D-dimer is a marker of coagulation and fibrinolysis widely used in clinical practice for assessing thrombotic activity. While it is commonly ordered in the Emergency Department (ED) for suspected venous thromboembolism (VTE), elevated D-dimer levels can occur due to various other disorders. The aim of this study was to find out the causes of elevated D-dimer in patients presenting to a large ED in Saudi Arabia and evaluate the accuracy of D-dimer in diagnosing these conditions. Methods Data was collected from an electronic hospital information system of patients who visited the ED from January 2016 to December 2022. Demographic information, comorbidities, D-dimer levels, and diagnoses were analyzed. Statistical analysis was performed using the SPSS software. The different diagnoses associated with D-dimer levels were analyzed by plotting the median D-dimer levels for each diagnosis category and their interquartile ranges (IQR). The receiver operating characteristic (ROC) curves were calculated and their area under the curve (AUC) values were demonstrated. The optimal cut-off points for specific diseases were determined based on the ROC analysis, along with their corresponding sensitivities and specificities. Results A total of 19,258 patients with D-dimer results were included in the study. The mean age of the participants was 50 years with a standard deviation of ± 18. Of the patients, 66% were female and 21.2% were aged 65 or above. Additionally, 21% had diabetes mellitus, 20.4% were hypertensive, and 15.1% had been diagnosed with dyslipidemia. The median D-dimer levels varied across different diagnoses, with the highest level observed in aortic aneurysm 5.46 g/L. Pulmonary embolism (PE) and deep vein thrombosis (DVT) were found in 729 patients (3.8%) of our study population and their median D-dimer levels 3.07 g/L (IQR: 1.35–7.05 g/L) and 3.36 g/L (IQR: 1.06–8.38 g/L) respectively. On the other hand, 1767 patients (9.2%) were diagnosed with respiratory infections and 936 patients (4.9%) were diagnosed with shortness of breath (not specified) with median D-dimer levels of 0.76 g/L (IQR: 0.40–1.47 g/L) and 0.51 g/L (IQR: 0.29–1.06 g/L), respectively. D-dimer levels showed superior or excellent discrimination for PE (AUC = 0.844), leukemia (AUC = 0.848), and aortic aneurysm (AUC = 0.963). DVT and aortic dissection demonstrated acceptable discrimination, with AUC values of 0.795 and 0.737, respectively. D-dimer levels in respiratory infections and shortness of breath (not specified) exhibited poor to discriminatory performance. Conclusion This is the first paper to identify multiple causes of elevated D-dimer levels in Saudi Arabia population within the ED and it clearly highlights their accurate and diagnostic values. These findings draw attention to the importance of considering the specific clinical context and utilizing additional diagnostic tools when evaluating patients with elevated D-dimer levels.

Masoumeh Farhadi Nia, Mohsen Ahmadi, Elyas Irankhah

Artificial intelligence has dramatically reshaped our interaction with digital technologies, ushering in an era where advancements in AI algorithms and Large Language Models (LLMs) have natural language processing (NLP) systems like ChatGPT. This study delves into the impact of cutting-edge LLMs, notably OpenAI's ChatGPT, on medical diagnostics, with a keen focus on the dental sector. Leveraging publicly accessible datasets, these models augment the diagnostic capabilities of medical professionals, streamline communication between patients and healthcare providers, and enhance the efficiency of clinical procedures. The advent of ChatGPT-4 is poised to make substantial inroads into dental practices, especially in the realm of oral surgery. This paper sheds light on the current landscape and explores potential future research directions in the burgeoning field of LLMs, offering valuable insights for both practitioners and developers. Furthermore, it critically assesses the broad implications and challenges within various sectors, including academia and healthcare, thus mapping out an overview of AI's role in transforming dental diagnostics for enhanced patient care.

Subhashis Das, Debashis Naskar, Sara Rodriguez Gonzalez

The rise of digital platforms has led to an increasing reliance on technology-driven, home-based healthcare solutions, enabling individuals to monitor their health and share information with healthcare professionals as needed. However, creating an efficient care plan management system requires more than just analyzing hospital summaries and Electronic Health Records (EHRs). Factors such as individual user needs and social determinants of health, including living conditions and the flow of healthcare information between different settings, must also be considered. Challenges in this complex healthcare network involve schema diversity (in EHRs, personal health records, etc.) and terminology diversity (e.g., ICD, SNOMED-CT) across ancillary healthcare operations. Establishing interoperability among various systems and applications is crucial, with the European Interoperability Framework (EIF) emphasizing the need for patient-centric access and control of healthcare data. In this paper, we propose an integrated ontological model, the Common Semantic Data Model for Social Determinants of Health (CSSDH), by combining ISO/DIS 13940:2024 ContSys with WHO Social Determinants of Health. CSSDH aims to achieve interoperability within the Continuity of Care Network.

