Evan S. Manning, MD, MPP, Gautam R. Shroff, MD, David R. Jacobs, Jr., PhD
et al.
Background: Inflammation plays a role in cardiovascular disease (CVD). We defined various noncardiovascular and noncancer conditions, both infectious and noninfectious, with a common basis of inflammation, collectively termed chronic inflammatory-related disease (ChrIRD). We describe ChrIRD and its interplay with CVD during follow-up in the Multi-Ethnic Study of Atherosclerosis. Objectives: The aim of the study was to describe ChrIRD, its associations with CVD, and its association with mortality. Methods: Participants were free of overt CVD at baseline with median 17.9 (Q1-Q3: 14.9-18.6) years of follow-up. ChrIRD was determined by review of hospitalization and death records of International Classification of Diseases codes. CVD diagnosis was adjudicated based on medical records. We performed time-dependent proportional hazard regressions to identify risks related to ChrIRD or CVD events. Results: MESA (Multi-Ethnic Study of Atherosclerosis) participants (n = 6,791) had a mean age of 62 ± 10 years, with 47% (3,201/6,791) men, 39% (2,617/6,791) White, 28% (1,882/6,791) Black, 22% (1,489/6,791) Hispanic, and 12% (803/6,791) Chinese race/ethnicity. ChrIRD was observed in 29% (1,965/6,791) and CVD in 21% (1,420/6,791); including 11% (761/6,791) with both conditions. Mortality after ChrIRD only was 47% (567/1,204; 95% CI: 44%-49%); after CVD only was 45% (300/659; 95% CI: 41%-49%); and after both conditions was 67% (510/761; 95% CI: 63%-70%). CVD was associated with increased risk of ChrIRD (HR: 1.48, 1.23-1.77) and ChrIRD was associated with increased risk of CVD (HR: 2.23, 1.97-2.52). Baseline inflammatory markers predicted both conditions. Conclusions: ChrIRD is common, present in all organ systems, and is associated with significant mortality, particularly in combination with CVD. The association between CVD and ChrIRD is bidirectional, and baseline inflammatory markers are associated with ChrIRD and CVD.
Diseases of the circulatory (Cardiovascular) system, Medical emergencies. Critical care. Intensive care. First aid
Malte Londschien, Manuel Burger, Gunnar Rätsch
et al.
The performance of predictive models in clinical settings often degrades when deployed in new hospitals due to distribution shifts. This paper presents a large-scale study of causality-inspired domain generalization on heterogeneous multi-center intensive care unit (ICU) data. We apply anchor regression and introduce anchor boosting, a novel, tree-based nonlinear extension, to a large dataset comprising 400,000 patients from nine distinct ICU databases. We find that anchor regularization yields improvements of out-of-distribution performance, particularly for the most dissimilar target domains. The methods appear robust to violations of theoretical assumptions, such as anchor exogeneity. Furthermore, we propose a novel conceptual framework to quantify the utility of large external data datasets. By evaluating performance as a function of available target-domain data, we identify three regimes: (i) a domain generalization regime, where only the external model should be used, (ii) a domain adaptation regime, where refitting the external model is optimal, and (iii) a data-rich regime, where external data provides no additional value.
In the intensive care unit, the underlying causes of critical illness vary substantially across diagnoses, yet prediction models accounting for diagnostic heterogeneity have not been systematically studied. To address the gap, we evaluate transfer learning approaches for diagnosis-specific mortality prediction and apply both GLM- and XGBoost-based models to the eICU Collaborative Research Database. Our results demonstrate that transfer learning consistently outperforms models trained only on diagnosis-specific data and those using a well-known ICU severity-of-illness score, i.e., APACHE IVa, alone, while also achieving better calibration than models trained on the pooled data. Our findings also suggest that the Youden cutoff is a more appropriate decision threshold than the conventional 0.5 for binary outcomes, and that transfer learning maintains consistently high predictive performance across various cutoff criteria.
