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

Mason Kadem, Sarah Masri, Anthea Innes et al.

We conducted a scoping review to map the rapidly evolving landscape of wearable and ambient sensing technologies for monitoring people with dementia across home and institutional settings. We analyzed empirical sensing studies (2015-2025) to identify and inform future technical and human-centered design requirements. Five key implementation principles emerge: (1) human-centered design involving all stakeholders to augment rather than replace caregivers; (2) personalized, adaptable solutions that support autonomy across settings and severity levels instead of standardized approaches; (3) integration with existing workflows with adequate training and support; (4) proactive privacy and consent considerations, especially for ambient monitoring of residents and caregivers; and (5) cost-effective, ethical, equitable, scalable solutions with quantifiable outcomes. This paper identifies gaps, trends and opportunities for developing sensing systems that address the complex challenges, while enhancing automation and autonomy, in dementia care.

Alice Zhong, Phoebe Chen, Anika Sharma et al.

Sexual and reproductive health (SRH) remains shaped by structural barriers that leave many without judgment-free information. AI chatbots offer anonymous alternatives, but access alone does not ensure equity when socioeconomic determinants shape whose capabilities these tools expand or constrain. Conventional methods for evaluating human-AI interaction were not designed to capture whether technologies holistically support reproductive autonomy. We introduce CARE, Capability Approach for Reproductive Equity, developing capabilities, functionings, and conversion factors into a Normative Design Lens and an Evaluation Lens for AI in SRH contexts. Evaluating SRH-specific non-LLM chatbots, general-use LLMs, and search engine features along credibility and reasoning, we identify two epistemic harms: source opacity and response rigidity. We conclude with design and evaluation recommendations, participatory auditing strategies, and policy implications for high-stakes domains where AI intersects with inequity.

Elahe Khatibi, Ziyu Wang, Ankita Sharma et al.

Large language models (LLMs) enable waveform-to-text ECG interpretation and interactive clinical questioning, yet most ECG-LLM systems still rely on weak signal-text alignment and retrieval without explicit physiological or causal structure. This limits grounding, temporal reasoning, and counterfactual "what-if" analysis central to clinical decision-making. We propose CARE-ECG, a causally structured ECG-language reasoning framework that unifies representation learning, diagnosis, and explanation in a single pipeline. CARE-ECG encodes multi-lead ECGs into temporally organized latent biomarkers, performs causal graph inference for probabilistic diagnosis, and supports counterfactual assessment via structural causal models. To improve faithfulness, CARE-ECG grounds language outputs through causal retrieval-augmented generation and a modular agentic pipeline that integrates history, diagnosis, and response with verification. Across multiple ECG benchmarks and expert QA settings, CARE-ECG improves diagnostic accuracy and explanation faithfulness while reducing hallucinations (e.g., 0.84 accuracy on Expert-ECG-QA and 0.76 on SCP-mapped PTB-XL under GPT-4). Overall, CARE-ECG provides traceable reasoning by exposing key latent drivers, causal evidence paths, and how alternative physiological states would change outcomes.

Adriana C. Mares, MS, MD, Rahul Gupta, MD, Bryan W. Kluck, DO

Ermira Muco, Najada Como, Ylber Vata et al.

Introduction: Fasciitis is an inflammation of the fascia, the connective tissue that surrounds muscles, blood vessels, and nerves, which can rapidly progress to necrosis of these structures, leading to life-threatening situations. It is a polymicrobial infection caused by bacterial agents, fungi, viruses, and parasites, resulting in septicemia, multiorgan failure, septic shock, and potentially death. The LRINEC score is a valuable tool in diagnosing necrotizing fasciitis (NF). Case Presentation: The patient is a 62-year-old woman with a history of Diabetes Mellitus type 2 and Bilinear Myelodysplasia, currently receiving high-dose corticosteroid therapy. She was recently diagnosed with Herpes Zoster affecting the S1, S2, S3, S4, and S5 dermatomes. Initially, she was treated for 8 days with antivirals and antibiotics for symptoms of fever, fatigue, and pain in the gluteal and lumbar regions. However, her condition worsened, prompting her readmission to the hospital, where she developed signs of sepsis. After imaging, laboratory tests, and consultations with surgeons, she was diagnosed with necrotizing fasciitis. Multidisciplinary management and complex therapy led to an excellent outcome. Conclusion: Necrotizing fasciitis presents a significant diagnostic challenge due to its rarity and its clinical resemblance to other critical conditions. It is an urgent pathology that requires not only surgical intervention but also immediate medical attention. It has always intrigued healthcare professionals both locally and internationally. Early diagnosis remains the primary challenge, and the management of these patients by experienced senior surgeons is crucial for successful outcomes.

