Real-time artificial intelligence predicts adverse outcomes in acute pancreatitis in the emergency department: Comparison with clinical decision rule.

OBJECTIVE: Artificial intelligence (AI) prediction is increasingly used for decision making in health care, but its application for adverse outcomes in emergency department (ED) patients with acute pancreatitis (AP) is not well understood. This study aimed to clarify this aspect. METHODS: Data from 8274 ED patients with AP in three hospitals from 2009 to 2018 were analyzed. Demographic data, comorbidities, laboratory results, and adverse outcomes were included. Six algorithms were evaluated, and the one with the highest area under the curve (AUC) was implemented into the hospital information system (HIS) for real-time prediction. Predictive accuracy was compared between the AI model and Bedside Index for Severity in Acute Pancreatitis (BISAP). RESULTS: The mean ± SD age was 56.1 ± 16.7 years, with 67.7% being male. The AI model was successfully implemented in the HIS, with Light Gradient Boosting Machine (LightGBM) showing the highest AUC for sepsis (AUC 0.961) and intensive care unit (ICU) admission (AUC 0.973), and eXtreme Gradient Boosting (XGBoost) showing the highest AUC for mortality (AUC 0.975). Compared to BISAP, the AI model had superior AUC for sepsis (BISAP 0.785), ICU admission (BISAP 0.778), and mortality (BISAP 0.817). CONCLUSIONS: The first real-time AI prediction model implemented in the HIS for predicting adverse outcomes in ED patients with AP shows favorable initial results. However, further external validation is needed to ensure its reliability and accuracy.

Implementing a Novel Machine Learning System for Nutrition Education in Diabetes Mellitus Nutritional Clinic: Predicting 1-Year Blood Glucose Control.

(1) Background: Persistent hyperglycemia in diabetes mellitus (DM) increases the risk of death and causes cardiovascular disease (CVD), resulting in significant social and economic costs. This study used a machine learning (ML) technique to build prediction models with the factors of lifestyle, medication compliance, and self-control in eating habits and then implemented a predictive system based on the best model to forecast whether blood glucose can be well-controlled within 1 year in diabetic patients attending a DM nutritional clinic. (2) Methods: Data were collected from outpatients aged 20 years or older with type 2 DM who received nutrition education in Chi Mei Medical Center. Multiple ML algorithms were used to build the predictive models. (3) Results: The predictive models achieved accuracies ranging from 0.611 to 0.690. The XGBoost model with the highest area under the curve (AUC) of 0.738 was regarded as the best and used for the predictive system implementation. SHAP analysis was performed to interpret the feature importance in the best model. The predictive system, evaluated by dietitians, received positive feedback as a beneficial tool for diabetes nutrition consultations. (4) Conclusions: The ML prediction model provides a promising approach for diabetes nutrition consultations to maintain good long-term blood glucose control, reduce diabetes-related complications, and enhance the quality of medical care.

Real-time artificial intelligence system for bacteremia prediction in adult febrile emergency department patients.

BACKGROUND: Artificial intelligence (AI) holds significant potential to be a valuable tool in healthcare. However, its application for predicting bacteremia among adult febrile patients in the emergency department (ED) remains unclear. Therefore, we conducted a study to provide clarity on this issue. METHODS: Adult febrile ED patients with blood cultures at Chi Mei Medical Center were divided into derivation (January 2017 to June 2019) and validation groups (July 2019 to December 2020). The derivation group was utilized to develop AI models using twenty-one feature variables and five algorithms to predict bacteremia. The performance of these models was compared with qSOFA score. The AI model with the highest area under the receiver operating characteristics curve (AUC) was chosen to implement the AI prediction system and tested on the validation group. RESULTS: The study included 5,647 febrile patients. In the derivation group, there were 3,369 patients with a mean age of 61.4 years, and 50.7% were female, including 508 (13.8%) with bacteremia. The model with the best AUC was built using the random forest algorithm (0.761), followed by logistic regression (0.755). All five models demonstrated better AUC than the qSOFA score (0.560). The random forest model was adopted to build a real-time AI prediction system integrated into the hospital information system, and the AUC achieved 0.709 in the validation group. CONCLUSION: The AI model shows promise to predict bacteremia in adult febrile ED patients; however, further external validation in different hospitals and populations is necessary to verify its effectiveness.

