Background: Elderly patients undergoing hip fracture repair surgery are at increased risk of delirium due to aging, comorbidities, and frailty. But current methods for identifying the high risk of delirium among hospitalized patients have moderate accuracy and require extra questionnaires. Artificial intelligence makes it possible to establish machine learning models that predict incident delirium risk based on electronic health data.Methods: We conducted a retrospective case-control study on elderly patients (≥65 years of age) who received orthopedic repair with hip fracture under spinal or general anesthesia between June 1, 2018, and May 31, 2019. Anesthesia records and medical charts were reviewed to collect demographic, surgical, anesthetic features, and frailty index to explore potential risk factors for postoperative delirium. Delirium was assessed by trained nurses using the Confusion Assessment Method (CAM) every 12 h during the hospital stay. Four machine learning risk models were constructed to predict the incidence of postoperative delirium: random forest, eXtreme Gradient Boosting (XGBoosting), support vector machine (SVM), and multilayer perception (MLP). K-fold cross-validation was deployed to accomplish internal validation and performance evaluation.Results: About 245 patients were included and postoperative delirium affected 12.2% (30/245) of the patients. Multiple logistic regression revealed that dementia/history of stroke [OR 3.063, 95% CI (1.231, 7.624)], blood transfusion [OR 2.631, 95% CI (1.055, 6.559)], and preparation time [OR 1.476, 95% CI (1.170, 1.862)] were associated with postoperative delirium, achieving an area under receiver operating curve (AUC) of 0.779, 95% CI (0.703, 0.856).The accuracy of machine learning models for predicting the occurrence of postoperative delirium ranged from 83.67 to 87.75%. Machine learning methods detected 16 risk factors contributing to the development of delirium. Preparation time, frailty index uses of vasopressors during the surgery, dementia/history of stroke, duration of surgery, and anesthesia were the six most important risk factors of delirium.Conclusion: Electronic chart-derived machine learning models could generate hospital-specific delirium prediction models and calculate the contribution of risk factors to the occurrence of delirium. Further research is needed to evaluate the significance and applicability of electronic chart-derived machine learning models for the detection risk of delirium in elderly patients undergoing hip fracture repair surgeries.

The reliable monitoring of the depth of anesthesia (DoA) is essential to control the anesthesia procedure. Electroencephalography (EEG) has been widely used to estimate DoA since EEG could reflect the effect of anesthetic drugs on the central nervous system (CNS). In this study, we propose that a deep learning model consisting mainly of a deep residual shrinkage network (DRSN) and a 1 × 1 convolution network could estimate DoA in terms of patient state index (PSI) values. First, we preprocessed the four raw channels of EEG signals to remove electrical noise and other physiological signals. The proposed model then takes the preprocessed EEG signals as inputs to predict PSI values. Then we extracted 14 features from the preprocessed EEG signals and implemented three conventional feature-based models as comparisons. A dataset of 18 patients was used to evaluate the models’ performances. The results of the five-fold cross-validation show that there is a relatively high similarity between the ground-truth PSI values and the predicted PSI values of our proposed model, which outperforms the conventional models, and further, that the Spearman’s rank correlation coefficient is 0.9344. In addition, an ablation experiment was conducted to demonstrate the effectiveness of the soft-thresholding module for EEG-signal processing, and a cross-subject validation was implemented to illustrate the robustness of the proposed method. In summary, the procedure is not merely feasible for estimating DoA by mimicking PSI values but also inspired us to develop a precise DoA-estimation system with more convincing assessments of anesthetization levels.

Many tools are used to predict difficult airway, including bedside screening tests, radiological variables, and ultrasonography. However, the “gold standard” to identify difficult airway before intubation has not been established. The assessment and prediction of difficult airway is receiving increasing attention in clinical practice due to the devastating results of failed oxygenation or intubation. A literature visualization study is necessary to understand the research trend and help tailor future research directions. Science citation index-expanded web of Science database were used to search for literature related to assessment and prediction of difficult airways published before May 9th, 2022. VOS viewer software was used for visual analysis, including literature statistics, and co-occurrence analysis. A total of 2609 articles were included. The amount of relevant research interest and literature is increasing every year. According to co-occurrence network analysis, the research results can be grouped into the following 5 clusters, intubation approaches, intubation in special populations, difficult airway assessment tests, intubation in critical care/emergency settings and education, and laryngoscopes. Co-occurrence overlay analysis showed that video laryngoscopes and index prediction (including computed tomography and ultrasonography), emerged recently and comprised an important percentage of current studies. It can be predicted that future studies should focus on understanding the upper airway anatomy and constructing risk index predictions. Based on current global research trends, risk index predictions are the next hot topics in the evaluation and prediction of difficult airways, and video laryngoscopes will continue to be a hot topic in this field.

Background: As a leading cause of childhood blindness, the epidemic of retinopathy of prematurity (ROP) in China is characterized by advanced stage of ROP in more mature infants than those in the West. More advanced stage of disease necessitates more complicated surgical procedures and consequently exposure to general anesthesia. These ex-prematurely born infants are at risk of developing desaturation especially after surgery under general anesthesia. Physical status, anesthetic management and surgical profile are three main facets of perioperative setting and need to be investigated to identify useful predictors for perioperative adverse events in this population of fragile infants. Methods: In this retrospective cohort study, we enrolled all infants undergoing ROP surgeries at Peking University People's Hospital, Beijing, China from November 1, 2016 to October 31, 2017. Physical status, anesthesia and surgical management were analyzed by exploratory factor analysis and component matrix to explore risk factors for adverse events. Results: During the 12 months, 267 cases were included, among whom 61 infants underwent two surgeries required by their ophthalmological conditions. The median postconceptual age at the time of surgery was 46 (40, 53) weeks, and median body weight was 4.0 (3.0, 6.5) kg. None of the infants was dependent on caffeine, oxygen or ventilator before surgery. Bradycardia (29/267, 10.9%) and postoperative desaturation (34/267, 13.4%) were identified as major cardiac and respiratory adverse events. Preoperative atropine, intubation and bigger body weight would prevent patients suffering from bradycardia. Infants with a body weight less than 3.15 kg had a significantly higher chance of desaturation and neonatal intensive care unit admission after ROP surgeries than those who weighed more than 3.15 kg (27.8 vs. 5.1%, OR 5.46 (95% CI 2.66-11.21), P = 0.000). Conclusion: This study found that preoperative atropine and intubation would prevent bradycardia and low body weight was a predictor for both bradycardia and postoperative desaturation in preterm infants undergoing ROP surgeries.