ABSTRACT
Objective:To develop a grading prediction model of traumatic hemorrhage volume based on deep learning and assist in predicting traumatic hemorrhage volume.Methods:A retrospective observational study was conducted based on the experimental data of pig gunshot wounds in the time-effect assessment database for experiments on war-traumatized animals constructed by the General Hospital of the Chinese People's Liberation Army. The hemorrhage volume data of the study population were extracted, and the animals were divided into 0-300 mL, 301-600 mL, and > 600 mL groups according to the hemorrhage volume. Using vital signs indexes as the predictive variables and hemorrhage volume grading as the outcome variable, trauma hemorrhage volume grading prediction models were developed based on four traditional machine learning and ten deep learning methods. Using laboratory test indexes as predictive variables and hemorrhage volume grading as outcome variables, trauma hemorrhage volume grading prediction models were developed based on the above fourteen methods. The effect of the two groups of models was evaluated by accuracy and area under the receiver operator characteristic curve (AUC), and the optimal models in the two groups were mixed to obtain hybrid model 1. Feature selection was conducted according to the genetic algorithm, and hybrid model 2 was constructed according to the best feature combination. Finally, hybrid model 2 was deployed in the animal experiment database system.Results:Ninety-six traumatic animals in the database were enrolled, including 27 pigs in the 0-300 mL group, 40 in the 301-600 mL group, and 29 in the > 600 mL group. Among the fourteen models based on vital signs indexes, fully convolutional network (FCN) model was the best [accuracy: 60.0%, AUC and 95% confidence interval (95% CI) was 0.699 (0.671-0.727)]. Among the fourteen models based on laboratory test indexes, recurrent neural network (RNN) model was the best [accuracy: 68.9%, AUC (95% CI) was 0.845 (0.829-0.860)]. After mixing the FCN and RNN models, the hybrid model 1, namely RNN-FCN model was obtained, and the performance of the model was improved [accuracy: 74.2%, AUC (95% CI) was 0.847 (0.833-0.862)]. Feature selection was carried out by genetic algorithm, and the hybrid model 2, namely RNN-FCN* model, was constructed according to the selected feature combination, which further improved the model performance [accuracy: 80.5%, AUC (95% CI) was 0.880 (0.868-0.893)]. The hybrid model 2 contained ten indexes, including mean arterial pressure (MAP), hematocrit (HCT), platelet count (PLT), lactic acid, arterial partial pressure of carbon dioxide (PaCO 2), Total CO 2, blood sodium, anion gap (AG), fibrinogen (FIB), international normalized ratio (INR). Finally, the RNN-FCN* model was deployed in the database system, which realized automatic, continuous, efficient, intelligent, and grading prediction of hemorrhage volume in traumatic animals. Conclusion:Based on deep learning, a grading prediction model of traumatic hemorrhage volume was developed and deployed in the information system to realize the intelligent grading prediction of traumatic animal hemorrhage volume.
ABSTRACT
Objective:To establish a stable fragment-induced penetrating liver injury model in landrace pigs and evaluate the characteristics of deep tissue injury.Methods:According to the different positioning methods of aiming points, twelve healthy adult landrace pigs were divided into group A (the relative height "h" of the aiming point and the highest point of the body surface on the tracing line was set to 5 cm) and group B ("h" was set to 6 cm). Ultrasonography was used to determine the direction of fragment projection, and an experimental ballistic gun was used to project high-velocity fragments to cause injury to animals. The vital signs of the two groups were monitored, and whole blood cell count, blood gas analysis, and liver and renal function were tested. Damages to the liver and adjacent organs, as well as the amount of bleeding and survival time were analyzed.Results:For the overall analysis of the two groups, the liver hit rate of fragment simulating projectiles was 100% (right anterior lobe and right lateral lobe injury), the hit rate of other organs in the abdominal cavity was 25% (3/12), and the incidence of hemothorax or pneumothorax was 8% (1/12). The wounds were mainly characterized by liver lacerations, with total or partial disconnection of the distal liver lobe. There was no significant difference in wound length and bleeding amount between groups A and B [wound length (cm): 9.8±1.7 vs. 11.2±3.8, bleeding amount (g): 597.0±477.1 vs. 1 032.0±390.3, both P > 0.05]. The depth of liver parenchymal laceration in group B with the aiming point closer to the anterior median line was significantly longer than that in group A (cm: 2.8±0.4 vs. 1.9±0.6, P = 0.015). Mean arterial pressure (MAP), pH value, residual arterial blood base (BE), hemoglobin (Hb) and hematocrit (HCT) levels decreased after the fragment-induced injury, and then reached a trough level [MAP (mmHg, 1 mmHg ≈ 0.133 kPa): 87.0±33.6, pH: 7.26±0.15, BE (mmol/L): -6.65±8.48, Hb (g/L): 9.86±1.10, HCT: 0.309±0.029, all P < 0.05] in the first hour. Blood lactate (Lac), lactate dehydrogenase (LDH) and aspartate aminotransferase (AST) levels increased over time, and reached a peak level [Lac (mmol/L): 10.21±4.40, LDH (U/L): 1 417.0±223.3, AST (U/L): 234.5 (162.5, 357.5), both P < 0.05] at 1 hour after injury. Pearson's correlation analysis showed that the total amount of bleeding was correlated with the depth of liver parenchyma laceration ( r = 0.684, P = 0.014). The Kaplan-Meier survival curve showed that the 3 hours survival rate in group A was higher than that in group B, but the difference was not statistically significant [83.3% (5/6) vs. 33.3% (2/6), P > 0.05]. Conclusions:The high-velocity fragment-induced penetrating liver injury model established by striking landrace pigs closer to the anterior median line with fragment simulating projectiles is reproducible and the degree of damage is controllable, and the model is applicable to further relevant research of hepatic ballistic trauma.
