Explore the vast range of titles available.

MIMIC-III (‘Medical Information Mart for Intensive Care’) is a large, single-center database comprising information relating to patients admitted to critical care units at a large tertiary care hospital. Data includes vital signs, medications, laboratory measurements, observations and notes charted by care providers, fluid balance, procedure codes, diagnostic codes, imaging reports, hospital length of stay, survival data, and more. The database supports applications including academic and industrial research, quality improvement initiatives, and higher education coursework. Design Type(s) data integration objective Measurement Type(s) Demographics • clinical measurement • intervention • Billing • Medical History Dictionary • Pharmacotherapy • clinical laboratory test • medical data Technology Type(s) Electronic Medical Record • Medical Record • Electronic Billing System • Medical Coding Process Document • Free Text Format Factor Type(s) Sample Characteristic(s) Homo sapiens Machine-accessible metadata file describing the reported data (ISA-Tab format)

PurposeThe new challenge in Artificial Intelligence (AI) is to understand the limitations of models to reduce potential harm. Particularly, unknown disparities based on demographic factors could encrypt currently existing inequalities worsening patient care for some groups.MethodsFollowing PRISMA guidelines, we present a systematic review of ‘fair’ deep learning modeling techniques for natural and medical image applications which were published between year 2011 to 2021. Our search used Covidence review management software and incorporates articles from PubMed, IEEE, and ACM search engines and three reviewers independently review the manuscripts.ResultsInter-rater agreement was 0.89 and conflicts were resolved by obtaining consensus between three reviewers. Our search initially retrieved 692 studies but after careful screening, our review included 22 manuscripts that carried four prevailing themes; ‘fair’ training dataset generation (4/22), representation learning (10/22), model disparity across institutions (5/22) and model fairness with respect to patient demographics (3/22). We benchmark the current literature regarding fairness in AI-based image analysis and highlighted the existing challenges. We observe that often discussion regarding fairness are limited to analyzing existing bias without further establishing methodologies to overcome model disparities.ConclusionBased on the current research trends, exploration of adversarial learning for demographic/camera/institution agnostic models is an important direction to minimize disparity gaps for imaging. Privacy preserving approaches also present encouraging performance for both natural and medical image domain.

Artificial intelligence (AI), specifically machine learning (ML), is increasingly applied in decision-making for dental diagnosis, prognosis, and treatment. However, the methodological completeness of published models has not been rigorously assessed. We performed a scoping review of PubMed-indexed articles (English, 1 January 2018‒31 December 2023) that used ML in any dental specialty. Each study was evaluated with the TRIPOD + AI rubric for key reporting elements such as data preprocessing, model validation, and clinical performance. Out of 1,506 identified studies, 280 met the inclusion criteria. Oral and maxillofacial radiology (27.5%), oral and maxillofacial surgery (15.0%), and general dentistry (14.3%) were the most represented specialties. Sixty-four studies (22.9%) lacked comparison with a clinical reference standard or existing model performing the same task. Most models focused on classification (59.6%), whereas generative applications were relatively rare (1.4%). Key gaps included limited assessment of model bias, poor outlier reporting, scarce calibration evaluation, low reproducibility, and restricted data access. ML could transform dental care, but robust calibration assessment and equity evaluation are critical for real-world adoption. Future research should prioritize error explainability, outlier reporting, reproducibility, fairness, and prospective validation.

There is an increase in admission rate for elderly patients to the ICU. Mortality rates are lower when more liberal ICU admission threshold are compared to more restrictive threshold. We sought to describe the temporal trends in elderly admissions and outcomes in a tertiary hospital before and after the addition of an 8-bed medical ICU. We conducted a retrospective analysis of a comprehensive longitudinal ICU database, from a large tertiary medical center, examining trends in patients' characteristics, severity of illness, intensity of care and mortality rates over the years 2001-2008. The study population consisted of elderly patients and the primary endpoints were 28 day and one year mortality from ICU admission. Between the years 2001 and 2008, 7,265 elderly patients had 8,916 admissions to ICU. The rate of admission to the ICU increased by 5.6% per year. After an eight bed MICU was added, the severity of disease on ICU admission dropped significantly and crude mortality rates decreased thereafter. Adjusting for severity of disease on presentation, there was a decreased mortality at 28- days but no improvement in one- year survival rates for elderly patient admitted to the ICU over the years of observation. Hospital mortality rates have been unchanged from 2001 through 2008. In a high capacity ICU bed hospital, there was a temporal decrease in severity of disease on ICU admission, more so after the addition of additional medical ICU beds. While crude mortality rates decreased over the study period, adjusted one-year survival in ICU survivors did not change with the addition of ICU beds. These findings suggest that outcome in critically ill elderly patients may not be influenced by ICU admission. Adding additional ICU beds to deal with the increasing age of the population may therefore not be effective.

