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

Bahadir Ciftci, Serkan Tulgar, Aybegüm Fazlioglu et al.

Ping Chen, Zicheng Huang, Xiangming Wang et al.

We propose VL-DUN, a principled framework for joint All-in-One Medical Image Restoration and Segmentation (AiOMIRS) that bridges the gap between low-level signal recovery and high-level semantic understanding. While standard pipelines treat these tasks in isolation, our core insight is that they are fundamentally synergistic: restoration provides clean anatomical structures to improve segmentation, while semantic priors regularize the restoration process. VL-DUN resolves the sub-optimality of sequential processing through two primary innovations. (1) We formulate AiOMIRS as a unified optimization problem, deriving an interpretable joint unfolding mechanism where restoration and segmentation are mathematically coupled for mutual refinement. (2) We introduce a frequency-aware Mamba mechanism to capture long-range dependencies for global segmentation while preserving the high-frequency textures necessary for restoration. This allows for efficient global context modeling with linear complexity, effectively mitigating the spectral bias of standard architectures. As a pioneering work in the AiOMIRS task, VL-DUN establishes a new state-of-the-art across multi-modal benchmarks, improving PSNR by 0.92 dB and the Dice coefficient by 9.76\%. Our results demonstrate that joint collaborative learning offers a superior, more robust solution for complex clinical workflows compared to isolated task processing. The codes are provided in https://github.com/cipi666/VLDUN.

Yuqing Xiao, John Grundy, Anuradha Madugalla et al.

Requirements engineering for aged-care digital health must account for human aspects, because requirement priorities are shaped not only by technical functionality but also by stakeholders' health conditions, socioeconomics, and lived experience. Knowing which human aspects matter most, and for whom, is critical for inclusive and evidence-based requirements prioritisation. Yet in practice, while some studies have examined human aspects in RE, they have largely relied on expert judgement or model-driven analysis rather than large-scale user studies with meaningful human-in-the-loop validation to determine which aspects matter most and why. To address this gap, we conducted a mixed-methods study with 103 older adults, 105 developers, and 41 caregivers. We first applied an explainable machine learning to identify the human aspects most strongly associated with requirement priorities across 8 aged-care digital health themes, and then conducted 12 semi-structured interviews to validate and interpret the quantitative patterns. The results identify the key human aspects shaping requirement priorities, reveal their directional effects, and expose substantial misalignment across stakeholder groups. Together, these findings show that human-centric requirements analysis should engage stakeholder groups explicitly rather than collapsing their perspectives into a single aggregate view. This paper contributes an identification of the key human aspects driving requirement priorities in aged-care digital health and an explainable, human-centric RE framework that combines ML-derived importance rankings with qualitative validation to surface the stakeholder misalignments that inclusive requirements engineering must address.

John E. Brush, Jr., MD, Chungsoo Kim, PharmD, PhD, Yuntian Liu, MPH et al.

Background: Neighborhood-level social vulnerability is associated with hypertension prevalence and severity and with cardiovascular complications in conditions other than hypertension, but its association with cardiovascular complications in patients with hypertension is understudied. Objectives: The aim of the study was to examine how the neighborhood-level social vulnerability index (SVI) influences cardiovascular outcomes in a large, diverse cohort of patients with hypertension. Methods: We used electronic health data to examine the association between census tract-level rankings for the SVI with a composite endpoint of incident myocardial infarction, congestive heart failure, or stroke. Results: In a longitudinal cohort of 330,972 patients with hypertension followed for a median of 6.6 years, the neighborhood-level SVI was significantly associated with the composite endpoint after adjustment for demographics, baseline body mass index and blood pressure (BP), and comorbidities (HR for quartile 4 [most disadvantaged group] vs quartile 1 = 1.31 [95% CI: 1.25-1.38], P < 0.001). Patients living in quartile 4 SVI areas had a significantly lower BP control rate compared with patients living in quartile 1 SVI areas (70.3% vs 74.8%, P < 0.001). Patients living in SVI quartile 4 areas were disproportionately Black (53.8%). Compared with the White race, the Black race was negatively associated with the composite outcome after adjustment for the SVI quartile, and other clinical factors (HR: 0.89 [95% CI: 0.86-0.92], P < 0.001). Conclusions: Neighborhood-level social vulnerability was strongly associated with adverse cardiovascular outcomes and poorer BP control and may be a driver of racial disparities in hypertension. These findings highlight the potential of leveraging social vulnerability indices for tailored interventions in hypertension management.