ABSTRACT
Objective:To observe the changes of arterial blood gas indexes in pigs with the free-field primary blast lung injury (PBLI) model, and to explore the value of arterial blood gas indexes in predicting moderate to severe PBLI.Methods:Nine adult healthy Landrace pigs were selected to construct the pig free-field PBLI model. Arterial blood samples were taken 15 minutes before the explosion (before injury) and 10, 30, 60, 120, and 180 minutes after the explosion (after injury). Arterial blood gas indexes and pulse oxygen saturation (SpO 2) were measured, compare the changes of blood gas analysis indexes and SpO 2 levels at different time points, and observe the changes of gross injury scores and pathological injury scores of lung tissue. Analyze the correlation between the blood gas indicators. Results:As time prolonged, at each time point, pH, arterial partial pressure of oxygen (PaO 2), and SpO 2 were lower than those before the injury, and blood lactic acid (Lac) and arterial partial pressure of carbon dioxide (PaCO 2) were higher than those before the injury. Compared with that before the injury, the pH value in the blood decreased significantly 10 minutes after the injury (7.39±0.06 vs. 7.46±0.02, P < 0.05), and the Lac increased significantly (mmol/L: 3.61±2.89 vs. 1.10±0.28, P < 0.05), and lasts until 180 minutes after injury (pH value: 7.37±0.07 vs. 7.46±0.02, Lac (mmol/L): 2.40±0.79 vs. 1.10±0.28, both P < 0.05); while PaO 2 and SpO 2 decreased significantly at 180 minutes after injury [PaO 2 (mmHg, 1 mmHg = 0.133 kPa): 59.40±10.94 vs. 74.81±9.39, P < 0.05; SpO 2: 0.75±0.11 vs. 0.89±0.08, P < 0.05], PaCO 2 increased significantly (mmHg: 56.17±5.38 vs. 48.42±4.93, P < 0.05). Correlation analysis showed that the gross injury score of lung blast injury animals was positively correlated with the pathological injury score ( r = 0.866, P = 0.005); PaO 2 and SpO 2 were positively correlated ( r = 0.703, P = 0.000); pH value and Lac were negative Correlation ( r = -0.400, P = 0.006); pH value is negatively correlated with PaCO 2 ( r = -0.844, P = 0.000). Conclusion:This study successfully established a large mammalian free-field PBLI model, arterial blood gas analysis is helpful for the early diagnosis of PBLI, whether SpO 2 can be used to evaluate the severity of lung injury remains to be further verified.
