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

Daniel Mejer Christensen, Katja Stougård Jørgensen, Josefine Palsgaard Wyrtz et al.

Research in Human-Computer Interaction (HCI) has shown that caring for others, including both humans (e.g., close friends) and computers (e.g., Tamagotchi), can have a positive effect on people's wellbeing. However, we know less about the potential role of conversational AI in such settings. In this work, we explore how AI chatbots can support plant care and, in turn, positively influence people's well-being. We developed a mobile application that allows users to `talk' to their plants via chatbots. We evaluated the application with ten participants and conducted semi-structured interviews based on Seligman's PERMA model, which identifies pillars of psychological well-being. Our findings suggest positive effects, with participants reflecting on a sense of connection to their plants and corresponding feelings of accomplishment. While our findings suggest that participants were generally positive about the app, they also raised concerns about the diverse preferences and expectations of users regarding interactions with chatbots representing plants.

Amir Khorram-Manesh

Abstract Background Disaster management is an inter-, intra-, and cross-disciplinary task in which different specialties partake. Triage is a crucial part of disaster education. A synchronized approach and mutual understanding of triaging and agreement on priorities are essential for saving lives. Case study Educational initiatives in disaster medicine aim to address issues that highlight the differences between more routine multi-casualty incidents and rarer mass casualty incidents. These differences are characterized by the number of victims, available resources, and environmental factors that may jeopardize the safety of victims and healthcare providers. While routine triage algorithms are often used in multiple casualty emergencies, considering environmental factors in mass casualty incidents caused by natural or human-made hazards should be equally important. Conclusions The impacts of environmental factors are usually not discussed in disaster medicine education, resulting in professionals having difficulties understanding the limitations of implementing routine triage algorithms during disaster response.

Philipp Dahlmann, Simone Böbel, Christian Elsenbast et al.

Waseem Nosair, MD, MPH, Liza Shaban, MD, Marc El Khoury, MD et al.

Background: Racial and ethnic disparities in catheter ablation (CA) utilization for atrial fibrillation have been reported, but inconsistent estimates complicate comparisons and understanding of root causes of disparity. Objectives: The purpose of this study was to quantify CA disparities by race/ethnicity, explore sources of heterogeneity, and offer recommendations for standardized definitions and methods to improve research. Methods: We systematically searched MEDLINE, Embase, Web of Science, and Cochrane’s CENTRAL from inception to January 15, 2024, for U.S.-based studies. Data were extracted on equity standards, cohort characteristics, methods, and risk of bias by 2 reviewers. We metaanalyzed cross-sectional and survival data separately using random effects models. We evaluated heterogeneity through qualitative synthesis, meta-regression, and sensitivity analyses. Results: Eighteen studies were included. None explicitly defined disparity or the source of race/ethnicity data. The most common analytic approach estimated disparity as the residual direct effect of race/ethnicity after adjusting for confounders, but key confounders were missing. Only one study evaluated mediators of disparity through sensitivity analysis. Compared to non-Hispanic White patients, non-Hispanic Black (OR: 0.65; 95% CI: 0.58-0.74), Hispanic/LatinX (OR: 0.78; 95% CI: 0.73-0.83), and Asian (OR: 0.74; 95% CI: 0.54-1.0) patients were less likely to receive CA. There was a high degree of between-study heterogeneity, likely from differences in source population, methods, and risk adjustment. Conclusions: While evidence confirms racial and ethnic disparities in CA utilization for atrial fibrillation, significant heterogeneity exists across studies. Standardized disparity definitions and consistent covariate adjustment may help confirm the scale of disparities and identify underlying mechanisms to inform interventions.

Saurabh Kumar, Sourav Bansal, Neeraj Agrawal et al.

Customer care is an essential pillar of the e-commerce shopping experience with companies spending millions of dollars each year, employing automation and human agents, across geographies (like US, Canada, Mexico, Chile), channels (like Chat, Interactive Voice Response (IVR)), and languages (like English, Spanish). SOTA pre-trained models like multilingual-BERT, fine-tuned on annotated data have shown good performance in downstream tasks relevant to Customer Care. However, model performance is largely subject to the availability of sufficient annotated domain-specific data. Cross-domain availability of data remains a bottleneck, thus building an intent classifier that generalizes across domains (defined by channel, geography, and language) with only a few annotations, is of great practical value. In this paper, we propose an embedder-cum-classifier model architecture which extends state-of-the-art domain-specific models to other domains with only a few labeled samples. We adopt a supervised fine-tuning approach with isotropic regularizers to train a domain-specific sentence embedder and a multilingual knowledge distillation strategy to generalize this embedder across multiple domains. The trained embedder, further augmented with a simple linear classifier can be deployed for new domains. Experiments on Canada and Mexico e-commerce Customer Care dataset with few-shot intent detection show an increase in accuracy by 20-23% against the existing state-of-the-art pre-trained models.