Marvellous Eyube, Courage Enuesueke, Marvellous Alimikhena

Biocompatibility is a critical factor in the application of nanomaterials in medical fields, as these materials must interact safely and effectively with biological systems to be viable for therapeutic and diagnostic use. This article investigates the biocompatibility of nanomaterials, focusing on their interactions with biological cells, tissues, and the immune system. Key properties such as surface chemistry, size, shape, and material composition are examined, as they significantly influence the biological response. The article explores the role of nanomaterials in medical applications, including drug delivery, diagnostic imaging, and tissue engineering, while discussing the challenges involved in enhancing their biocompatibility. A case study on the CaO-CaP binary system is presented, showcasing the use of calcium oxide (CaO) and calcium phosphate (CaP) nanoparticles in bone tissue engineering. This system is widely investigated for its ability to mimic the mineral content of bone and promote osteogenesis, highlighting both its therapeutic potential and challenges in ensuring safe biocompatibility in clinical settings. The article concludes by reviewing strategies to optimize the biocompatibility of nanomaterials and discussing future directions for research in advancing their applications in medical treatments.

Heyuan Huang, Alexandra DeLucia, Vijay Murari Tiyyala et al.

While Large Language Models (LLMs) can generate fluent and convincing responses, they are not necessarily correct. This is especially apparent in the popular decompose-then-verify factuality evaluation pipeline, where LLMs evaluate generations by decomposing the generations into individual, valid claims. Factuality evaluation is especially important for medical answers, since incorrect medical information could seriously harm the patient. However, existing factuality systems are a poor match for the medical domain, as they are typically only evaluated on objective, entity-centric, formulaic texts such as biographies and historical topics. This differs from condition-dependent, conversational, hypothetical, sentence-structure diverse, and subjective medical answers, which makes decomposition into valid facts challenging. We propose MedScore, a new pipeline to decompose medical answers into condition-aware valid facts and verify against in-domain corpora. Our method extracts up to three times more valid facts than existing methods, reducing hallucination and vague references, and retaining condition-dependency in facts. The resulting factuality score substantially varies by decomposition method, verification corpus, and used backbone LLM, highlighting the importance of customizing each step for reliable factuality evaluation by using our generalizable and modularized pipeline for domain adaptation.

Yuting Zhang, Karina V. Bunting, Asgher Champsi et al.

Cardiovascular diseases (CVDs) remain the foremost cause of mortality worldwide, a burden worsened by a severe deficit of healthcare workers. Artificial intelligence (AI) agents have shown potential to alleviate this gap through automated detection and proactive screening, yet their clinical application remains limited by: 1) rigid sequential workflows, whereas clinical care often requires adaptive reasoning that select specific tests and, based on their results, guides personalised next steps; 2) reliance solely on intrinsic model capabilities to perform role assignment without domain-specific tool support; 3) general and static knowledge bases without continuous learning capability; and 4) fixed unimodal or bimodal inputs and lack of on-demand visual outputs when clinicians require visual clarification. In response, a multimodal framework, CardAIc-Agents, was proposed to augment models with external tools and adaptively support diverse cardiac tasks. First, a CardiacRAG agent generated task-aware plans from updatable cardiac knowledge, while the Chief agent integrated tools to autonomously execute these plans and deliver decisions. Second, to enable adaptive and case-specific customization, a stepwise update strategy was developed to dynamically refine plans based on preceding execution results, once the task was assessed as complex. Third, a multidisciplinary discussion team was proposed which was automatically invoked to interpret challenging cases, thereby supporting further adaptation. In addition, visual review panels were provided to assist validation when clinicians raised concerns. Experiments across three datasets showed the efficiency of CardAIc-Agents compared to mainstream Vision-Language Models (VLMs) and state-of-the-art agentic systems.

Marie-Laure Nisolle, Arnaud Bourguignon

Esra Adiyeke, Yuanfang Ren, Benjamin Shickel et al.