ABSTRACT
Background& AimsThe Coronavirus Disease 2019 (COVID-19) has become a global epidemic and has caused a lasting and huge loss of life security, economic development and social stability in more than 180 countries around the world. Unfortunately, there is still no specific treatment for COVID-19 till now, therefore, at this point, all potential therapies need to be critically considered. LL-37 is one of the best-studied human antimicrobial peptide (AMPs) that has a broad-spectrum activity against bacteria and viruses. The use of living, genetically modified organisms (GMOs) is an effective approach for delivery of therapeutic proteins. The aim of this study was to determine the safety and efficacy of the Lactococcus lactis which has been genetically modified to produce the therapeutic human antimicrobial peptide LL-37 (herein after referred to cas001) in the patients of COVID-19. MethodsFirstly we constructed genetically modified food-grade probiotic, Lactococcus lactis, with sequence of seven tandem repeats of mature human LL-37 under control of the nisin-inducible nisA promoter to produce the cas001. A total of 20 healthy SD rats, half male and half female (There were five male and five female in the control group, the same in treatment group) were used to observe the acute toxic reaction and death after daily administration of cas001 for three weeks, which helps to provide necessary reference basis for clinical dose selection, verificaition of toxic reaction and possible target organs. According to the estimated clinical dosage of 1 x 108CFU /kg/day, considering the conversion of body surface area, the dose for rats should be multiplied by 6.17 to 6 x 108 CFU/kg/day. We administrated 100 times higher dose at 6 x 1010 CFU/ /kg/day to rats. In order to investigate the pharmacokinetics of cas001, male SD rats (body weight 250-300g, 1 x 1010 /animal, n=3) were given oral administration of LL-37 bacteria powder. The concentration of LL-37 in the blood before and after gavage was detected by ELISA kit (Hycult biotechnology Cat# HK321). Human clinical study was approved by Ethics committee of Chinese PLA General Hospital (S2020-074-04) and a total of 11 patients with mild symptoms were enrolled in Wuhan hankou hospital and Huoshenshan hospital. They were enrolled voluntarily and all patients signed informed consent. Among them, there were 5 males and 6 females, aged 55 {+/-} 12 (36-70) years old, and the duration from onset to medication enrollment was 35 {+/-} 19 (5-68) days. 6 patients were nucleic acid positive and 5 patients were nucleic acid negative when they were enrolled. All patients received the oral drug cas001 treatment according to requirement(1 x 109 CFU/capsule, 3 capsules/time, three times a day for 3weeks), with an average follow-up time of 33 {+/-} 15 days (see table 1 for the results). O_TBL View this table: org.highwire.dtl.DTLVardef@1845d29org.highwire.dtl.DTLVardef@1004b38org.highwire.dtl.DTLVardef@4a741aorg.highwire.dtl.DTLVardef@c73d5org.highwire.dtl.DTLVardef@188b563_HPS_FORMAT_FIGEXP M_TBL O_FLOATNOTable 1.C_FLOATNO O_TABLECAPTIONSerum biochemical detection in rats two weeks after gavage Serum biochemistry detection of rats at the end of two weeks (Vehicle[male] v.s. LL-37[male], n=5; Vehicle[female] v.s. LL-37[female], n=5). *represent p value<0.05. C_TABLECAPTION C_TBL FindingsWestern blot analysis shows that reasonable amount of LL-37 were induced by different concentrations of nisin, which means we have successfully constructed cas001. In the pre-clinical safety evaluation test, after three weeks administration of cas001, no adverse effects were observed on the rats body weight, food and water intake, hematological or serum biochemical parameters. The results showed that the LD50 of cas001 was higher than that of the 100 times of the expected clinical dose of 6 x 1010 CFU/day. These results showed that cas001 could be safe in animal experiments. In addition, rat pharmacokinetics results showed that the serum concentration of LL-37 reached peak 2 hours after gavage of cas001 and returned to basal level 6 hours after gavage. During study period, the volunteers did not feel any discomfort while taking the cas001 capsules, and two hours after oral administration, the concentration of LL-37 were increased in healthy volunteers. cas001 shows definite effect in the improvement of gastrointestinal symptoms and is possible to have effects in improving the systemic symptoms and respiratory symptoms and may play a role in the improvement of results of nucleic acid test and lung CT test. 11 patients enrolled showed good compliance, tolerance, subjective feeling and actively interacted with the doctors. None of the patients had any adverse reactions. ConclusionsBased on above observations, we conclude here that as an oral anti-viral agent, cas001 displayed good safety profiles. It is very hard to reach conclusion of clinical outcomes related to the cas001, although changes of several symptoms indicate encouraging findings.