Erina Seh-Young Moon, Devansh Saxena, Dipto Das et al.

Homelessness systems in North America adopt coordinated data-driven approaches to efficiently match support services to clients based on their assessed needs and available resources. AI tools are increasingly being implemented to allocate resources, reduce costs and predict risks in this space. In this study, we conducted an ethnographic case study on the City of Toronto's homelessness system's data practices across different critical points. We show how the City's data practices offer standardized processes for client care but frontline workers also engage in heuristic decision-making in their work to navigate uncertainties, client resistance to sharing information, and resource constraints. From these findings, we show the temporality of client data which constrain the validity of predictive AI models. Additionally, we highlight how the City adopts an iterative and holistic client assessment approach which contrasts to commonly used risk assessment tools in homelessness, providing future directions to design holistic decision-making tools for homelessness.

Subham Kumar, Lekhansh Shukla, Animesh Mukherjee et al.

Determining the appropriate locus of care for addiction patients is one of the most critical clinical decisions that affects patient treatment outcomes and effective use of resources. With a lack of sufficient specialized treatment resources, such as inpatient beds or staff, there is an unmet need to develop an automated framework for the same. Current decision-making approaches suffer from severe class imbalances in addiction datasets. To address this limitation, we propose a novel graph neural network (GRACE) framework that formalizes locus of care prediction as a structured learning problem. In addition, we propose a new approach of obtaining an unbiased meta-graph to train a GNN to overcome the class imbalance problem. Experimental results with real-world data show an improvement of 11-35% in terms of the F1 score of the minority class over competitive baselines. Further, if we jointly finetune the base embedding fed into GRACE as input together with the rest of the GNN component of GRACE, there is a remarkable boost of 15.8% in performance.

Koji Ochiai, Yuya Tahara-Arai, Akari Kato et al.

The automation of experiments in life sciences and chemistry has significantly advanced with the development of various instruments and AI technologies. However, achieving full laboratory automation, where experiments conceived by scientists are seamlessly executed in automated laboratories, remains a challenge. We identify the lack of automation in planning and operational tasks--critical human-managed processes collectively termed "care"--as a major barrier. Automating care is the key enabler for full laboratory automation. To address this, we propose the concept of self-maintainability (SeM): the ability of a laboratory system to autonomously adapt to internal and external disturbances, maintaining operational readiness akin to living cells. A SeM-enabled laboratory features autonomous recognition of its state, dynamic resource and information management, and adaptive responses to unexpected conditions. This shifts the planning and execution of experimental workflows, including scheduling and reagent allocation, from humans to the system. We present a conceptual framework for implementing SeM-enabled laboratories, comprising three modules--Requirement manager, Labware manager, and Device manager--and a Central manager. SeM not only enables scientists to execute envisioned experiments seamlessly but also provides developers with a design concept that drives the technological innovations needed for full automation.

Smiljana Antonijevic Ubois

Large language models are commonly trained on dominant languages like English, and their representation of low resource languages typically reflects cultural and linguistic biases present in the source language materials. Using the Serbian language as a case, this study examines the structural, historical, and sociotechnical factors shaping language technology development for low resource languages in the AI age. Drawing on semi structured interviews with ten scholars and practitioners, including linguists, digital humanists, and AI developers, it traces challenges rooted in historical destruction of Serbian textual heritage, intensified by contemporary issues that drive reductive, engineering first approaches prioritizing functionality over linguistic nuance. These include superficial transliteration, reliance on English-trained models, data bias, and dataset curation lacking cultural specificity. To address these challenges, the study proposes Data Care, a framework grounded in CARE principles (Collective Benefit, Authority to Control, Responsibility, and Ethics), that reframes bias mitigation from a post hoc technical fix to an integral component of corpus design, annotation, and governance, and positions Data Care as a replicable model for building inclusive, sustainable, and culturally grounded language technologies in contexts where traditional LLM development reproduces existing power imbalances and cultural blind spots.

Neil K. R. Sehgal, Hita Kambhamettu, Allen Chang et al.