Kerol Djoumessi, Philipp Berens

Convolutional neural networks (CNNs) are widely used for high-stakes applications like medicine, often surpassing human performance. However, most explanation methods rely on post-hoc attribution, approximating the decision-making process of already trained black-box models. These methods are often sensitive, unreliable, and fail to reflect true model reasoning, limiting their trustworthiness in critical applications. In this work, we introduce SoftCAM, a straightforward yet effective approach that makes standard CNN architectures inherently interpretable. By removing the global average pooling layer and replacing the fully connected classification layer with a convolution-based class evidence layer, SoftCAM preserves spatial information and produces explicit class activation maps that form the basis of the model's predictions. Evaluated on three medical datasets, SoftCAM maintains classification performance while significantly improving both the qualitative and quantitative explanation compared to existing post-hoc methods. Our results demonstrate that CNNs can be inherently interpretable without compromising performance, advancing the development of self-explainable deep learning for high-stakes decision-making. The code is available at https://github.com/kdjoumessi/SoftCAM

Heather M. Whitney, Hui Li, Karen Drukker et al.

Health disparities (differences in non-genetic conditions that influence health) can be associated with differences in burden of disease by groups within a population. Social determinants of health (SDOH) are domains such as health care access, dietary access, and economics frequently studied for potential association with health disparities. Evaluating SDOH-related phenotypes using routine medical images as data sources may enhance health disparities research. We developed a pipeline for using quantitative measures automatically extracted from medical images as inputs into health disparities index calculations. Our study focused on the use case of two SDOH demographic correlates (sex and race) and data extracted from chest radiographs of 1,571 unique patients. The likelihood of severe disease within the lung parenchyma from each image type, measured using an established deep learning model, was merged into a single numerical image-based phenotype for each patient. Patients were then separated into phenogroups by unsupervised clustering of the image-based phenotypes. The health rate for each phenogroup was defined as the median image-based phenotype for each SDOH used as inputs to four imaging-derived health disparities indices (iHDIs): one absolute measure (between-group variance) and three relative measures (index of disparity, Theil index, and mean log deviation). The iHDI measures demonstrated feasible values for each SDOH demographic correlate, showing potential for medical images to serve as a novel probe for health disparities. Large-scale AI analysis of medical images can serve as a probe for a novel data source for health disparities research.

Ifrat Ikhtear Uddin, Longwei Wang, KC Santosh

Medical image analysis often faces significant challenges due to limited expert-annotated data, hindering both model generalization and clinical adoption. We propose an expert-guided explainable few-shot learning framework that integrates radiologist-provided regions of interest (ROIs) into model training to simultaneously enhance classification performance and interpretability. Leveraging Grad-CAM for spatial attention supervision, we introduce an explanation loss based on Dice similarity to align model attention with diagnostically relevant regions during training. This explanation loss is jointly optimized with a standard prototypical network objective, encouraging the model to focus on clinically meaningful features even under limited data conditions. We evaluate our framework on two distinct datasets: BraTS (MRI) and VinDr-CXR (Chest X-ray), achieving significant accuracy improvements from 77.09% to 83.61% on BraTS and from 54.33% to 73.29% on VinDr-CXR compared to non-guided models. Grad-CAM visualizations further confirm that expert-guided training consistently aligns attention with diagnostic regions, improving both predictive reliability and clinical trustworthiness. Our findings demonstrate the effectiveness of incorporating expert-guided attention supervision to bridge the gap between performance and interpretability in few-shot medical image diagnosis.

V.Yu. Sadovyi, Yu.L. Kuchyn, K.Yu. Bielka et al.