Artificial intelligence (AI) algorithms have the potential to revolutionize healthcare, but their successful translation into clinical practice has been limited. One crucial factor is the data used to train these algorithms, which must be representative of the population. However, most healthcare databases are derived from high-income countries, leading to non-representative models and potentially exacerbating health inequities. This review focuses on the landscape of health-related open datasets in Latin America, aiming to identify existing datasets, examine data-sharing frameworks, techniques, platforms, and formats, and identify best practices in Latin America. The review found 61 datasets from 23 countries, with the DATASUS dataset from Brazil contributing to the majority of articles. The analysis revealed a dearth of datasets created by the authors themselves, indicating a reliance on existing open datasets. The findings underscore the importance of promoting open data in Latin America. We provide recommendations for enhancing data sharing in the region.

Healthcare has long struggled to improve services through technology without further widening health disparities. With the significant expansion of digital health, a group of healthcare professionals and scholars from across the globe are proposing the official usage of the term “Digital Determinants of Health” (DDOH) to explicitly call out the relationship between technology, healthcare, and equity. This is the final paper in a series published in PLOS Digital Health that seeks to understand and summarize current knowledge of the strategies and solutions that help to mitigate the negative effects of DDOH for underinvested communities. Through a search of English-language Medline, Scopus, and Google Scholar articles published since 2010, 345 articles were identified that discussed the application of digital health technology among underinvested communities. A group of 8 reviewers assessed 132 articles selected at random for the mention of solutions that minimize differences in DDOH. Solutions were then organized by categories of policy; design and development; implementation and adoption; and evaluation and ongoing monitoring. The data were then assessed by category and the findings summarized. The reviewers also looked for common themes across the solutions and evidence of effectiveness. From this limited scoping review, the authors found numerous solutions mentioned across the papers for addressing DDOH and many common themes emerged regardless of the specific community or digital health technology under review. There was notably less information on solutions regarding ongoing evaluation and monitoring which corresponded with a lack of research evidence regarding effectiveness. The findings directionally suggest that universal strategies and solutions can be developed to address DDOH independent of the specific community under focus. With the need for the further development of DDOH measures, we also provide a framework for DDOH assessment.

There have been suggestions that patients with subarachnoid hemorrhage (SAH) have a better outcome when treated in high-volume centers. Much of the published literature on the subject is limited by an inability to control for severity of SAH. This is a nationwide retrospective cohort analysis using the Nationwide Inpatient Sample (NIS). The NIS Subarachnoid Severity Scale was used to adjust for severity of SAH in multivariate logistic regression modeling. The records of 47 911 414 hospital admissions from the 2006-2011 NIS samples were examined. There were 11 607 patients who met inclusion criteria for the study. Of these, 7787 (67.0%) were treated at a high-volume center compared with 3820 (32.9%) treated at a low-volume center. Patients treated at high-volume centers compared with low-volume centers were more likely to receive endovascular aneurysm control (58.5% vs 51.2%, P=.04), be transferred from another hospital (35.4% vs 19.7%, P<.01), be treated in a teaching facility (97.3% vs 72.9%, P<.01), and have a longer length of stay (14.9 days [interquartile range 10.3-21.7] vs 13.9 days [interquartile range, 8.9-20.1], P<.01). After adjustment for all baseline covariates, including severity of SAH, treatment in a high-volume center was associated with an odds ratio for death of 0.82 (95% confidence interval, 0.72-0.95; P<.01) and a higher odds of a good functional outcome (odds ratio, 1.16; 95% confidence interval, 1.04-1.28; P<.01). After adjustment for severity of SAH, treatment in a high-volume center was associated with a lower risk of in-hospital mortality and a higher odds of a good functional outcome.