Effective communication in serious illness and palliative care is essential but often under-taught due to limited access to training resources like standardized patients. We present PAL (Palliative Assisted Learning-bot), a conversational system that simulates emotionally nuanced patient interactions and delivers structured feedback grounded in an existing empathy-based framework. PAL supports text and voice modalities and is designed to scaffold clinical skill-building through repeated, low-cost practice. Through a mixed-methods study with 17 U.S. medical trainees and clinicians, we explore user engagement with PAL, evaluate usability, and examine design tensions around modalities, emotional realism, and feedback delivery. Participants found PAL helpful for reflection and skill refinement, though some noted limitations in emotional authenticity and the adaptability of feedback. We contribute: (1) empirical evidence that large language models can support palliative communication training; (2) design insights for modality-aware, emotionally sensitive simulation tools; and (3) implications for systems that support emotional labor, cooperative learning, and AI-augmented training in high-stakes care settings.

Stanley Mugisha, Rashid Kisitu, Francis Komakech et al.

Preterm birth remains a leading cause of neonatal mortality, disproportionately affecting low-resource settings with limited access to advanced neonatal intensive care units (NICUs).Continuous monitoring of infant behavior, such as sleep/awake states and crying episodes, is critical but relies on manual observation or invasive sensors, which are prone to error, impractical, and can cause skin damage. This paper presents a novel, noninvasive, and automated vision-based framework to address this gap. We introduce an embedded monitoring system that utilizes a quantized MobileNet model deployed on a Raspberry Pi for real-time behavioral state detection. When trained and evaluated on public neonatal image datasets, our system achieves state-of-the-art accuracy (91.8% for sleep detection and 97.7% for crying/normal classification) while maintaining computational efficiency suitable for edge deployment. Through comparative benchmarking, we provide a critical analysis of the trade-offs between model size, inference latency, and diagnostic accuracy. Our findings demonstrate that while larger architectures (e.g., ResNet152, VGG19) offer marginal gains in accuracy, their computational cost is prohibitive for real-time edge use. The proposed framework integrates three key innovations: model quantization for memory-efficient inference (68% reduction in size), Raspberry Pi-optimized vision pipelines, and secure IoT communication for clinical alerts. This work conclusively shows that lightweight, optimized models such as the MobileNet offer the most viable foundation for scalable, low-cost, and clinically actionable NICU monitoring systems, paving the way for improved preterm care in resource-constrained environments.

Sujit J. Kshirsagar, Anandkumar H. Pande, Sanyogita V. Naik et al.

Background Optic nerve sheath diameter (ONSD) is an emerging non-invasive, easily accessible, and possibly useful measurement for evaluating changes in intracranial pressure (ICP). The utilization of bedside ultrasonography (USG) to measure ONSD has garnered increased attention due to its portability, real-time capability, and lack of ionizing radiation. The primary aim of the study was to assess whether bedside USG-guided ONSD measurement can reliably predict increased ICP in traumatic brain injury (TBI) patients. Methods A total of 95 patients admitted to the trauma intensive care unit was included in this cross sectional study. Patient brain computed tomography (CT) scans and Glasgow Coma Scale (GCS) scores were assessed at the time of admission. Bedside USG-guided binocular ONSD was measured and the mean ONSD was noted. Microsoft Excel was used for statistical analysis. Results Patients with low GCS had higher mean ONSD values (6.4±1.0 mm). A highly significant association was found among the GCS, CT results, and ONSD measurements (P<0.001). Compared to CT scans, the bedside USG ONSD had 86.42% sensitivity and 64.29% specificity for detecting elevated ICP. The positive predictive value of ONSD to identify elevated ICP was 93.33%, and its negative predictive value was 45.00%. ONSD measurement accuracy was 83.16%. Conclusions Increased ICP can be accurately predicted by bedside USG measurement of ONSD and can be a valuable adjunctive tool in the management of TBI patients.

Christel Sirocchi, Muhammad Suffian, Federico Sabbatini et al.