Background: Acute kidney injury (AKI), the decline of kidney excretory function, occurs in up to 18% of hospitalized admissions. Progression of AKI may lead to irreversible kidney damage. Methods: This retrospective cohort study includes adult patients admitted to a non-intensive care unit at the University of Pittsburgh Medical Center (UPMC) (n = 46,815) and University of Florida Health (UFH) (n = 127,202). We developed and compared deep learning and conventional machine learning models to predict progression to Stage 2 or higher AKI within the next 48 hours. We trained local models for each site (UFH Model trained on UFH, UPMC Model trained on UPMC) and a separate model with a development cohort of patients from both sites (UFH-UPMC Model). We internally and externally validated the models on each site and performed subgroup analyses across sex and race. Results: Stage 2 or higher AKI occurred in 3% (n=3,257) and 8% (n=2,296) of UFH and UPMC patients, respectively. Area under the receiver operating curve values (AUROC) for the UFH test cohort ranged between 0.77 (UPMC Model) and 0.81 (UFH Model), while AUROC values ranged between 0.79 (UFH Model) and 0.83 (UPMC Model) for the UPMC test cohort. UFH-UPMC Model achieved an AUROC of 0.81 (95% confidence interval [CI] [0.80, 0.83]) for UFH and 0.82 (95% CI [0.81,0.84]) for UPMC test cohorts; an area under the precision recall curve values (AUPRC) of 0.6 (95% CI, [0.05, 0.06]) for UFH and 0.13 (95% CI, [0.11,0.15]) for UPMC test cohorts. Kinetic estimated glomerular filtration rate, nephrotoxic drug burden and blood urea nitrogen remained the top three features with the highest influence across the models and health centers. Conclusion: Locally developed models displayed marginally reduced discrimination when tested on another institution, while the top set of influencing features remained the same across the models and sites.

Oliver Cassidy, Marta Andronic, Samuel Coward et al.

Lookup tables (LUTs) are frequently used to efficiently store arrays of precomputed values for complex mathematical computations. When used in the context of neural networks, these functions exhibit a lack of recognizable patterns which presents an unusual challenge for conventional logic synthesis techniques. Several approaches are known to break down a single large lookup table into multiple smaller ones that can be recombined. Traditional methods, such as plain tabulation, piecewise linear approximation, and multipartite table methods, often yield inefficient hardware solutions when applied to LUT-based NNs. This paper introduces ReducedLUT, a novel method to reduce the footprint of the LUTs by injecting don't cares into the compression process. This additional freedom introduces more self-similarities which can be exploited using known decomposition techniques. We then demonstrate a particular application to machine learning; by replacing unobserved patterns within the training data of neural network models with don't cares, we enable greater compression with minimal model accuracy degradation. In practice, we achieve up to $1.63\times$ reduction in Physical LUT utilization, with a test accuracy drop of no more than $0.01$ accuracy points.

Emmanuel Tung, Ali Mostafavi, Maoxu Li et al.

Disruptions to medical infrastructure during disasters pose significant risks to critically ill patients with advanced chronic kidney disease or end-stage renal disease. To enhance patient access to dialysis treatment under such conditions, it is crucial to assess the vulnerabilities of critical care facilities to hazardous events. This study proposes optimization models for patient reallocation and the strategic placement of temporary medical facilities to bolster the resilience of the critical care system, with a focus on equitable outcomes. Utilizing human mobility data from Texas, we evaluate patient access to critical care and dialysis centers under simulated hazard scenarios. The proposed bio-inspired optimization model, based on the Ant Colony optimization method, efficiently reallocates patients to mitigate disrupted access to dialysis facilities. The model outputs offer valuable insights into patient and hospital preparedness for disasters. Overall, the study presents a data-driven, analytics-based decision support tool designed to proactively mitigate potential disruptions in access to critical care facilities during disasters, tailored to the needs of health officials, emergency managers, and hospital system administrators in both the private and public sectors.

Yugo Wakayama, Satoshi Yoshimura, Katsuyuki Hanabusa et al.

Eshaan J. Daas, Coleman S. Cowart, Amanda Balmages et al.

Introduction: Acute aortic syndrome (AAS) includes the disease processes of aortic dissection, penetrating atherosclerotic ulcer, and intramural hematoma. This case demonstrates an atypical presentation of the disease and offers approaches to potentially prevent missed diagnoses. Case Report: An 87-year-old female with hypertension and Alzheimer’s dementia presented to the emergency department with stable vital signs and a chief complaint of throat pain. Initial work-up was significant for ischemia on electrocardiogram and elevated troponin. Computed tomography of the soft tissue neck revealed evidence of a ruptured aorta. Conclusion: Aortic rupture is a fatal complication of AAS. In an elderly patient with a history of hypertension, ischemic changes on electrocardiogram, and nonspecific pain, AAS should be on the emergency physician’s differential even in the setting of a benign or limited history and exam.

Menghan Yu, Sourabh Kulhare, Courosh Mehanian et al.