Renata de Souza Mendes, Pedro Leme Silva, Chiara Robba
et al.
Abstract This narrative review delves into the intricate interplay between the lungs and the kidneys, with a focus on elucidating the pathogenesis of diseases influenced by immunological factors, acid–base regulation, and blood gas disturbances, as well as assessing the effects of various therapeutic modalities on these interactions. Key disorders, such as anti-glomerular basement membrane (anti-GBM) disease, the syndrome of inappropriate antidiuretic hormone secretion (SIADH), and Anti-neutrophil Cytoplasmic Antibodies (ANCA) associated vasculitis (AAV), are also examined to shed light on their underlying mechanisms. This review also explores the relationship between acute respiratory distress syndrome (ARDS) and acute kidney injury (AKI), emphasizing how inflammatory mediators can lead to systemic damage and impact multiple organs. In ARDS, fluid overload exacerbates pulmonary edema, while imbalances in blood volume, such as hypovolemia or hypervolemia, can precipitate renal dysfunction. The review highlights how mechanical ventilation strategies can compromise renal blood flow, trigger systemic inflammation, and induce hemodynamic and neurohormonal alterations, all contributing to lung and kidney damage. The impact of extracorporeal membrane oxygenation (ECMO) on lung–kidney interactions is evaluated, highlighting its role in severe respiratory failure and its renal implications. Emerging therapies, such as mesenchymal stem cells and extracellular vesicles, are discussed as promising avenues to mitigate organ damage and enhance outcomes in critically ill patients. Overall, this review offers a nuanced exploration of lung–kidney dynamics, bridging historical insights with contemporary perspectives. It underscores the clinical significance of these interactions in critically ill patients and advocates for integrated management approaches to optimize patient outcomes.
Medical emergencies. Critical care. Intensive care. First aid
Artificial intelligence has significantly impacted medical applications, particularly with the advent of Medical Large Vision Language Models (Med-LVLMs), sparking optimism for the future of automated and personalized healthcare. However, the trustworthiness of Med-LVLMs remains unverified, posing significant risks for future model deployment. In this paper, we introduce CARES and aim to comprehensively evaluate the Trustworthiness of Med-LVLMs across the medical domain. We assess the trustworthiness of Med-LVLMs across five dimensions, including trustfulness, fairness, safety, privacy, and robustness. CARES comprises about 41K question-answer pairs in both closed and open-ended formats, covering 16 medical image modalities and 27 anatomical regions. Our analysis reveals that the models consistently exhibit concerns regarding trustworthiness, often displaying factual inaccuracies and failing to maintain fairness across different demographic groups. Furthermore, they are vulnerable to attacks and demonstrate a lack of privacy awareness. We publicly release our benchmark and code in https://cares-ai.github.io/.
Disparities in access to healthcare have been well-documented in the United States, but their effects on electronic health record (EHR) data reliability and resulting clinical models are poorly understood. Using an All of Us dataset of 134,513 participants, we investigate the effects of access to care on the medical machine learning pipeline, including medical condition rates, data quality, outcome label accuracy, and prediction performance. Our findings reveal that patients with cost constrained or delayed care have worse EHR reliability as measured by patient self-reported conditions for 78% of examined medical conditions. We demonstrate in a prediction task of Type II diabetes incidence that clinical risk predictive performance can be worse for patients without standard care, with balanced accuracy gaps of 3.6 and sensitivity gaps of 9.4 percentage points for those with cost-constrained or delayed care. We evaluate solutions to mitigate these disparities and find that including patient self-reported conditions improved performance for patients with lower access to care, with 11.2 percentage points higher sensitivity, effectively decreasing the performance gap between standard versus delayed or cost-constrained care. These findings provide the first large-scale evidence that healthcare access systematically affects both data reliability and clinical prediction performance. By revealing how access barriers propagate through the medical machine learning pipeline, our work suggests that improving model equity requires addressing both data collection biases and algorithmic limitations. More broadly, this analysis provides an empirical foundation for developing clinical prediction systems that work effectively for all patients, regardless of their access to care.