Hanyu Wang, Daniel Bochen Tan, Jason Cong

Quantum state preparation initializes the quantum registers and is essential for running quantum algorithms. Designing state preparation circuits that entangle qubits efficiently with fewer two-qubit gates enhances accuracy and alleviates coupling constraints on devices. Existing methods synthesize an initial circuit and leverage compilers to reduce the circuit's gate count while preserving the unitary equivalency. In this study, we identify numerous conditions within the quantum circuit where breaking local unitary equivalences does not alter the overall outcome of the state preparation (i.e., don't cares). We introduce a peephole optimization algorithm that identifies such unitaries for replacement in the original circuit. Exploiting these don't care conditions, our algorithm achieves a 36% reduction in the number of two-qubit gates compared to prior methods.

Yi Zhang, Yidong Zhao, Hui Xue et al.

Image registration is essential for medical image applications where alignment of voxels across multiple images is needed for qualitative or quantitative analysis. With recent advancements in deep neural networks and parallel computing, deep learning-based medical image registration methods become competitive with their flexible modelling and fast inference capabilities. However, compared to traditional optimization-based registration methods, the speed advantage may come at the cost of registration performance at inference time. Besides, deep neural networks ideally demand large training datasets while optimization-based methods are training-free. To improve registration accuracy and data efficiency, we propose a novel image registration method, termed Recurrent Inference Image Registration (RIIR) network. RIIR is formulated as a meta-learning solver to the registration problem in an iterative manner. RIIR addresses the accuracy and data efficiency issues, by learning the update rule of optimization, with implicit regularization combined with explicit gradient input. We evaluated RIIR extensively on brain MRI and quantitative cardiac MRI datasets, in terms of both registration accuracy and training data efficiency. Our experiments showed that RIIR outperformed a range of deep learning-based methods, even with only $5\%$ of the training data, demonstrating high data efficiency. Key findings from our ablation studies highlighted the important added value of the hidden states introduced in the recurrent inference framework for meta-learning. Our proposed RIIR offers a highly data-efficient framework for deep learning-based medical image registration.

R. Mehta, M. T. Pascual, Sharon B. Soroko et al.

Mei Fong Liew, W. Siow, Y. Yau et al.

Dear Editor, Although COVID-19 has not been officially labelled as a pandemic yet, the global burden of disease is significant and continues to rise. The virus has a high humanto-human transmissibility via airborne, droplet and contact routes [1]. Patient numbers can surge, and hospitals should be ready not just with the infrastructure, but also staff to be familiar with workflows. Kain and Fowler [2] have eloquently detailed influenza pandemic preparations for hospitals and intensive care units, and we feel the principles described in the article are relevant to COVID-19. Staff must consider patient transfers in between wards, as COVID-19 patients are admitted in isolation facilities to contain infected cases and to avoid nosocomial spread [1]. Infectious cases may be intentionally brought out of isolation rooms for various reasons. Intra-hospital transfer may be required from emergency departments to the wards, from the general floor to the intensive care unit and from the wards to radiology suites. Inter-hospital transfer may be required for extracorporeal membrane oxygenation (ECMO) if patients with COVID-19 develop severe acute respiratory distress syndrome within hospitals with only basic ventilation facilities. During episodes of patient transport outside of isolation, potential breaches of infection control can occur. At the same time, when COVID-19 patients turn ill during transport, their management is exceptionally challenging as accompanying staff would be wearing cumbersome personal protective equipment (PPE) [3]. Mitigating the spread of COVID-19 is a national priority in Singapore [4], and part of this effort involves planning and conducting safe patient transport for suspected or confirmed cases. HCWs who handle the transport of COVID-19 patients must consider the following principles (see Table 1): firstly, early recognition of the deteriorating patient; secondly, HCW safety; thirdly, bystander safety; fourthly, contingency plans for medical emergencies during transport; fifthly, post-transport decontamination. Specific action steps require designated zones for transport [5], sufficient supplies of PPE, staff training and support personnel like security officers and cleaning crews. Powered air-purifying respirators add a layer of safety on top of N95 respirators [3] and should be used if possible for high-risk cases, such as those requiring ambulance transport to ECMO centres. Given the continued global spread of COVID-19, we expect that more hospitals will need to deal with this disease. Haphazard transport of infected cases leading to nosocomial spread can stymie efforts to break the chains of transmission. We hope that our suggestions can aid others in ensuring safe patient transport for COVID-19 and reduce nosocomial spread.