Background. The endotracheal tube insertion, which is a veritable patient care procedure, carries risks and possible adverse effects. Such complications vary in severity and range from mild symptoms such as sore throat or cough, which are often considered normal post-intubation events, to severe damage, including damage to the vocal cords, trachea, malacia, and fistula formation. A critical factor influencing the manifestation of these complications is the pressure exerted by the endotracheal tube cuff on the trachea. The purpose was to study the influence of high pressure of the intubation tube cuff on the incidence and severity of complications in an in vivo model. Materials and methods. A total of 12 rabbits were studied and divided into three groups according to target cuff pressure: 30, 50, and 100 cm H2O. Sevoflurane was used to maintain anesthesia, monitoring of vital functions included heart rate, pulse oximetry, capnography, and electrocardiography. Intubation was performed after induction with ketamine and administration of relaxants, and cuff pressure was measured with a mechanical manometer. The duration of pressure exposure was 20 minutes for each group, then the degree of tracheal damage was assessed by autopsy. Results. Increasing pressure generally leads to an increased risk of injury. In the third group (pressure of 100 cm H2O), all subjects had damage to the trachea, in 75 % of cases, it was a complete tracheal rupture. However, the dependence on pressure is not unambiguous for all types of injuries: the frequency of pneumothorax does not correlate directionally with the incidence of iatrogenic injuries and does not increase proportionally to pressure. So, for example, subcutaneous emphysema developed only in a third of cases of tracheal rupture. This emphasizes possible difficulties in timely diagnosis of this type of complications. Conclusions. The work highlights the risks associated with endotracheal intubation and emphasizes the need to maintain recommended practices and individual approach to each case. The most severe damage was observed at a pressure of 100 cm H2O, while at a pressure of 30 cm H2O, a smaller proportion of rabbits showed signs of damage.

Amal H. Aljohani, Hamdi Alsufiani, Ghousia Ahmed

Abstract Background Spontaneous pneumomediastinum (SPM) and subcutaneous emphysema (SE) are rare, severe, and potentially life-threatening complications associated with asthma exacerbation. Most of these conditions are benign and self-limiting. However, the overlapping symptoms between asthma exacerbation and pneumomediastinum (PM) may delay diagnosis. These conditions can usually be managed through conservative treatment, although unfamiliarity with this presentation may lead some physicians to consider surgical intervention. Case presentation We report a unique case involving a 9-year-old patient experiencing a severe bronchial asthma attack and right lobe atelectasis complicated by PM and severe SE that extended to his left eye. The condition was successfully treated conservatively, with aggressive management of asthma exacerbation and close monitoring in the intensive care unit. Conclusion This case highlights the effectiveness of conservative management of PM and SE with appropriate asthma exacerbation treatment. Early diagnosis and management can lead to a favorable prognosis and a relatively brief hospital stay. Clinical trial number Not applicable.

G. A. Zabunyan, A. A. Martirosyan, A. G. Baryshev et al.

Objective: To present a case of reconstruction of a full-thickness lower eyelid defect.Case report: Male patient Y. aged 71 years was admitted to the Scientific Research Institute – Ochapovsky Regional Clinical Hospital No. 1 (Krasnodar, Russian Federation) with the diagnosis: cancer of skin of the lateral canthus of the left eye, T1N0M0, stage I, clinical group II. According to the medical history, the neoplasm was self-detected 5 years ago. Histology findings revealed micronodular basal cell carcinoma.After a general clinical examination, the patient underwent surgery to remove the neoplasm of the lower eyelid skin and lateral canthus and reconstruct the defect using a lateral infraorbital fat pad flap, local tissues, and a free autologous cartilage graft. The configuration of the soft tissues of the eyelid was restored. Soft tissue edema was not observed. The functional state of the lacrimal apparatus was fully preserved. The esthetics of the soft tissue was rehabilitated.Conclusions: Reconstruction in such esthetically significant area as the middle third of the face is associated with special requirements for graft formation and positioning. A gentle and functionally justified method for restoring the parameters of the tarsal plate, conjunctival sac, and lower eyelid skin ensures complete rehabilitation of patients and improves their quality of life.

Anvar Kurmukov, Bogdan Zavolovich, Aleksandra Dalechina et al.

Image compression is a critical tool in decreasing the cost of storage and improving the speed of transmission over the internet. While deep learning applications for natural images widely adopts the usage of lossy compression techniques, it is not widespread for 3D medical images. Using three CT datasets (17 tasks) and one MRI dataset (3 tasks) we demonstrate that lossy compression up to 20 times have no negative impact on segmentation quality with deep neural networks (DNN). In addition, we demonstrate the ability of DNN models trained on compressed data to predict on uncompressed data and vice versa with no quality deterioration.

Qiang Qiao, Wenyu Wang, Meixia Qu et al.