Recently, the state space model Mamba has demonstrated efficient long-sequence modeling capabilities, particularly for addressing long-sequence visual tasks in 3D medical imaging. However, existing generative self-supervised learning methods have not yet fully unleashed Mamba's potential for handling long-range dependencies because they overlook the inherent causal properties of state space sequences in masked modeling. To address this challenge, we propose a general-purpose pre-training framework called MambaMIM, a masked image modeling method based on a novel TOKen-Interpolation strategy (TOKI) for the selective structure state space sequence, which learns causal relationships of state space within the masked sequence. Further, MambaMIM introduces a bottom-up 3D hybrid masking strategy to maintain a masking consistency across different architectures and can be used on any single or hybrid Mamba architecture to enhance its multi-scale and long-range representation capability. We pre-train MambaMIM on a large-scale dataset of 6.8K CT scans and evaluate its performance across eight public medical segmentation benchmarks. Extensive downstream experiments reveal the feasibility and advancement of using Mamba for medical image pre-training. In particular, when we apply the MambaMIM to a customized architecture that hybridizes MedNeXt and Vision Mamba, we consistently obtain the state-of-the-art segmentation performance. The code is available at: https://github.com/FengheTan9/MambaMIM.
Case presentation: An 84-year-old man presented to the emergency department with sudden, left lower quadrant cramping pain. Because critical hypotension was noted, point-of-care ultrasonography (POCUS) was performed immediately. The study revealed a pulsatile flow extravasating from the left common iliac artery into the left psoas muscle with hypoechoic para-aortic fluid collection. Discussion: Common iliac artery rupture is rare and has nonspecific clinical presentations. A quick disposition can be made with a combination of clinical manifestations and POCUS results.
Medical emergencies. Critical care. Intensive care. First aid
Scarcity of health care resources could result in the unavoidable consequence of rationing. For example, ventilators are often limited in supply, especially during public health emergencies or in resource-constrained health care settings, such as amid the pandemic of COVID-19. Currently, there is no universally accepted standard for health care resource allocation protocols, resulting in different governments prioritizing patients based on various criteria and heuristic-based protocols. In this study, we investigate the use of reinforcement learning for critical care resource allocation policy optimization to fairly and effectively ration resources. We propose a transformer-based deep Q-network to integrate the disease progression of individual patients and the interaction effects among patients during the critical care resource allocation. We aim to improve both fairness of allocation and overall patient outcomes. Our experiments demonstrate that our method significantly reduces excess deaths and achieves a more equitable distribution under different levels of ventilator shortage, when compared to existing severity-based and comorbidity-based methods in use by different governments. Our source code is included in the supplement and will be released on Github upon publication.
E. Vasilevskis, R. Chandrasekhar, C. Holtze
et al.
Rationale: Intensive care unit (ICU) delirium is highly prevalent and a potentially avoidable hospital complication. The current cost of ICU delirium is unknown. Objectives: To specify the association between the daily occurrence of delirium in the ICU with costs of ICU care accounting for time-varying illness severity and death. Research Design: We performed a prospective cohort study within medical and surgical ICUs in a large academic medical center. Subjects: We analyzed critically ill patients (N=479) with respiratory failure and/or shock. Measures: Covariates included baseline factors (age, insurance, cognitive impairment, comorbidities, Acute Physiology and Chronic Health Evaluation II Score) and time-varying factors (sequential organ failure assessment score, mechanical ventilation, and severe sepsis). The primary analysis used a novel 3-stage regression method: first, estimation of the cumulative cost of delirium over 30 ICU days and then costs separated into those attributable to increased resource utilization among survivors and those that were avoided on the account of delirium’s association with early mortality in the ICU. Results: The patient-level 30-day cumulative cost of ICU delirium attributable to increased resource utilization was $17,838 (95% confidence interval, $11,132–$23,497). A combination of professional, dialysis, and bed costs accounted for the largest percentage of the incremental costs associated with ICU delirium. The 30-day cumulative incremental costs of ICU delirium that were avoided due to delirium-associated early mortality was $4654 (95% confidence interval, $2056–7869). Conclusions: Delirium is associated with substantial costs after accounting for time-varying illness severity and could be 20% higher (∼$22,500) if not for its association with early ICU mortality.