Shangqi Gao, Hangqi Zhou, Yibo Gao et al.

Due to the cross-domain distribution shift aroused from diverse medical imaging systems, many deep learning segmentation methods fail to perform well on unseen data, which limits their real-world applicability. Recent works have shown the benefits of extracting domain-invariant representations on domain generalization. However, the interpretability of domain-invariant features remains a great challenge. To address this problem, we propose an interpretable Bayesian framework (BayeSeg) through Bayesian modeling of image and label statistics to enhance model generalizability for medical image segmentation. Specifically, we first decompose an image into a spatial-correlated variable and a spatial-variant variable, assigning hierarchical Bayesian priors to explicitly force them to model the domain-stable shape and domain-specific appearance information respectively. Then, we model the segmentation as a locally smooth variable only related to the shape. Finally, we develop a variational Bayesian framework to infer the posterior distributions of these explainable variables. The framework is implemented with neural networks, and thus is referred to as deep Bayesian segmentation. Quantitative and qualitative experimental results on prostate segmentation and cardiac segmentation tasks have shown the effectiveness of our proposed method. Moreover, we investigated the interpretability of BayeSeg by explaining the posteriors and analyzed certain factors that affect the generalization ability through further ablation studies. Our code will be released via https://zmiclab.github.io/projects.html, once the manuscript is accepted for publication.

Yuhao Huang, Xin Yang, Lian Liu et al.

The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It has achieved impressive results on various natural image segmentation tasks. However, medical image segmentation (MIS) is more challenging because of the complex modalities, fine anatomical structures, uncertain and complex object boundaries, and wide-range object scales. To fully validate SAM's performance on medical data, we collected and sorted 53 open-source datasets and built a large medical segmentation dataset with 18 modalities, 84 objects, 125 object-modality paired targets, 1050K 2D images, and 6033K masks. We comprehensively analyzed different models and strategies on the so-called COSMOS 1050K dataset. Our findings mainly include the following: 1) SAM showed remarkable performance in some specific objects but was unstable, imperfect, or even totally failed in other situations. 2) SAM with the large ViT-H showed better overall performance than that with the small ViT-B. 3) SAM performed better with manual hints, especially box, than the Everything mode. 4) SAM could help human annotation with high labeling quality and less time. 5) SAM was sensitive to the randomness in the center point and tight box prompts, and may suffer from a serious performance drop. 6) SAM performed better than interactive methods with one or a few points, but will be outpaced as the number of points increases. 7) SAM's performance correlated to different factors, including boundary complexity, intensity differences, etc. 8) Finetuning the SAM on specific medical tasks could improve its average DICE performance by 4.39% and 6.68% for ViT-B and ViT-H, respectively. We hope that this comprehensive report can help researchers explore the potential of SAM applications in MIS, and guide how to appropriately use and develop SAM.

Sergio Tascon-Morales, Pablo Márquez-Neila, Raphael Sznitman

Visual Question Answering (VQA) models aim to answer natural language questions about given images. Due to its ability to ask questions that differ from those used when training the model, medical VQA has received substantial attention in recent years. However, existing medical VQA models typically focus on answering questions that refer to an entire image rather than where the relevant content may be located in the image. Consequently, VQA models are limited in their interpretability power and the possibility to probe the model about specific image regions. This paper proposes a novel approach for medical VQA that addresses this limitation by developing a model that can answer questions about image regions while considering the context necessary to answer the questions. Our experimental results demonstrate the effectiveness of our proposed model, outperforming existing methods on three datasets. Our code and data are available at https://github.com/sergiotasconmorales/locvqa.