ABSTRACT
BackgroundCurrently, the prevention and control of the novel coronavirus disease (COVID-19) outside Hubei province in China, and other countries have become more and more critically serious. We developed and validated a diagnosis aid model without computed tomography (CT) images for early identification of suspected COVID-19 pneumonia (S-COVID-19-P) on admission in adult fever patients and made the validated model available via an online triage calculator. MethodsPatients admitted from Jan 14 to February 26, 2020 with the epidemiological history of exposure to COVID-19 were included [Model development (n = 132) and validation (n = 32)]. Candidate features included clinical symptoms, routine laboratory tests, and other clinical information on admission. Features selection and model development were based on the least absolute shrinkage and selection operator (LASSO) regression. The primary outcome was the development and validation of a diagnostic aid model for S-COVID-19-P early identification on admission. ResultsThe development cohort contained 26 S-COVID-19-P and 7 confirmed COVID-19 pneumonia cases. The final selected features included 1 variable of demographic information, 4 variables of vital signs, 5 variables of blood routine values, 7 variables of clinical signs and symptoms, and 1 infection-related biomarker. The model performance in the testing set and the validation cohort resulted in the area under the receiver operating characteristic (ROC) curves (AUCs) of 0.841 and 0.938, the F-1 score of 0.571 and 0.667, the recall of 1.000 and 1.000, the specificity of 0.727 and 0.778, and the precision of 0.400 and 0.500. The top 5 most important features were Age, IL-6, SYS_BP, MONO%, and Fever classification. Based on this model, an optimized strategy for S-COVID-19-P early identification in fever clinics has also been designed. ConclusionsS-COVID-19-P could be identified early by a machine-learning model only used collected clinical information without CT images on admission in fever clinics with a 100% recall score. The well-performed and validated model has been deployed as an online triage tool, which is available at https://intensivecare.shinyapps.io/COVID19/.
ABSTRACT
Blast injury is the main cause of injury in the battlefield, which also occurs frequently in the civil field and modern society. The damage caused by blast is more complicated than other types of trauma. Primary blast injury is a common type of blast injury, which can cause multiple organ damage with complex mechanism. Tissue and vascular endothelium damage and organ hypoperfusion are the consistent manifestations of most organ damage. However, due to the concealed damage caused by the primary blast injury, it is difficult to recognize it in time. The study of coagulation function and acid-base balance change after primary blast injury can bring benefits to its early diagnosis and intervention, thus improving the prognosis and mortality of blast injury. However, at present, the research on primary blast injury mostly focuses on single organ damage. Lack of research on systemic coagulation and acid-base balance changes calls for further research. Such research has a practical significance for the early diagnosis and optimization of tactical care for primary blast injury. This article reviews the injury characteristics, epidemiology, mechanism and the relationship with trauma-induced coagulopathy (TIC) in primary blast injury to provide reference for related researches.
ABSTRACT
Medical big data is a hot research topic in China,and it is also the main research direction in the field of emergency medicine.The current situation of the construction of the first-aid big data platform and the construction of the first-aid clinical decision support system were analyzed,the problems existing in the development of the first-aid big data research field were enumerated,to explore the theoretical methods for promoting the development of domestic first-aid big data,so as to provide references for the research in related fields.
ABSTRACT
On?the?premise?of?fully?studying?the?disaster?medical?rescue?monitoring?mechanism?in?emergencies?at?home?and?abroad,?the?functional?requirements?of?the?domestic?disaster?medical?rescue?monitoring?system?was?analyzed?in?this?paper,?the?logical?framework?and?data?structure?of?disaster?medical?rescue?monitoring?system?with?privacy?protection?mechanism?was?designed?by?department?of?emergency?in?Chinese?PLA?General?Hospital,?department?of?information?management?in?School?of?Economics?and?Management?of?Beijing?Jiaotong?University,?the?School?of?Information?Management?of?Nanjing?University.?Three?major?functional?modules?were?realized?in?the?system:?reporter?information?management,?disaster?medical?rescue?data?upload,?and?disaster?medical?rescue?data?search.?Android?client?and?Web?client?were?developed?for?easy?access?to?the?system.?The?system?also?had?the?function?of?privacy?protection.?Based?on?symmetric?searchable?encryption?algorithm,?the?system?realized?the?encryption?storage?of?untrusted?servers?and?ensured?the?security?of?medical?and?health?data.?It?is?beneficial?for?the?further?development?and?improvement?of?disaster?medical?rescue?data?collection?in?China.
ABSTRACT
The analysis of big data in medical field cannot be isolated from the high quality clinical database, and the construction of first aid database in our country is still in the early stage of exploration. This paper introduces the idea and key technology of the construction of multi-parameter first aid database. By combining emergency business flow with information flow, an emergency data integration model was designed with reference to the architecture of the Medical Information Mart for Intensive Care III (MIMIC-III), created by Computational Physiology Laboratory of Massachusetts Institute of Technology (MIT), and a high-quality first-aid database was built. The database currently covers 22 941 medical records for 19 814 different patients from May 2015 to October 2017, including relatively complete information on physiology, biochemistry, treatment, examination, nursing, etc. And based on the database, the first First-Aid Big Data Datathon event, which 13 teams from all over the country participated in, was launched. The First-Aid database provides a reference for the construction and application of clinical database in China. And it could provide powerful data support for scientific research, clinical decision making and the improvement of medical quality, which will further promote secondary analysis of clinical data in our country.
