Background: Respiratory support is crucial for newborns with underdeveloped lung. The clinical outcomes of patients depend on the clinician’s ability to recognize the status underlying the presented symptoms and signs. With the increasing number of high-risk infants, artificial intelligence (AI) should be considered as a tool for personalized neonatal care. Continuous monitoring of vital signs is essential in cardiorespiratory care. In this study, we developed deep learning (DL) prediction models for rapid and accurate detection of mechanical ventilation requirements in neonates using electronic health records (EHR). Methods: We utilized data from the neonatal intensive care unit in a single center, collected between March 3, 2012, and March 4, 2022, including 1,394 patient records used for model development, consisting of 505 and 889 patients with and without invasive mechanical ventilation (IMV) support, respectively. The proposed model architecture includes feature embedding using feature-wise fully connected (FC) layers, followed by three bidirectional long short-term memory (LSTM) layers. Results: A mean gestational age (GA) was 36.61 ± 3.25 weeks, and the mean birth weight was 2,734.01 ± 784.98 g. The IMV group had lower GA, birth weight, and longer hospitalization duration than the non-IMV group (P < 0.05). Our proposed model, tested on a dataset from March 4, 2019, to March 4, 2022. The mean AUROC of our proposed model for IMV support prediction performance demonstrated 0.861 (95%CI, 0.853–0.869). It is superior to conventional approaches, such as newborn early warning score systems (NEWS), Random Forest, and eXtreme gradient boosting (XGBoost) with 0.611 (95%CI, 0.600–0.622), 0.837 (95%CI, 0.828–0.845), and 0.0.831 (95%CI, 0.821–0.845), respectively. The highest AUPRC value is shown in the proposed model at 0.327 (95%CI, 0.308–0.347). The proposed model performed more accurate predictions as gestational age decreased. Additionally, the model exhibited the lowest alarm rate while maintaining the same sensitivity level. Conclusion: Deep learning approaches can help accurately standardize the prediction of invasive mechanical ventilation for neonatal patients and facilitate advanced neonatal care. The results of predictive, recall, and alarm performances of the proposed model outperformed the other models.

Background: Successful sepsis treatment depends on early diagnosis. We aimed to develop and validate a system to predict sepsis and septic shock in real time using deep learning. Methods: Clinical data were retrospectively collected from electronic medical records (EMRs). Data from 2010 to 2019 were used as development data, and data from 2020 to 2021 were used as validation data. The collected EMRs consisted of eight vital signs, 13 laboratory data points, and three demographic information items. We validated the deep-learning-based sepsis and septic shock early prediction system (DeepSEPS) using the validation datasets and compared our system with other traditional early warning scoring systems, such as the national early warning score, sequential organ failure assessment (SOFA), and quick sequential organ failure assessment. Results: DeepSEPS achieved even higher area under receiver operating characteristic curve (AUROC) values (0.7888 and 0.8494 for sepsis and septic shock, respectively) than SOFA. The prediction performance of traditional scoring systems was enhanced because the early prediction time point was close to the onset time of sepsis; however, the DeepSEPS scoring system consistently outperformed all conventional scoring systems at all time points. Furthermore, at the time of onset of sepsis and septic shock, DeepSEPS showed the highest AUROC (0.9346). Conclusions: The sepsis and septic shock early warning system developed in this study exhibited a performance that is worth considering when predicting sepsis and septic shock compared to other traditional early warning scoring systems. DeepSEPS showed better performance than existing sepsis prediction programs. This novel real-time system that simultaneously predicts sepsis and septic shock requires further validation.

Background: Ward patients who experience clinical deterioration are at high risk of mortality. Conventional rapid response systems (RRS) using track-and-trigger protocols have not consistently demonstrated improved outcomes. This study evaluated the impact of an artificial intelligence (AI)-based cardiac arrest prediction model. Methods: This 1-year, prospective, non-randomized interventional trial assigned hospitalized patients with AI-based software as a medical device (AI-SaMD) high-risk alerts to groups based on their subsequent clinical response; those reassessed or treated within 24 h comprised the AI-SaMD-guided cohort, while the remainder formed the usual care cohort. Alerts prompted an optional but not mandatory treatment review. The primary outcome was ward-based cardiac arrest; the secondary outcome was in-hospital mortality. Multivariable regression analysis was used to adjust for potential confounders. Results: Of 35,627 general ward admissions, 2906 triggered an AI-SaMD alert. Among these, 1409 (48.4%) were allocated to the AI-SaMD-guided cohort. The incidence of cardiac arrest significantly decreased from 2.07% to 1.06% (adjusted risk ratio (RR), 0.54; 95% confidence interval (CI), 0.20–0.88; p < 0.01). In-hospital mortality also significantly declined (adjusted RR, 0.65; 95% CI, 0.32–0.98; p < 0.05). Conclusions: AI-SaMD-guided alerts were associated with reductions in cardiac arrest and in-hospital mortality without requiring additional resources, supporting their integration into current clinical workflows to improve patient safety and optimize RRS performance.