Using an Artificial Intelligence Approach to Predict the Adverse Effects and Prognosis of Tuberculosis.

BACKGROUND: Tuberculosis (TB) is one of the leading causes of death worldwide and a major cause of ill health. Without treatment, the mortality rate of TB is approximately 50%; with treatment, most patients with TB can be cured. However, anti-TB drug treatments may result in many adverse effects. Therefore, it is important to detect and predict these adverse effects early. Our study aimed to build models using an artificial intelligence/machine learning approach to predict acute hepatitis, acute respiratory failure, and mortality after TB treatment. MATERIALS AND METHODS: Adult patients (age ≥ 20 years) who had a TB diagnosis and received treatment from January 2004 to December 2021 were enrolled in the present study. Thirty-six feature variables were used to develop the predictive models with AI. The data were randomly stratified into a training dataset for model building (70%) and a testing dataset for model validation (30%). These algorithms included XGBoost, random forest, MLP, light GBM, logistic regression, and SVM. RESULTS: A total of 2248 TB patients in Chi Mei Medical Center were included in the study; 71.7% were males, and the other 28.3% were females. The mean age was 67.7 ± 16.4 years. The results showed that our models using the six AI algorithms all had a high area under the receiver operating characteristic curve (AUC) in predicting acute hepatitis, respiratory failure, and mortality, and the AUCs ranged from 0.920 to 0.766, 0.884 to 0.797, and 0.834 to 0.737, respectively. CONCLUSIONS: Our AI models were good predictors and can provide clinicians with a valuable tool to detect the adverse prognosis in TB patients early.

Design and Implementation of a Comprehensive AI Dashboard for Real-Time Prediction of Adverse Prognosis of ED Patients.

The emergency department (ED) is at the forefront of medical care, and the medical team needs to make outright judgments and treatment decisions under time constraints. Thus, knowing how to make personalized and precise predictions is a very challenging task. With the advancement of artificial intelligence (AI) technology, Chi Mei Medical Center (CMMC) adopted AI, the Internet of Things (IoT), and interaction technologies to establish diverse prognosis prediction models for eight diseases based on the ED electronic medical records of three branch hospitals. CMMC integrated these predictive models to form a digital AI dashboard, showing the risk status of all ED patients diagnosed with any of these eight diseases. This study first explored the methodology of CMMC's AI development and proposed a four-tier AI dashboard architecture for ED implementation. The AI dashboard's ease of use, usefulness, and acceptance was also strongly affirmed by the ED medical staff. The ED AI dashboard is an effective tool in the implementation of real-time risk monitoring of patients in the ED and could improve the quality of care as a part of best practice. Based on the results of this study, it is suggested that healthcare institutions thoughtfully consider tailoring their ED dashboard designs to adapt to their unique workflows and environments.

Anesthesia management for cesarean section in a pregnant woman with odontogenic infection: A case report.

BACKGROUND: In recent years, people have paid more attention to oral health with the development of stomatology. Due to the various physiological changes during pregnancy, such as changing hormone levels and immune functions, oral diseases have a high incidence during pregnancy, and the prevention and treatment of oral diseases have also received the attention of both dentists and obstetricians. However, the anesthetic management of pregnant patients with oral disease, especially severe maxillofacial infections, and patients who need surgical treatment or have obstetric emergencies and need to terminate their pregnancy is not clear. CASE SUMMARY: This article describes a parturient patient with a severe masseteric space infection who had an emergency cesarean section. CONCLUSION: This case report aims to discuss the important anesthetic considerations for these patients.

Real-time interactive artificial intelligence of things-based prediction for adverse outcomes in adult patients with pneumonia in the emergency department.