In clinical practice, decision-making relies heavily on established protocols, often formalised as rules. Concurrently, Machine Learning (ML) models, trained on clinical data, aspire to integrate into medical decision-making processes. However, despite the growing number of ML applications, their adoption into clinical practice remains limited. Two critical concerns arise, relevant to the notions of consistency and continuity of care: (a) accuracy - the ML model, albeit more accurate, might introduce errors that would not have occurred by applying the protocol; (b) interpretability - ML models operating as black boxes might make predictions based on relationships that contradict established clinical knowledge. In this context, the literature suggests using ML models integrating domain knowledge for improved accuracy and interpretability. However, there is a lack of appropriate metrics for comparing ML models with clinical rules in addressing these challenges. Accordingly, in this article, we first propose metrics to assess the accuracy of ML models with respect to the established protocol. Secondly, we propose an approach to measure the distance of explanations provided by two rule sets, with the goal of comparing the explanation similarity between clinical rule-based systems and rules extracted from ML models. The approach is validated on the Pima Indians Diabetes dataset by training two neural networks - one exclusively on data, and the other integrating a clinical protocol. Our findings demonstrate that the integrated ML model achieves comparable performance to that of a fully data-driven model while exhibiting superior accuracy relative to the clinical protocol, ensuring enhanced continuity of care. Furthermore, we show that our integrated model provides explanations for predictions that align more closely with the clinical protocol compared to the data-driven model.

Jonathan Weblin, Adam Harriman, Katrina Butler et al.

Keisuke Yoshida, Yuko Nakano, Satoki Inoue

Marceli Wac, Raul Santos-Rodriguez, Chris McWilliams et al.

Intensive care units (ICUs) are complex and data-rich environments. Data routinely collected in the ICUs provides tremendous opportunities for machine learning, but their use comes with significant challenges. Complex problems may require additional input from humans which can be provided through a process of data annotation. Annotation is a complex, time-consuming process that requires domain expertise and technical proficiency. Existing data annotation tools fail to provide an effective solution to this problem. In this study, we investigated clinicians' approach to the annotation task. We focused on establishing the characteristics of the annotation process in the context of clinical data and identifying differences in the annotation workflow between different staff roles. The overall goal was to elicit requirements for a software tool that could facilitate an effective and time-efficient data annotation. We conducted an experiment involving clinicians from the ICUs annotating printed sheets of data. The participants were observed during the task and their actions were analysed in the context of Norman's Interaction Cycle to establish the requirements for the digital tool. The annotation process followed a constant loop of annotation and evaluation, during which participants incrementally analysed and annotated the data. No distinguishable differences were identified between how different staff roles annotate data. We observed preferences towards different methods for applying annotation which varied between different participants and admissions. We established 11 requirements for the digital data annotation tool for the healthcare setting. We conducted a manual data annotation activity to establish the requirements for a digital data annotation tool, characterised the clinicians' approach to annotation and elicited 11 key requirements for effective data annotation software.

Xianghong Lai, Anmei Wang

Objective. To study the clinical efficacy of hot ironing of the Tianshu and Shangjuxu with moxa salt packet to prevent irinotecan (CPT-11)-induced delayed-onset diarrhea (IIDD). Methods. A randomized controlled study was conducted on a sample of 120 patients with advanced colorectal cancer who were hospitalized in our oncology department and treated with FOLFIRI chemotherapy regimen from February 2018 to July 2021. They were equally divided into study group (n = 60) and control group (n = 60) according to whether they were treated with hot ironing with moxa salt packs or not. The general conditions, occurrence of IIDD, occurrence of delayed chemotherapy due to IIDD, time of occurrence and duration of IIDD, Karnofsky performance score (KPS) score, occurrence of leukopenia, and myelosuppression were compared between the two groups. Result. The incidence of grade 1, 2, 3, and 4 diarrhea in the study group was 11.67% (7/60), 5.00% (3/60), 3.33% (2/60), and 0.00% (0/60), respectively, while the incidence of grade 1, 2, 3, and 4 diarrhea in the control group was 21.67% (13/60), 8.33% (5/60), 10.00% (6/60), and 3.33% (2/60). The incidence of severe diarrhea and total diarrhea in the study group was (3.33% and 20.00%) lower than that in the control group (13.33% and 43.33%) (P<0.05). The incidence of delayed chemotherapy was lower in the study group (8.33%) (1/12) than in the control group (23.08%) (6/26) but the difference between the groups was not statistically significant (P>0.05). The time to onset of IIDD in the study group (6.45 ± 1.53) days was comparable to that in the control group (6.40 ± 1.77 days) (P>0.05), but the duration of IIDD in the study group (3.25 ± 1.05 days) was shorter than that in the control group (5.70 ± 1.72 days) (P<0.05). After treatment, the incidence of KPS improvement, stabilization, and reduction in the study group was 38.33% (23/60), 51.67% (31/60), and 10.00% (6/60), respectively, the incidence of KPS improvement, stabilization, and reduction in the control group was 23.33% (14/60), 50.00% (30/60), and 26.67% (16/60), respectively, and the percentage of KPS reduction in the study group was less than that in the control group (P<0.05). During the observation period after treatment, the total incidence of leucopenia in the study group was 11.67% (7/60) which is lower than 31.67% (19/60) in the control group (P<0.05). During the observation period after treatment, the incidence of III°+°IV myelosuppression in the study group was 5.00% (3/60) which is lower than 25.00% (15/60) in the control group (P<0.05). Conclusion. The hot ironing with moxa salt packet on Tianshu and Shangjuxu was more effective in preventing IIDD, which could reduce the incidence and severity of IIDD, shorten the duration of diarrhea and significantly increase the quality of life of patients with no significant adverse effects.