B. M. Munasinghe, U. P. M. Fernando, Thileep Kumar
et al.
Abstract Background The spleen is one of the most frequently injured abdominal organs during trauma, which can result in intraperitoneal bleeding of life-threatening magnitude. Although splenic injury secondary to trivial trauma comprises a minor fraction of abdominal injuries, undiagnosed or delayed diagnosis may result in a complicated clinical course. Case presentation One such event is presented here, wherein a late diagnosis of an advanced grade splenic injury following a trivial trauma initially presented in disguise as acute myocardial ischaemia in a previously healthy South Asian woman in her late 30s. Emergency laparotomy and splenectomy were performed with simultaneous massive transfusion for a 3.5-L blood loss. She subsequently had an uncomplicated clinical course with regular surgical follow-up. Conclusion Splenic injuries might present with ambiguous symptoms such as atypical chest pain and shoulder pain, necessitating attending clinicians to have a high degree of suspicion, especially in busy units such as the emergency department (ED).
Medical emergencies. Critical care. Intensive care. First aid
The Critical Assessment of Genome Interpretation Consortium
The Critical Assessment of Genome Interpretation (CAGI) aims to advance the state of the art for computational prediction of genetic variant impact, particularly those relevant to disease. The five complete editions of the CAGI community experiment comprised 50 challenges, in which participants made blind predictions of phenotypes from genetic data, and these were evaluated by independent assessors. Overall, results show that while current methods are imperfect, they have major utility for research and clinical applications. Missense variant interpretation methods are able to estimate biochemical effects with increasing accuracy. Performance is particularly strong for clinical pathogenic variants, including some difficult-to-diagnose cases, and extends to interpretation of cancer-related variants. Assessment of methods for regulatory variants and complex trait disease risk is less definitive, and indicates performance potentially suitable for auxiliary use in the clinic. Emerging methods and increasingly large, robust datasets for training and assessment promise further progress ahead.
The COVID-19 pandemic has posed a heavy burden to the healthcare system worldwide and caused huge social disruption and economic loss. Many deep learning models have been proposed to conduct clinical predictive tasks such as mortality prediction for COVID-19 patients in intensive care units using Electronic Health Record (EHR) data. Despite their initial success in certain clinical applications, there is currently a lack of benchmarking results to achieve a fair comparison so that we can select the optimal model for clinical use. Furthermore, there is a discrepancy between the formulation of traditional prediction tasks and real-world clinical practice in intensive care. To fill these gaps, we propose two clinical prediction tasks, Outcome-specific length-of-stay prediction and Early mortality prediction for COVID-19 patients in intensive care units. The two tasks are adapted from the naive length-of-stay and mortality prediction tasks to accommodate the clinical practice for COVID-19 patients. We propose fair, detailed, open-source data-preprocessing pipelines and evaluate 17 state-of-the-art predictive models on two tasks, including 5 machine learning models, 6 basic deep learning models and 6 deep learning predictive models specifically designed for EHR data. We provide benchmarking results using data from two real-world COVID-19 EHR datasets. One dataset is publicly available without needing any inquiry and another dataset can be accessed on request. We provide fair, reproducible benchmarking results for two tasks. We deploy all experiment results and models on an online platform. We also allow clinicians and researchers to upload their data to the platform and get quick prediction results using our trained models. We hope our efforts can further facilitate deep learning and machine learning research for COVID-19 predictive modeling.