Subject(s)
Humans , Big Data , Critical Care , Databases, Factual , Medical InformaticsABSTRACT
OBJECTIVE@#To propose a method of prediction for fatal gastrointestinal bleeding recurrence in hospital and a method of feature selection via machine learning models.@*METHODS@#728 digestive tract hemorrhage samples were extracted from the first aid database of PLA General Hospital, and 343 patients among them were diagnosed as fatal gastrointestinal bleeding recurrence in hospital. A total of 64 physiological or laboratory indicators were extracted and screened. Based on the ten-fold cross-validation, Logistic regression, AdaBoost and XGBoost were used for classification prediction and comparison. XGBoost was used to search sequence features, and the key indicators for predicting fatal gastrointestinal bleeding recurrence in hospital were screened out according to the importance of the indicators during training.@*RESULTS@#Logistic regression, AdaBoost and XGBoost all get better F1.5 score under each feature input dimension, among which XGBoost had the best effect and the highest score, which was able to identify as many patients as possible who might have fatal gastrointestinal bleeding recurrence in hospital. Through XGBoost iteration results, the Top 30 indicators with high importance for predicting fatal gastrointestinal bleeding recurrence in hospital were ranked. The F1.5 scores of the first 12 key indicators peaked at iteration (0.893), including hemoglobin (Hb), calcium (CA), red blood cell count (RBC), mean platelet volume (MPV), mean erythrocyte hemoglobin concentration (MCH), systolic blood pressure (SBP), platelet count (PLT), magnesium (MG), lymphocyte (LYM), glucose (GLU, blood gas analysis), glucose (GLU, blood biochemistry) and diastolic blood pressure (DBP).@*CONCLUSIONS@#Logistic regression, AdaBoost and XGBoost could achieve the purpose of early warning for predicting fatal gastrointestinal bleeding recurrence in hospital, and XGBoost is the most suitable. The 12 most important indicators were screened out by sequential forward selection.
Subject(s)
Humans , Gastrointestinal Hemorrhage/mortality , Health Status Indicators , Hospital Mortality , Logistic Models , Machine Learning , RecurrenceABSTRACT
Objective To propose a method of prediction for fatal gastrointestinal bleeding recurrence in hospital and a method of feature selection via machine learning models. Methods 728 digestive tract hemorrhage samples were extracted from the first aid database of PLA General Hospital, and 343 patients among them were diagnosed as fatal gastrointestinal bleeding recurrence in hospital. A total of 64 physiological or laboratory indicators were extracted and screened. Based on the ten-fold cross-validation, Logistic regression, AdaBoost and XGBoost were used for classification prediction and comparison. XGBoost was used to search sequence features, and the key indicators for predicting fatal gastrointestinal bleeding recurrence in hospital were screened out according to the importance of the indicators during training. Results Logistic regression, AdaBoost and XGBoost all get better F1.5 score under each feature input dimension, among which XGBoost had the best effect and the highest score, which was able to identify as many patients as possible who might have fatal gastrointestinal bleeding recurrence in hospital. Through XGBoost iteration results, the Top 30 indicators with high importance for predicting fatal gastrointestinal bleeding recurrence in hospital were ranked. The F1.5 scores of the first 12 key indicators peaked at iteration (0.893), including hemoglobin (Hb), calcium (CA), red blood cell count (RBC), mean platelet volume (MPV), mean erythrocyte hemoglobin concentration (MCH), systolic blood pressure (SBP), platelet count (PLT), magnesium (MG), lymphocyte (LYM), glucose (GLU, blood gas analysis), glucose (GLU, blood biochemistry) and diastolic blood pressure (DBP). Conclusions Logistic regression, AdaBoost and XGBoost could achieve the purpose of early warning for predicting fatal gastrointestinal bleeding recurrence in hospital, and XGBoost is the most suitable. The 12 most important indicators were screened out by sequential forward selection.