Ke Lin, Yonghua Hu, G. Kong

OBJECTIVES We aimed to construct a mortality prediction model using the random forest (RF) algorithm for acute kidney injury (AKI) patients in the intensive care unit (ICU), and compared its performance with that of two other machine learning models and the customized simplified acute physiology score (SAPS) II model. METHODS We used medical information mart for intensive care (MIMIC) III database for model development and performance comparison. The RF model uses the same predictor variable set as that of the SAPS II model. We also developed three other models and compared the RF model with the other three models in prediction performance. Three other models include support vector machine (SVM) model, artificial neural network (ANN) model and customized SAPS II model. In model comparison, the prediction performance of each model was assessed by the Brier score, the area under the receiver operating characteristic curve (AUROC), accuracy and F1 score. RESULTS The final cohort consisted of 19044 patients with AKI in the ICU. The observed in-hospital mortality of AKI patients is 13.6% in the final cohort. The results of model performance comparison show that the Brier score of the RF model is 0.085 (95%CI: 0.084-0.086) and AUROC of the RF model is 0.866 (95%CI: 0.862-0.870). The accuracy of the RF model is 0.728 (95%CI: 0.715-0.741). The F1 score of the RF model is 0.459 (95%CI: 0.449-0.470). The calibration plots show that the RF model slightly overestimates mortality in patients with low risk of death and underestimates mortality in patients with high risk of death. CONCLUSION There is great potential for the RF model in mortality prediction for AKI patients in ICU. The RF model may be helpful to aid ICU clinicians to make timely clinical intervention decisions for AKI patients, which is critical to help reduce the in-hospital mortality of AKI patients. A prospective study is necessary to evaluate the clinical utility of the RF model.

Y. Kong, Junyan Han, Xueying Wu et al.

As the coronavirus 2019 (COVID-19) continues to spread globally, hundreds of thousands have been infected, among whom approximately 15% of COVID-19 patients develop severe disease, and 5 to 6% are critically ill [1]. Critical patients of COVID-19 have a dramatically higher case fatality rate than severe cases. Thus, it is increasingly urgent to develop early and effective predictors to distinguish critical patients from severe patients. Storms of inflammatory cytokines and blood clots were reported to associate with severe disease and fatality of COVID-19 patients [2, 3]. We aimed to identify a biomarker for the detection of COVID-19 progression from numerous cytokines and coagulation indicators. We conducted a retrospective study based on patients with a laboratory-confirmed diagnosis of COVID-19 admitted to the intensive care unit in Beijing Ditan Hospital from January 20, 2020, to March 23, 2020. This study was approved by the Ethics Committee of Beijing Ditan Hospital. The severity of COVID-19 was defined according to the guidelines on the diagnosis and treatment of new coronavirus pneumonia (version 7). All baseline medical record information including demographic, data, complications, and laboratory results were obtained within the first day after hospital admission. Blood samples were collected at baseline and once every 3–7 days during hospitalization. Forty-five cytokines/ chemokines/growth factors in serum were measured using Luminex multiplex assay. Random forests machine learning classifier in Python environment was used for variable importance of the feature rankings. A receiver operating characteristic (ROC) curve was generated to evaluate the diagnostic accuracy of a protein. A total of 24 COVID-19 patients were enrolled in this study, including 14 (58.3%) severe patients and 10 (41.7%) critical patients (Table 1). Compared to the severe group, critical cases were of significantly older ages and showed higher white blood cell counts and neutrophil counts. Levels of VEGF-D, TNF-α, SCF, LIF, IL-2, IL-4, IL-6, IL-8, IL-10, IL-15, IL-17A, IL-18, IL-1β, and IFN-γ were significantly higher in the critical group than in the severe group (Table 1). Additionally, lymphocyte count, CRP, LDH, and coagulation indicators (D-dimer, platelet count, PT, and APTT), which were reported to associate with clinical outcome [4, 5], were also included in the random forests model. Strikingly, VEGF-D was identified as the most important indicator related to the severity of COVID-19 (ranked as 1, Fig. 1a). As expected, D-dimer, age, IL-6, and lymphocyte count associated with clinical outcomes of COVID-19 patients reported previously were also highly ranked. VEGF-D had a higher area under the curve (AUC) (0.836 (95% CI 0.648–1); Fig. 1b) than Ddimer (0.755 (95% CI 0.527–0.982); Fig. 1c). Consistently, VEGF-D levels were positively correlated with sequential organ failure assessment (SOFA) scores (Fig. 1d). As shown in Fig. 1e, critical patients had higher levels of VEGF-D than the severe cases during the whole course of hospitalization. To our knowledge, this is the first report of VEGF-D as a potential biomarker for detecting the progression of COVID-19. Despite limited evidence in COVID-19, previous studies demonstrated an important role of VEGF in the pathogenesis of acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) by its properties to increase vascular permeability. Furthermore, VEGF is regarded as an indirect procoagulant for altering the