BACKGROUND: Artificial intelligence of things (AIoT) may be a solution for predicting adverse outcomes in emergency department (ED) patients with pneumonia; however, this issue remains unclear. Therefore, we conducted this study to clarify it. METHODS: We identified 52,626 adult ED patients with pneumonia from three hospitals between 2010 and 2019 for this study. Thirty-three feature variables from electronic medical records were used to construct an artificial intelligence (AI) model to predict sepsis or septic shock, respiratory failure, and mortality. After comparisons of the predictive accuracies among logistic regression, random forest, support-vector machine (SVM), light gradient boosting machine (LightGBM), multilayer perceptron (MLP), and eXtreme Gradient Boosting (XGBoost), we selected the best one to build the model. We further combined the AI model with the Internet of things as AIoT, added an interactive mode, and implemented it in the hospital information system to assist clinicians with decision making in real time. We also compared the AIoT-based model with the confusion-urea-respiratory rate-blood pressure-65 (CURB-65) and pneumonia severity index (PSI) for predicting mortality. RESULTS: The best AI algorithms were random forest for sepsis or septic shock (area under the curve [AUC] = 0.781), LightGBM for respiratory failure (AUC = 0.847), and mortality (AUC = 0.835). The AIoT-based model represented better performance than CURB-65 and PSI indicators for predicting mortality (0.835 vs. 0.681 and 0.835 vs. 0.728). CONCLUSIONS: A real-time interactive AIoT-based model might be a better tool for predicting adverse outcomes in ED patients with pneumonia. Further validation in other populations is warranted.

[ED50 of Intrathecal Isobaric Bupivacaine with Epidural Volume Extension for Cesarean Delivery].

OBJECTIVE: To determine the 50% effective dose (ED50) of intrathecal isobaric bupivacaine in combined spinal-epidural anaesthesia with epidural volume extension for caesarean surgery. METHODS: Eighty-four women undergoing elective caesarean sections were randomly assigned to seven groups, receiving intrathecally isobaric bupivacaine in a dosage of 6, 7, 8, 9, 10, 11 or 12 mg (n=12 per group), respectively. Ten mL normal saline was injected through an epidural catheter at a rate of 0.5 mL/s. Successful induction was defined as Hollmen pinprick block grade 2 at a height equal to or higher than T6. Success (surgery) was defined as success (induction) plus no requirement for epidural supplementation throughout surgery. The ED50 values were determined using a logistic regression model. RESULTS: The EDs for success (induction) and success (surgery) were 8.121 mg (95% confidence interval: 7.421-8.821) and 9.012 mg (95% confidence interval: 7.747-10.280), respectively. With increase in amount of intrathecal bupivacaine, time to success (induction) was shortened (P<0.05), resulting in increased successful numbers for both induction and surgery. No differences were found in the use of phenylephrine, haemodynamic stability and the occurrence of nausea and vomiting among groups. CONCLUSION: When combined with epidural volume extension, intrathecal bupivacaine has an ED50 of 8.121 mg and 9.012 mg for success (induction) and success (surgery), respectively, in women undergoing caesarean sections.

[Liver injury and Nrf2 expression in neonatal rats with hyperoxia].

OBJECTIVE: To investigate the effect of continuous exposure to hyperoxia on livers and changes in nuclear factor erythroid 2-related factor 2 (Nrf2) expression in hepatic tissues in neonatal rats. METHODS: 100 neonatal Sprague-Dawley rats were randomized divided into hyperoxia group (FiO = 95%, O group) and normal control group (FiOG2 = 21%, N group) immediately after birth. Hepatic apoptotic index and Nrf2 expression were detected by immunohistochemical methods on 4 d, 7 d, and 14 d. RESULTS: Rats with hyperoxia had higher levels of hepatic apoptotic index and Nrf2 expression compared with the controls (P < 0.01). The levels of hepatic apoptotic index and Nrf2 expression in rats with hyperoxia were higher on 7 d and 14 d than those on 4 d (P < 0.01). CONCLUSION: Continuous exposure to hyperoxia could result in hepatic damage to neonatal rats, with increased expression of Nrf2 as a mechanism of anti-oxidant.