ABSTRACT
Objective To analyze the correlation between acute coronary syndrome (ACS) and stress differentiation factors (GDF-15), catecholamines, and heat shock proteins (HSP-70). Methods A total of 40 patients with ACS were selected from the Emergency Department of the PLA General Hospital from September 10, 2016 to October 10, 2016. 40 healthy volunteers were selected as the control group. The information of age, gender, history of smoking, drinking, hyperlipidemia, hypertension and diabetes. Inspection indicators of blood biochemistry (Creation kinase Isoenzyme, Total cholesterol, Triglyceride, High-density lipoprotein, Blood glucose, Total bilirubin, Direct bilirubin), serum level of GDF-15, catecholamine (Adrenaline,norepinephrine,dopamine)and HSP-70 were collected. Evaluation of Coronary Stenosis used with Coronary Artery Lesions and Gensini Score. Statistical analysis using SPSS 17.0 statistical software, measurement data are expressed as mean ± standard deviation (x±s),count data to the number of cases and percentage, measurement using t test, count data using chisquare test. Results Serum levels of GDF-15[(21.94±14.23) vs. (7.06±5.53), P=0.007],catecholami ne[(46592.15±30931.27) vs. (5507.14±2083.28), P<0.01], HSP-70 [(369.56±300.44) vs. (07.76±54.23),P<0.001],all higher than the control group. GDF-15 serum levels of Gensini scores> 40 compare with <20group was significantly higher [(324.27 ± 198.81) vs. (77.43 ± 699.22), P=0.035], serum catecholaminelevels of > 40 group compare with <20 group significantly increased [(18.71 ± 7.32) vs. (18.6±46.1),P=0.017], GDF-15 levels were significantly higher in the multi-vessel stenosis group than in the doublevessel stenosis group[ (618.40±434.42) vs. (292.07±219.65), P=0.033]. Conclusions GDF-15,catecholamine and HSP-70 are correlated with ACS, as well as the severity of coronary artery lesions.
ABSTRACT
To introduce Medical Information Mart for Intensive Care (MIMIC) database and elaborate the approach of critically emergent research with big data based on the feature of MIMIC and updated studies both domestic and overseas, we put forward the feasibility and necessity of introducing medical big data to research in emergency. Then we discuss the role of MIMIC database in emergency clinical study, as well as the principles and key notes of experimental design and implementation under the medical big data circumstance. The implementation of MIMIC database in emergency medical research provides a brand new field for the early diagnosis, risk warning and prognosis of critical illness, however there are also limitations. To meet the era of big data, emergency medical database which is in accordance with our national condition is needed, which will provide new energy to the development of emergency medicine.
ABSTRACT
Objective The detailed analysis of the surveillance in post extreme emergencies and disasters (SPEED) provides practical reference for China to establish a disaster medical rescue information monitoring system with Chinese characteristics. Methods The SPEED system under the scene of disaster medical rescue information monitoring is analyzed in detail. The SPEED system design, work flows, system implementation and other aspects are analyzed and summarized in this paper, and suggests the enlightenment of SPEED system for Chinese disaster medical rescue information monitoring work. Results The SPEED system is an information monitoring system for the early stages of disasters. It provides monitoring for diseases caused by disasters, and life and health trends. It has a complete data collection mechanism, a comprehensive personnel training system, a complete system function, and an implementation strategy involving multi-layer, multi-region, and multi -sector. It is a powerful tool for disaster medical rescue and management personnel to obtain information in time. In the field of disaster medical rescue, a similar public-facing information monitoring system in China is still not perfect. Conclusion Learning the design flows and establishment mode of the SPEED system can provide reference for China to establish a disaster medical rescue information monitoring system with Chinese characteristics.
ABSTRACT
Objective To study the distribution of diseases in Medical Information Mart for Intensive Care Ⅲ(MIMIC-Ⅲ) database in order to provide reference for clinicians and engineers who use MIMIC-Ⅲ database to solve clinical research problems. Methods The exploratory data analysis technologies were used to explore the distribution characteristics of diseases and emergencies of patients (excluding newborns) in MIMIC-Ⅲ database were explored; then, neonatal gestational age, weight, length of hospital stay in intensive care unit (ICU) were analyzed with the same method. Results In the MIMIC-Ⅲ database, 46 428 patients were admitted for the first time, and 49 214 ICU records were recorded. There were 26 076 males and 20 352 females; the median age was 60.5 (38.6, 75.6) years, and most patients were between 60 and 80 years old. The first diagnosis in the disease spectrum analysis was firstly ranked by circulatory diseases (32%), followed by injury and poisoning (14%), digestive system disease (8%), tumor (7%), respiratory disease (6%) and so on. Patients with ischemic heart disease accounted for the largest proportion of circulatory disease (42%), the proportion of these patients gradually increased with age of 60-70 years old, then decreased. However, the proportion of patients with cerebrovascular disease declined first and then increased with age, which was the main cause of death of circulatory system disease (ICU mortality was 22.5%). Injury and poisoning patients showed a significant decrease with age. Digestive system diseases were younger than the general population (most people aged between 50 to 60 years), and non-infectious enteritis and colitis were the main causes of death (ICU mortality was 18.3%). Respiratory infections were predominant in infected patients (34%), but circulatory system infections were the main cause of death (ICU mortality was 25.6%). Secondly, in the neonatal care unit, premature infants accounted for the vast majority (82%). As the gestational age increased, the duration of ICU was decreased, and the mortality was decreased. Conclusions The diseases distribution of patients can be provided by MIMIC-Ⅲ database, which helps to grasp the overview of the volume and age distribution of the target patients in advance, and carry out the next step of research. Meanwhile, it points out the important role of exploratory data analysis in electronic health records analysis.