Acquiring large quantities of data and annotations is known to be effective for developing high-performing deep learning models, but is difficult and expensive to do in the healthcare context. Adding synthetic training data using generative models offers a low-cost method to deal effectively with the data scarcity challenge, and can also address data imbalance and patient privacy issues. In this study, we propose a comprehensive framework that fits seamlessly into model development workflows for medical image analysis. We demonstrate, with datasets of varying size, (i) the benefits of generative models as a data augmentation method; (ii) how adversarial methods can protect patient privacy via data substitution; (iii) novel performance metrics for these use cases by testing models on real holdout data. We show that training with both synthetic and real data outperforms training with real data alone, and that models trained solely with synthetic data approach their real-only counterparts. Code is available at https://github.com/Global-Health-Labs/US-DCGAN.

Hao Du, Qihua Dong, Yan Xu et al.

Medical image segmentation is almost the most important pre-processing procedure in computer-aided diagnosis but is also a very challenging task due to the complex shapes of segments and various artifacts caused by medical imaging, (i.e., low-contrast tissues, and non-homogenous textures). In this paper, we propose a simple yet effective segmentation framework that incorporates the geometric prior and contrastive similarity into the weakly-supervised segmentation framework in a loss-based fashion. The proposed geometric prior built on point cloud provides meticulous geometry to the weakly-supervised segmentation proposal, which serves as better supervision than the inherent property of the bounding-box annotation (i.e., height and width). Furthermore, we propose contrastive similarity to encourage organ pixels to gather around in the contrastive embedding space, which helps better distinguish low-contrast tissues. The proposed contrastive embedding space can make up for the poor representation of the conventionally-used gray space. Extensive experiments are conducted to verify the effectiveness and the robustness of the proposed weakly-supervised segmentation framework. The proposed framework is superior to state-of-the-art weakly-supervised methods on the following publicly accessible datasets: LiTS 2017 Challenge, KiTS 2021 Challenge, and LPBA40. We also dissect our method and evaluate the performance of each component.

Döndü Şanlıtürk

Louise Rose, Andreas Xyrichis, Joel Meyer

Adèle Rohée-Traoré, Jean-Daniel Kün-Darbois, Sophie Boucher

Mervat Alhaffar, Huda Basaleem, Fouad Othman et al.

Abstract Introduction Widespread armed conflict has affected Yemen since 2014. To date, the mortality toll of seven years of crisis, and any excess due to the COVID-19 pandemic, are not well quantified. We attempted to estimate population mortality during the pre-pandemic and pandemic periods in nine purposively selected urban and rural communities of southern and central Yemen (Aden and Ta’iz governorates), totalling > 100,000 people. Methods Within each study site, we collected lists of decedents between January 2014–March 2021 by interviewing different categories of key community informants, including community leaders, imams, healthcare workers, senior citizens and others. After linking records across lists based on key variables, we applied two-, three- or four-list capture-recapture analysis to estimate total death tolls. We also computed death rates by combining these estimates with population denominators, themselves subject to estimation. Results After interviewing 138 disproportionately (74.6%) male informants, we identified 2445 unique decedents. While informants recalled deaths throughout the study period, reported deaths among children were sparse: we thus restricted analysis to persons aged ≥ 15 years old. We noted a peak in reported deaths during May–July 2020, plausibly coinciding with the first COVID-19 wave. Death rate estimates featured uninformatively large confidence intervals, but appeared elevated compared to the non-crisis baseline, particularly in two sites where a large proportion of deaths were attributed to war injuries. There was no clear-cut evidence of excess mortality during the pandemic period. Conclusions We found some evidence of a peak in mortality during the early phase of the pandemic, but death rate estimates were otherwise too imprecise to enable strong inference on trends. Estimates suggested substantial mortality elevations from baseline during the crisis period, but are subject to serious potential biases. The study highlighted challenges of data collection in this insecure, politically contested environment.

Tarek Abdel hay Mostafa, Mohammed Osama Tommom, Naglaa Khalil Khalil

Abstract Background The main anesthetic goal in the postoperative period of functional endoscopic sinus surgery (FESS) is early and pain-free recovery, with return of protective airway reflex. The aim of this study is to evaluate the role of dexmedetomidine as an adjuvant to regional blocks in patients undergoing functional endoscopic sinus surgery. Results Group bupivacaine + dexmedetomidine (BD) showed statistically significantly lower postoperative pain measurements than group bupivacaine (B), longer time to the first request of rescue analgesia, less patients needing rescue analgesia, and less rescue morphine analgesic requirements. Other measurements were insignificantly different between both groups. Conclusions Dexmedetomidine can be used as an adjuvant to regional nerve block in patients undergoing functional endoscopic sinus surgery. It effectively prolonged postoperative analgesia, decreased postoperative opioid analgesic requirements, and reduced the number of patients needing rescue analgesia.