The field of medical image segmentation is challenged by domain generalization (DG) due to domain shifts in clinical datasets. The DG challenge is exacerbated by the scarcity of medical data and privacy concerns. Traditional single-source domain generalization (SSDG) methods primarily rely on stacking data augmentation techniques to minimize domain discrepancies. In this paper, we propose Random Amplitude Spectrum Synthesis (RASS) as a training augmentation for medical images. RASS enhances model generalization by simulating distribution changes from a frequency perspective. This strategy introduces variability by applying amplitude-dependent perturbations to ensure broad coverage of potential domain variations. Furthermore, we propose random mask shuffle and reconstruction components, which can enhance the ability of the backbone to process structural information and increase resilience intra- and cross-domain changes. The proposed Random Amplitude Spectrum Synthesis for Single-Source Domain Generalization (RAS^4DG) is validated on 3D fetal brain images and 2D fundus photography, and achieves an improved DG segmentation performance compared to other SSDG models.

Harriet Tucker, Karim Brohi, Joachim Tan et al.

Abstract Background In-hospital acute resuscitation in trauma has evolved toward early and balanced transfusion resuscitation with red blood cells (RBC) and plasma being transfused in equal ratios. Being able to deliver this ratio in prehospital environments is a challenge. A combined component, like leukocyte-depleted red cell and plasma (RCP), could facilitate early prehospital resuscitation with RBC and plasma, while at the same time improving logistics for the team. However, there is limited evidence on the clinical benefits of RCP. Objective To compare prehospital transfusion of combined RCP versus RBC alone or RBC and plasma separately (RBC + P) on mortality in trauma bleeding patients. Methods Data were collected prospectively on patients who received prehospital transfusion (RBC + thawed plasma/Lyoplas or RCP) for traumatic hemorrhage from six prehospital services in England (2018–2020). Retrospective data on patients who transfused RBC from 2015 to 2018 were included for comparison. The association between transfusion arms and 24-h and 30-day mortality, adjusting for age, injury mechanism, age, prehospital heart rate and blood pressure, was evaluated using generalized estimating equations. Results Out of 970 recruited patients, 909 fulfilled the study criteria (RBC + P = 391, RCP = 295, RBC = 223). RBC + P patients were older (mean age 42 vs 35 years for RCP and RBC), and 80% had a blunt injury (RCP = 52%, RBC = 56%). RCP and RBC + P were associated with lower odds of death at 24-h, compared to RBC alone (adjusted odds ratio [aOR] 0.69 [95%CI: 0.52; 0.92] and 0.60 [95%CI: 0.32; 1.13], respectively). The lower odds of death for RBC + P and RCP vs RBC were driven by penetrating injury (aOR 0.22 [95%CI: 0.10; 0.53] and 0.39 [95%CI: 0.20; 0.76], respectively). There was no association between RCP or RBC + P with 30-day survival vs RBC. Conclusion Prehospital plasma transfusion for penetrating injury was associated with lower odds of death at 24-h compared to RBC alone. Large trials are needed to confirm these findings.

Marzieh Momennasab, Zohreh Homayoon, Camellia Torabizadeh

Background. Adherence to ethical codes is a major pillar of nursing care that is affected by various factors. Identifying these factors can lead to better ethical performance. The present study was conducted to determine critical care nurses’ adherence to ethical codes and its association with spiritual well-being (SWB) and moral sensitivity (MS). Methods. In this descriptive-correlational study, data were collected using the moral sensitivity questionnaire (MSQ) by Lützén et al., Paloutzian and Ellison’s spiritual well-being scale (SWBS), and the adherence to ethical codes questionnaire. The study was conducted on 298 nurses working in critical care units of hospitals affiliated with Shiraz University of Medical Sciences in southern Iran in 2019. This study was examined and approved by the Ethics Committee of Shiraz University of Medical Sciences. Results. The majority of the participants were female (76.2%) and single (60.1%), with a mean age of 30.69 ± 5.74 years. The mean scores of adherence to ethical codes, SWB, and MS were 64.06 (good), 91.94 (moderate), and 134.08 (moderate), respectively. Adherence to ethical codes had a positive correlation with the total score of SWB (P<0.001, r = 0.25) and MS (P<0.001, r = 0.27). A positive correlation was also observed between MS and SWB (P<0.001, r = 0.41). Meanwhile, MS (β = 0.21) had a greater effect than SWB (β = 0.157) on adherence to ethical codes. Conclusion. Critical care nurses showed a good adherence to ethical codes. MS and SWB also positively affected their adherence to ethical codes. Nursing managers can use these findings to devise plans for the promotion of MS and SWB in nurses and thus help improve their ethical performance.