ABSTRACT
A detailed, high-scale clinical data can be generated in the process of diagnosis and treatment of emergency critically ill patients. The integration and analysis and utilization of these data are of great value for improving the treatment level and efficiency and developing the data-driven clinical assistant decision support. China has large volume of health information resources, however, the construction of healthcare databases and subsequent secondary analysis has just started. With the effort of the Chinese PLA General Hospital in building an emergency database and promoting data sharing, the first emergency database was published in China and a health Datathon was organized utilizing this database, providing experience for clinical data integration, database construction, cross-disciplinary collaboration and data sharing. Referring to the development at home and abroad, this review discussed work in this area and further proposed establishing a big data cooperation for emergency medicine and building a learning healthcare system to integrate more clinical resources and form a closed loop of "clinical database construction-analysis-applications", and enhance the effectiveness of medical big data in reducing medical costs and improving healthcare delivery.
ABSTRACT
Medical practice generates and stores immense amounts of clinical process data, while integrating and utilization of these data requires interdisciplinary cooperation together with novel models and methods to further promote applications of medical big data and research of artificial intelligence. A "Datathon" model is a novel event of data analysis and is typically organized as intense, short-duration, competitions in which participants with various knowledge and skills cooperate to address clinical questions based on "real world" data. This article introduces the origin of Datathon, organization of the events and relevant practice. The Datathon approach provides innovative solutions to promote cross-disciplinary collaboration and new methods for conducting research of big data in healthcare. It also offers insight into teaming up multi-expertise experts to investigate relevant clinical questions and further accelerate the application of medical big data.
ABSTRACT
Objective To construct a database containing multiple kinds of diseases that can provide "real world"data for first-aid clinical research. Methods Structured or non-structured information from hospital information system, laboratory information system, emergency medical system, emergency nursing system and bedside monitoring instruments of patients who visited department of emergency in PLA General Hospital from January 2014 to January 2018 were extracted. Database was created by forms, code writing, and data process. Results Emergency Rescue Database is a single center database established by PLA General Hospital. The information was collected from the patients who had visited the emergency department in PLA General Hospital since January 2014 to January 2018. The database included 530 585 patients' information of triage and 22 941 patients' information of treatment in critical rescue room, including information related to human demography, triage, medical records, vital signs, lab tests, image and biological examinations and so on. There were 12 tables (PATIENTS, TRIAGE_PATIENTS, EMG_PATIENTS_VISIT, VITAL_SIGNS, CHARTEVENTS, MEDICAL_ORDER, MEDICAL_RECORD, NURSING_RECORD, LAB_TEST_MASTER, LAB_RESULT, MEDICAL_EXAMINATION, EMG_INOUT_RECORD) that containing different kinds of patients' information. Conclusions The setup of high quality emergency databases lay solid ground for scientific researches based on data. The model of constructing Emergency Rescue Database could be the reference for other medical institutions to build multiple-diseases databases.
ABSTRACT
Objective To explore a method of screening the core indicators in the emergency database that can be used to evaluate the in-hospital fatal gastrointestinal rebleeding by using the big data algorithm. Methods Based on the emergency database of the Chinese PLA General Hospital, through the big data retrieval technology, all the 647 patients diagnosed as gastrointestinal bleeding in the emergency database were enrolled, except those who were admitted to the hospital for the first time and whose hemoglobin (Hb) was less than 90 g/L or did not undergo Hb test. Among them, there were 313 in the rebleeding group (fatal rebleeding in the hospital) and 334 in the non-rebleeding group (no fatal rebleeding in the hospital). General data of patients were collected, including gender, age, physical signs, blood gas, test index collection data, and the identification of gastrointestinal rebleeding. The fusion algorithm of rough set algorithm, genetic algorithm, and cellular automaton algorithm were used to calculate the key indicators that affect gastrointestinal rebleeding. Results A total of 499 indicators were calculated by machine fusion algorithm, after screening 5 times repeatedly, 24 key indicators were screened out, 3 of which were vital signs, including systolic blood pressure (SBP), diastolic blood pressure (DBP), temperature (T); 7 key indicators of blood routine, including white blood cell count (WBC), eosinophil (EOS), monocyte (MONO), Hb, hematocrit (HCT), red cell distribution width (RDW), mean corpuscular hemoglobin (MCH); 3 key indicators of coagulation, including prothrombin time (PT), plasma fibrinogen (FIB), activated partial thromboplastin time (APTT); 5 key indicators of biochemical, including myoglobin (MYO), chloride, glucose (GLU), serum albumin (ALB), total bilirubin (TBil); and 6 key indicators of blood gas, including pH, lactate (Lac), oxygen saturation (SO2), base excess (BE), bicarbonate (HCO3-), partial pressure of carbon dioxide (PaCO2). Conclusions Using big data technology, 24 core indicators for evaluating the fatal gastrointestinal rebleeding in hospitals can be screened out from the emergency database, providing new ideas and methods for clinical diagnosis of the disease.
