Computer vision (CV), a type of artificial intelligence (AI) that uses digital videos or a sequence of images to recognize content, has been used extensively across industries in recent years. However, in the healthcare industry, its applications are limited by factors like privacy, safety, and ethical concerns. Despite this, CV has the potential to improve patient monitoring, and system efficiencies, while reducing workload. In contrast to previous reviews, we focus on the end-user applications of CV. First, we briefly review and categorize CV applications in other industries (job enhancement, surveillance and monitoring, automation, and augmented reality). We then review the developments of CV in the hospital setting, outpatient, and community settings. The recent advances in monitoring delirium, pain and sedation, patient deterioration, mechanical ventilation, mobility, patient safety, surgical applications, quantification of workload in the hospital, and monitoring for patient events outside the hospital are highlighted. To identify opportunities for future applications, we also completed journey mapping at different system levels. Lastly, we discuss the privacy, safety, and ethical considerations associated with CV and outline processes in algorithm development and testing that limit CV expansion in healthcare. This comprehensive review highlights CV applications and ideas for its expanded use in healthcare.

Pulmonary auscultation is essential for detecting abnormal lung sounds during physical assessments, but its reliability depends on the operator. Machine learning (ML) models offer an alternative by automatically classifying lung sounds. ML models require substantial data, and public databases aim to address this limitation. This systematic review compares characteristics, diagnostic accuracy, concerns, and data sources of existing models in the literature. Papers published from five major databases between 1990 and 2022 were assessed. Quality assessment was accomplished with a modified QUADAS-2 tool. The review encompassed 62 studies utilizing ML models and public-access databases for lung sound classification. Artificial neural networks (ANN) and support vector machines (SVM) were frequently employed in the ML classifiers. The accuracy ranged from 49.43% to 100% for discriminating abnormal sound types and 69.40% to 99.62% for disease class classification. Seventeen public databases were identified, with the ICBHI 2017 database being the most used (66%). The majority of studies exhibited a high risk of bias and concerns related to patient selection and reference standards. Summarizing, ML models can effectively classify abnormal lung sounds using publicly available data sources. Nevertheless, inconsistent reporting and methodologies pose limitations to advancing the field, and therefore, public databases should adhere to standardized recording and labeling procedures.

The King LT(S)-D™ laryngeal tube (King LT) has gained popularity as a bridge airway for pre-hospital airway management. In this study, we retrospectively reviewed the use of the King LT and its associated airway outcomes at a single Level 1 trauma centre.

Background: Endotracheal intubation can occur in up to 60% of critically ill patients. Despite the frequency with which endotracheal intubation occurs, the current practice is largely unknown. This is relevant, as advances in airway equipment (ie, video laryngoscopes) have become more prevalent, leading to possible improvement of care delivered during this process. In addition to new devices, a greater emphasis on airway plans and choices in sedation have evolved, although the influence on patient morbidity and mortality is largely unknown. Objective: This study aims to derive and validate prediction models for immediate airway and hemodynamic complications of intensive care unit intubations. Methods: A multicenter, observational, prospective study of adult critically ill patients admitted to both medical and surgical intensive care units (ICUs) was conducted. Participating ICU sites were located throughout eight health and human services regions of the United States for which endotracheal intubation was needed. A steering committee composed of both anesthesia and pulmonary critical care physicians proposed a core set of data variables. These variables were incorporated into a data collection form to be used within the multiple, participating ICUs across the United States during the time of intubation. The data collection form consisted of two basic components, focusing on airway management and hemodynamic management. The form was generated using RedCap and distributed to the participating centers. Quality checks on the dataset were performed several times with each center, such that they arrived at less than 10% missing values for each data variable; the checks were subsequently entered into a database. Results: The study is currently undergoing data analysis. Results are expected in November 2018 with publication to follow thereafter. The study protocol has not yet undergone peer review by a funding body. Conclusions: The overall goal of this multicenter prospective study is to develop a scoring system for peri-intubation, hemodynamic, and airway-related complications so we can stratify those patients at greatest risk for decompensation as a result of these complications. This will allow critical care physicians to be better prepared in addressing these occurrences and will allow them to improve the quality of care delivered to the critically ill. Trial Registration: ClinicalTrials.gov NCT02508948; https://clinicaltrials.gov/ct2/show/NCT02508948 (Archived by WebCite at http://www.webcitation.org/73Oj6cTFu) International Registered Report Identifier (IRRID): RR1-10.2196/11101

Background: Arterial desaturations experienced during endotracheal intubation (ETI) may lead to poor outcomes. Thus, our primary aim was to identify predictors of arterial desaturation (pulse oximetry <90%) during the peri-intubation period and to assess outcomes of those who developed arterial hypoxemia. Methods: Adult patients admitted to a medical and/or surgical intensive care unit (ICU) over the time period of January 1st 2013 through December 31st 2014 who required ETI were included. Only the first intubation was captured. Arterial desaturation was defined as pulse oximetry readings of <90% (hypoxemia) in the immediate peri-intubation period. Patients were then grouped in cases (those who developed desaturation) and controls (those who did not develop this complication). Results: The final cohort included 420 patients. Arterial desaturations occurred in 74 (18%) patients. When adjusting for significant predictors on univariate analysis and known predictors of a difficult airway, only acute respiratory failure (OR 2.38; 95% CI: 1.15–4.93; P=0.02) and provider training level (OR 7.12; 95% CI: 1.65–30.67; P=0.016) remained significant. Higher pulse oximetry readings prior to intubation was found to be protective on multivariate analysis (OR 0.92; 95% CI: 0.89–0.96; P<0.01; per one percent increase). Conclusions: Patients who were intubated for acute respiratory failure and those who were intubated by junior level trainees had increased odds of experiencing arterial desaturation in the peri-intubation period. Patients experiencing arterial desaturation had lower pulse oximetry readings prior to intubation suggesting a possible delay at intubation.