Catherine Willmott, Jonathan Reyes, Jack VK Nguyen et al.

Aim: To explore soft-shell padded headgear (HG) use, player behavior and injuries associated with HG in junior Australian football. Methods: Prospective case-crossover with head impact measurement,...

Nicolás Gaggion, Rodrigo Echeveste, Lucas Mansilla et al.

It has recently been shown that deep learning models for anatomical segmentation in medical images can exhibit biases against certain sub-populations defined in terms of protected attributes like sex or ethnicity. In this context, auditing fairness of deep segmentation models becomes crucial. However, such audit process generally requires access to ground-truth segmentation masks for the target population, which may not always be available, especially when going from development to deployment. Here we propose a new method to anticipate model biases in biomedical image segmentation in the absence of ground-truth annotations. Our unsupervised bias discovery method leverages the reverse classification accuracy framework to estimate segmentation quality. Through numerical experiments in synthetic and realistic scenarios we show how our method is able to successfully anticipate fairness issues in the absence of ground-truth labels, constituting a novel and valuable tool in this field.

Niketha Sabesan, Nivethitha, J. N Shreyah et al.

Web scraping is a technique that allows us to extract data from websites automatically. in the field of medicine, web scraping can be used to collect information about medical procedures, treatments, and healthcare providers. this information can be used to improve patient care, monitor the quality of healthcare services, and identify areas for improvement. one area where web scraping can be particularly useful is in medical ministrations. medical ministrations are the actions taken to provide medical care to patients, and web scraping can help healthcare providers identify the most effective ministrations for their patients. for example, healthcare providers can use web scraping to collect data about the symptoms and medical histories of their patients, and then use this information to determine the most appropriate ministrations. they can also use web scraping to gather information about the latest medical research and clinical trials, which can help them stay up-to-date with the latest treatments and procedures.

Franz Thaler, Matthias A. F. Gsell, Gernot Plank et al.

Late gadolinium enhanced (LGE) magnetic resonance (MR) imaging is widely established to assess the viability of myocardial tissue of patients after acute myocardial infarction (MI). We propose the Cascading Refinement CNN (CaRe-CNN), which is a fully 3D, end-to-end trained, 3-stage CNN cascade that exploits the hierarchical structure of such labeled cardiac data. Throughout the three stages of the cascade, the label definition changes and CaRe-CNN learns to gradually refine its intermediate predictions accordingly. Furthermore, to obtain more consistent qualitative predictions, we propose a series of post-processing steps that take anatomical constraints into account. Our CaRe-CNN was submitted to the FIMH 2023 MYOSAIQ challenge, where it ranked second out of 18 participating teams. CaRe-CNN showed great improvements most notably when segmenting the difficult but clinically most relevant myocardial infarct tissue (MIT) as well as microvascular obstructions (MVO). When computing the average scores over all labels, our method obtained the best score in eight out of ten metrics. Thus, accurate cardiac segmentation after acute MI via our CaRe-CNN allows generating patient-specific models of the heart serving as an important step towards personalized medicine.

Linda Lastrico, Nuno Ferreira Duarte, Alessandro Carfì et al.

Implicit communication plays such a crucial role during social exchanges that it must be considered for a good experience in human-robot interaction. This work addresses implicit communication associated with the detection of physical properties, transport, and manipulation of objects. We propose an ecological approach to infer object characteristics from subtle modulations of the natural kinematics occurring during human object manipulation. Similarly, we take inspiration from human strategies to shape robot movements to be communicative of the object properties while pursuing the action goals. In a realistic HRI scenario, participants handed over cups - filled with water or empty - to a robotic manipulator that sorted them. We implemented an online classifier to differentiate careful/not careful human movements, associated with the cups' content. We compared our proposed "expressive" controller, which modulates the movements according to the cup filling, against a neutral motion controller. Results show that human kinematics is adjusted during the task, as a function of the cup content, even in reach-to-grasp motion. Moreover, the carefulness during the handover of full cups can be reliably inferred online, well before action completion. Finally, although questionnaires did not reveal explicit preferences from participants, the expressive robot condition improved task efficiency.