Alexandre Egea, Claire Dupuis, Etienne de Montmollin et al.

Abstract Background Augmented renal clearance (ARC) remains poorly evaluated in ICU. The objective of this study is to provide a full description of ARC in ICU including prevalence, evolution profile, risk factors and outcomes. Methods This was a retrospective, single-center, observational study. All the patients older than 18 years admitted for the first time in Medical ICU, Bichat, University Hospital, APHP, France, between January 1, 2017, and November 31, 2020 and included into the Outcomerea database with an ICU length of stay longer than 72 h were included. Patients with chronic kidney disease were excluded. Glomerular filtration rate was estimated each day during ICU stay using the measured creatinine renal clearance (CrCl). Augmented renal clearance (ARC) was defined as a 24 h CrCl greater than 130 ml/min/m2. Results 312 patients were included, with a median age of 62.7 years [51.4; 71.8], 106(31.9%) had chronic cardiovascular disease. The main reason for admission was acute respiratory failure (184(59%)) and 196(62.8%) patients had SARS-COV2. The median value for SAPS II score was 32[24; 42.5]; 146(44%) and 154(46.4%) patients were under vasopressors and invasive mechanical ventilation, respectively. The overall prevalence of ARC was 24.6% with a peak prevalence on Day 5 of ICU stay. The risk factors for the occurrence of ARC were young age and absence of cardiovascular comorbidities. The persistence of ARC during more than 10% of the time spent in ICU was significantly associated with a lower risk of death at Day 30. Conclusion ARC is a frequent phenomenon in the ICU with an increased incidence during the first week of ICU stay. Further studies are needed to assess its impact on patient prognosis.

Joel Shor, Nick Johnston

Compression is essential to storing and transmitting medical videos, but the effect of compression on downstream medical tasks is often ignored. Furthermore, systems in practice rely on standard video codecs, which naively allocate bits between medically relevant frames or parts of frames. In this work, we present an empirical study of some deficiencies of classical codecs on gastroenterology videos, and motivate our ongoing work to train a learned compression model for colonoscopy videos. We show that two of the most common classical codecs, H264 and HEVC, compress medically relevant frames statistically significantly worse than medically nonrelevant ones, and that polyp detector performance degrades rapidly as compression increases. We explain how a learned compressor could allocate bits to important regions and allow detection performance to degrade more gracefully. Many of our proposed techniques generalize to medical video domains beyond gastroenterology

Viet-Khoa Vo-Ho, Kashu Yamazaki, Hieu Hoang et al.

Deep learning methods have been successful in solving tasks in machine learning and have made breakthroughs in many sectors owing to their ability to automatically extract features from unstructured data. However, their performance relies on manual trial-and-error processes for selecting an appropriate network architecture, hyperparameters for training, and pre-/post-procedures. Even though it has been shown that network architecture plays a critical role in learning feature representation feature from data and the final performance, searching for the best network architecture is computationally intensive and heavily relies on researchers' experience. Automated machine learning (AutoML) and its advanced techniques i.e. Neural Architecture Search (NAS) have been promoted to address those limitations. Not only in general computer vision tasks, but NAS has also motivated various applications in multiple areas including medical imaging. In medical imaging, NAS has significant progress in improving the accuracy of image classification, segmentation, reconstruction, and more. However, NAS requires the availability of large annotated data, considerable computation resources, and pre-defined tasks. To address such limitations, meta-learning has been adopted in the scenarios of few-shot learning and multiple tasks. In this book chapter, we first present a brief review of NAS by discussing well-known approaches in search space, search strategy, and evaluation strategy. We then introduce various NAS approaches in medical imaging with different applications such as classification, segmentation, detection, reconstruction, etc. Meta-learning in NAS for few-shot learning and multiple tasks is then explained. Finally, we describe several open problems in NAS.