[Protection effect of recombiant human erythropoietin preconditioning against intrauterine hypoxic-ischemic brain injury and its influence on expression of caspase-3 protein in brain tissue].

OBJECTIVE: To investigate the effects of recombine human erythropoietin (rhEPO) on neural cells apoptosis and the expression of Caspase-3 protein in brain tissue of fetal rats after intrauterine hypoxic-ischemic brain injury. METHODS: Forty-four Sprague-Dawley rats on 19 days of pregnancy were divided into rhEPO treated group, ischemia-reperfusion group and sham-operated group. Intrauterine hypoxic-ischemic injury of fetal rats was induced by bilateral occlusion of the utero-ovarian artery for 20 min. rhEPO (5000 U/kg) was injected into rats through caudal vein in rhEPO treated group while saline was injected into rats in hypoxic-ischemic group 30 min before hypoxic-ischemic injury. The brain samples in rhEPO treated group and hypoxic-ischemic group were obtained at 30 min, 3 h, 6 h, 24 h and 48 h respectively after artery clamping. There was no hypoxic-ischemic injury in sham-operated group, so the brain samples were obtained at 24 hours after sham operation. Neuroapoptosis in brain tissue was measured by TdT mediated dUTP-biotin nick end labeling (Tunel) staining. The expression of Caspase-3 protein was observed by immunohistochemistry. RESULTS: The number of apoptosis cells in fetal rat hippocampus after intrauterine hypoxic-ischemic increased progressively with reperfusion. Compared with the I/R group, the number of apoptosis cells decreased in rhEPO treated group (P < 0.01). The expression of Caspase-3 increased rapidly after 3 hours from the reperfusion in the I/R group. Compared with the I/R group, there was less expression of Caspase-3 in rhEPO treated group (P < 0.01). CONCLUSION: rhEPO showed the effects to inhibit the apoptosis of fetal neural cells and the expression of Caspase-3 protein due to intrauterine hypoxic-ischemic brain injury.

[Effect of recombinant human erythropoietin on apoptosis of neural cells in fetal rats after intrauterine hypoxic-ischemic injury].

OBJECTIVE: To investigate the effect of recombine human erythropoietin (rhEPO) on apoptosis of neural cells in fetal rats after intrauterine hypoxic-ischemic injury. METHODS: Twenty SD rats on 19 days of pregnancy were divided into rhEPO (2500 U/kg, 5000 U/kg, 7500 U/kg) treated groups, ischemia-reperfusion (I/R) group and sham-operated group (4 rats in each group). Intrauterine hypoxic-ischemic injury of fetal rat was induced by bilateral occlusion of utero-ovarian artery for 20 min. rhEPO was injected into the rats in rhEPO treated group through the caudal vein 30 min before hypoxic-ischemic injury while saline was used in the other two groups. There was no hypoxic-ischemic injury in sham-operated group. The death rate of fetal rats was evaluated at 24 h after the operation, and then the brain samples of fetal rats were harvested. The expression of Caspase-3 protein was observed by immunohistochemistry. Neuroapoptosis was measured by TdT mediated dUTP-biotin nick end labeling (TUNEL) staining. RESULTS: Death rates of fetal rats in rhEPO treated groups decreased compared with the I/R group (P < 0.05). Compared with the I/R group, there was less expression of copious Caspase-3 in rhEPO treated group (P < 0.01). The expression of Caspase-3 was decreased in the rhEPO treated groups with the increase of rhEPO dose (P < 0.01). Compared with the I/R group, the death rate of fetal rats in rhEPO treated groups decreased (P < 0.05), the number of apoptosis cells also decreased obviously (P < 0.01). The anti-apoptosis effect of 5000 U/kg rhEPO was similar to 7500 U/kg rhEPO, but better than 2500 U/kg rhEPO (P < 0.01). CONCLUSION: rhEPO can inhibit the apoptosis of fetal rat brain cells after intrauterine hypoxic-ischemic injury.

[Erythropoietin through the placenta barrier and fetal blood-brain barrier with transient uteroplacental ischemia].