Over recent years, the escalation of patient volumes in emergency departments (ED) worldwide has posed to the delivery of timely critical care. Intensive Care Unit (ICU) services became essential due to increasing acuity in EDs, and previous studies revealed a strong association between prolonged boarding times and unfavorable outcomes. Innovative strategies such as Emergency Department-based Intensive Care Units (ED-ICUs) have been introduced to optimize critical care delivery. Given the higher acuity and mortality rates in ED-ICU patients, the prediction of certain events, such as In-Hospital Cardiac Arrest (IHCA), has become abstruse. Conventional Early Warning Scores (EWSs) were developed to stratify the risk of conventional ICUs, but have never been validated in ED-ICU patients with higher acuity. Moreover, EWSs are predominantly focused on forecasting mortality and lack capability for real-time prediction. Our study aimed to develop and validate a deep-learning-based model to predict IHCA within 24 h in ED-ICU. We included 1975 patients admitted to ED-ICU. The study period was from 01 January 2019 to 31 December 2020. Our model, the Deep-ICU CMS (Central Monitoring System), uses four classic vital signs (blood pressure, heart rate, respiratory rate, and body temperature) as input. The model outperformed conventional EWSs in predicting IHCA and maintained performance even with extended prediction windows; it provided robust prediction within a 24-h window, setting it apart from models with restricted prediction horizons. It achieved notably high sensitivity and specificity, overcoming the alarm fatigue issue that is common in EWSs. This study pioneered IHCA risk stratification in ED-ICU and showcases Deep-ICU CMS as a robust prediction tool that overcomes the limitations of conventional EWSs. Prospective and external validation are now warranted to confirm the impact of Deep-ICU CMS in real-world practice. Given the scarcity of research in ED-ICU, our findings contribute valuable insights to optimizing critical care delivery.

Background/Objectives: In-hospital cardiac arrest (IHCA) remains a devastating event associated with high morbidity and mortality among general ward patients. While Rapid Response Systems (RRS) can help identify deteriorating patients, maintaining these systems in secondary hospitals is frequently hindered by severe fiscal and personnel constraints. Consequently, evidence regarding the real-world clinical effectiveness of artificial intelligence software as a medical device (AI-SaMD) for predicting deterioration in such resource-constrained settings remains limited. Methods: We conducted a retrospective analysis on a multicenter, staggered-implementation study evaluating 164,761 eligible adult general ward admissions across three secondary hospitals in South Korea. The intervention involved deploying an AI-SaMD (DeepCARS), which utilizes four routine vital signs to predict ward IHCA within 24 h. The primary outcome was ward IHCA. Secondary outcomes included in-hospital mortality and length of stay (LOS). Exploratory analyses investigated the mechanisms of clinical outcomes by evaluating lead-times to interventions, outcomes in sepsis subgroups, changes in care directives, and post-arrest neurological outcomes. Results: AI-SaMD implementation was associated with a 21% reduction in ward IHCA incidence (adjusted rate ratio 0.79; 95% CI, 0.65–0.96; p = 0.016) and a 15% reduction in in-hospital mortality (aRR 0.85; 95% CI, 0.79–0.90; p < 0.001), alongside significantly shorter hospital and intensive care unit LOS. These associations were also observed in patients with sepsis (IHCA aRR 0.71; 95% CI, 0.54–0.93; p = 0.013). Lead-times to critical care intervention and to antibiotic escalation were numerically shorter in the AI-SaMD group by 16.3 h (p = 0.066) and 2.6 h (p = 0.523); poor neurological outcome at discharge among ward IHCA cases was 85/108 (78.7%) in the AI-SaMD group versus 63/102 (61.8%) in the standard-care group (aRR 1.13; 95% CI, 0.99–1.33; p = 0.058); and the full-code death rate did not differ between groups (aRR 0.94; 95% CI, 0.76–1.15)—none of these additional analyses reached statistical significance. Conclusions: In secondary hospitals unable to operate an RRS due to fiscal limitations, implementation of an AI-SaMD as an additional informational layer was associated with lower ward IHCA and in-hospital mortality. The AI-SaMD may serve as an actionable and scalable additional safety layer for general-ward patients in resource-constrained environments where RRS infrastructure is not feasible. Although this was a multicenter, large-scale study, the present analysis was retrospective and quasi-experimental in design; rigorous randomized studies are needed to confirm these associations.

The use of airplanes is becoming increasingly frequent worldwide. However, although the number of flight passengers is steadily increasing, there is no comprehensive database that accurately reflects the frequency of emergency medical events that occur in airports. We evaluated the characteristics and treatment outcomes of patients who had emergency medical events at an airport. We retrospectively reviewed cases of patients who had emergency medical events at International Airport between May 2013 and April 2018. A comparative analysis was conducted on gender, age, disease, temperament, and average length of stay between patients visiting the airport and the general population visiting the emergency room (ED). Among the 258,823 patients who visited our ED during the study period. A total of 846 patients (0.3%) were transferred from the airport; the proportion of men was 59.3%, the mean age of the subjects was 43.7 ± 20.1 years. The admission and mortality rates of the patients in the airport group were relatively higher (35.1% and 2.6%, respectively) than that of those in the direct ED visit group (21.6% and 0.5%, respectively). Abdominal disease was the most common medical problem, and the most common causes of death were sudden cardiac arrest and acute myocardial infarction. Future prospective studies are necessary to affirm its findings.