Decision support systems based on reinforcement learning (RL) have been implemented to facilitate the delivery of personalized care. This paper aimed to provide a comprehensive review of RL applications in the critical care setting. This review aimed to survey the literature on RL applications for clinical decision support in critical care and to provide insight into the challenges of applying various RL models. We performed an extensive search of the following databases: PubMed, Google Scholar, Institute of Electrical and Electronics Engineers (IEEE), ScienceDirect, Web of Science, Medical Literature Analysis and Retrieval System Online (MEDLINE), and Excerpta Medica Database (EMBASE). Studies published over the past 10 years (2010-2019) that have applied RL for critical care were included. We included 21 papers and found that RL has been used to optimize the choice of medications, drug dosing, and timing of interventions and to target personalized laboratory values. We further compared and contrasted the design of the RL models and the evaluation metrics for each application. RL has great potential for enhancing decision making in critical care. Challenges regarding RL system design, evaluation metrics, and model choice exist. More importantly, further work is required to validate RL in authentic clinical environments.

Contemporary medical AI systems exhibit a critical vulnerability: they deliver confident predictions without mechanisms to express uncertainty or acknowledge limitations, leading to dangerous overreliance in clinical settings. This paper introduces the BODHI (Bridging, Open, Discerning, Humble, Inquiring) framework, a dual-reflective architecture grounded in two essential epistemic virtues: curiosity and humility, as foundational design principles for healthcare AI. Curiosity drives systems to actively explore diagnostic uncertainty, seek additional information when faced with ambiguous presentations, and recognize when training distributions fail to match clinical reality. Humility provides complementary restraint, enabling uncertainty quantification, boundary recognition, and appropriate deference to human expertise. We demonstrate how these virtues function synergistically in a dynamic feedback loop, preventing both reckless exploration and excessive caution while supporting collaborative clinical decision-making. Drawing from psychological theories of curiosity and cross-species evidence of epistemic humility, we argue that these capacities represent fundamental biological design principles essential for systems operating in high-stakes, uncertain environments. The BODHI framework addresses systemic failures in medical AI deployment, from biased training data to institutional workflow pressures, by embedding uncertainty awareness and collaborative restraint into foundational system architecture. Key implementation features include calibrated confidence measures, out-of-distribution detection, curiosity-driven escalation protocols, and transparency mechanisms that adapt to clinical context. Rather than pursuing algorithmic perfection through pure optimization, we advocate for human-AI partnerships that enhance clinical reasoning through mutual accountability and calibrated trust. This approach represents a paradigm shift from overconfident automation toward collaborative systems that embody the wisdom to pause, reflect, and defer when appropriate.

Serum sodium derangement is the most common electrolyte disturbance among patients admitted to intensive care. This study aims to validate the association between dysnatremia and serum sodium fluctuation with mortality in surgical intensive care patients. We performed a retrospective analysis of the Medical Information Mart for Intensive Care II database. Dysnatremia was defined as a sodium concentration outside physiologic range (135-145mmol/L) and subjects were categorized by severity of dysnatremia and sodium fluctuation. Univariate and multivariable logistic regressions were used to test for associations between sodium fluctuations and mortality. We identified 8600 subjects, 39% of whom were female, with a median age of 66years for analysis. Subjects with dysnatremia were more likely to be dead at 28 days (17% vs 7%; P<.001). There was a significant association between sodium fluctuation and mortality at 28 days (adjusted odds ratio per 1mmol/L change, 1.10 [95% confidence interval, 1.08-1.12; P<.001]), even in patients who remained normotremic during their intensive care unit stay (1.12 [95% confidence interval, 1.09-1.16; P<.001]) This observational study validates previous findings of an association between serum sodium fluctuations and mortality in surgical intensive care patients. This association was also present in subjects who remained normonatremic throughout their intensive care unit admission. •Dysnatremia is common and is associated with increased risk of mortality in postoperative patients requiring intensive care.•There is a significant association between sodium fluctuation and 28-day mortality, even in patients who remained normotremic throughout their ICU stay.•Subjects with dysnatremia were more likely to be dead at 28 days.•Severity of dysnatremia was associated with 28-day mortality.