OBJECTIVE: To observe the permeability of recombinant human erythropoietin through placenta barrier and fetal blood-brain barrier after transient uteroplacental ischemia. METHODS: Rats on days 19 of pregnancy were divided into rhEPO treated group, ischemia-reperfusion group and sham operated group. Fetal ischemia in rhEPO treated group and ischemia-reperfusion group was induced by bilateral occlusion of the utero-ovarian artery for 20 minutes. Different dosage of 125I-rhEPO (2500 U/kg, 5000 U/kg, 7500 U/kg) was injected into the rats through caudal veins 30 min before injury in rhEPO treated group and sham-operated group. Saline was administered intravenously 30 min before the induction of hypoxic-ischemic injury in ischemia-reperfusion group. The amniotic fluid, placenta and fetal organs including brain, liver, heart, lung and kidney were collected to measure the radioactivity at 24h after injury. RESULTS: 125I-rhEPO was detected in amniotic fluid, placenta and fetal organs. The radioactivity of 125I-rhEPO in these tissues increased gradually with the increased dose injected in rhEPO treated group and sham-operated group. There were significant differences in the radioactivity of 125I-rhEPO between rhEPO treated group and sham-operated group (P < 0.05). CONCLUSION: The permeability of rhEPO through placental barrier and blood-brain barrier increased under the condition of fetal ischemia and hypoxia.

[Permeability of exogenetic recombine human erythropoietin through placenta barrier to fetal rats with transient uteroplacental ischemia].

OBJECTIVE: To investigate whether recombine human erythropoietin can cross the placenta barrier in rats with transient uteroplacental ischemia. METHODS: Rats on day 19 of pregnancy were divided into ischemia-reperfusion group, sham-operated group and rhEPO treated group. Fetal ischemia was induced by bilateral occlusion of the utero-ovarian artery for 20 minutes. 125I-rhEPO or saline were administered intravenously 30 min before the induction of hypoxic-ischemic injury. Fetal rat organs were removed to measure the radioactivity post injury. RESULTS: A small amount of radioactive activity (1.26 +/- 0.28) pg/g was detected in the fetal rats with ischemia-reperfusion. 125I-rhEPO radioactivity increased gradually with time in the placenta, amniotic fluid and fetal tissues of vital organs in the rats of sham-operated group and rhEPO treatment group. There were significant differences in 125I-rhEPO between placental organs and other organs (P < 0.05). The permeability of 125I-rhEPO through the blood brain barrier changed with reperfusion time and peaked about 6 h after reperfusion, and significant differences were found between rhEPO treatment group and sham operation group (P < 0.05). CONCLUSION: Exogenous rhEPO can cross the placenta barrier and blood-brain barrier and reach hypoxic-ischemic fetal rats.

[Effects of granisetron/lidocaine combination on propofol injection-induced pain: a double-blind randomized clinical trial].

OBJECTIVE: To investigate the alleviation effect of vein pretreatment and granisetron/lidocaine combination on propofol injection-induced pain. METHODS: Two hundreds patients scheduled for gynaecological laparoscopic operations were randomly divided into four groups: control group (group I), lidocaine group (group II), granisetron group (group III) and granisetron/lidocaine combination group (group IV), with 50 patients in each group. The patients in the above four groups received placebo (saline), lidocaine 20 mg, granisetron 2 mg and granisetron 2 mg plus lidocaine intravenously respectively. The patients were injected with one-forth of scheduled propofol via a dorsal hand vein after one minute of venous occlusion. The pain during the injection of propofol was evaluated. RESULTS: Pain occurred in 84% of patients in the control group, 46% in the lidocaine group, 52% in the granisetron group and 24% in the granisetron/lidocaine combination group. There was a significant reduction in pain incidence in the three experimental groups compared with the control group (P<0.05). The incidence of pain, nausea, vomiting and shivering was less in the granisetron/lidocaine combination group than in the control group (P<0.05). CONCLUSION: Pretreatment with granisetron/lidocaine may be effective not only in attenuating pains during i.v. injection of propofol, but also in preventing postoperative nausea, vomiting and shivering.