ABSTRACT
Objective To explore the effect of toll-like receptor 4 (TLR4), myeloid differentiation protein-2 (MD2), and stromal interaction molecular 1 (STIM1) for regulating human vascular endothelial calcium overload injury and inflammatory reaction induced by bacterial endotoxin (LPS).Methods Human umbilical vein endothelial cells (HUVECs) were cultured in Dulbecco's modification of Eagle's medium (DMEM). ① The levels of TLR4, MD2 and nuclear factor-κB (NF-κB) were detected by reverse transcriotion-polymerase chain reaction (RT-PCR) before and 0.5, 1, 6, 12, 24 hours after LPS stimulation. ② Intracellular calcium peak level was detected by confocal following probe fluo-3 AM loading in HUVEC cells induced with LPS and transfected by psiSTIM or psiTLR. ③ MD2, STIM1 or NF-κB protein level was detected by immunoprecipitation (IP) and immuno-blotting in HUVEC cells which were transfected by TLR4 inhibited expression (psiTLR) for 12 hours and followed by LPS stimulation for 6 hours. ④ HUVEC cells were randomly divided into 6 groups: control group, LPS group, PDTC 0.1 mg/L group, PDTC 1 mg/L group, psiTLR 1 h group and psiTLR 12 h group. Tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) were detected by enzyme linked immunosorbent assay (ELISA) in supernatant. The mRNA levels of STIM1 and NF-κB were detected by RT-PCR.Results ① The mRNA levels of TLR4, MD2, and NF-κB gradually increased after LPS induction and peaked at 6 hours (2-ΔΔCt: 23.52±2.88, 17.43±3.43, 18.13±2.99, respectively), which were statistically significant before the stimulation with LPS (2-ΔΔCt: 7.02±2.81, 5.19±3.22, 8.11±1.42, allP < 0.05). ② Extracellular calcium influx in LPS group was increased significantly higher than control group (nmol/L: 108.13±22.33 vs. 41.57±13.19, P < 0.01). Extracellular calcium influx in psiSTIM+LPS group (nmol/L: 62.61±14.12 vs. 108.13±22.33,P < 0.05) and psiTLR+LPS group (nmol/L: 50.78±8.05 vs. 109.43±20.21,P < 0.01) were both suppressed as compared with LPS group. While extracellular calcium peak level in psiTLR+psiSTIM+LPS group further decreased (nmol/L: 39.31±6.42 vs. 109.43±20.21,P < 0.01). ③ MD2 protein but not STIM1 or NF-κB can be detected in anti-TLR4 precipitates in control (ctrl-) by immunoprecipitation. MD2 protein level increased in anti-TLR4 precipitates in LPS group (ctrl+) and was suppressed in TLR4 inhibiting group (psiTLR). ④ The levels of TNF-α in PDTC 1mg/L group were significantly lower than those of LPS group (ng/L: 0.60±0.24 vs. 1.77±0.66,P < 0.01). The levels of IL-6 in PDTC 0.1 mg/L, 1 mg/L group and psiTLR 12 h group decreased significantly lower than that of LPS group (ng/L: 232.10±63.54, 134.32±37.23, 284.23±56.14 vs. 510.22±89.23, allP < 0.05). Compared to LPS group, the mRNA levels of NF-κB and STIM1 were obviously inhibited in PDTC1 mg/L group and psiTLR 12 h group [NF-κB mRNA (2-ΔΔCt): 17.22±2.35, 13.24±3.54 vs. 30.16±2.06; STIM1 mRNA (2-ΔΔCt): 12.57±2.43, 12.21±2.46 vs. 25.12±2.02, allP < 0.05]. Conclusions TLR4, MD2, NF-κB signal and SOC calcium channel STIM1 mediate LPS induced-calcium influx and inflammatory mediators level in HUVEC cells. Extracellular calcium overload and inflammatory response by endotoxin induction can be effectively inhibited by down-regulation of TLR4, NF-κB and/or STIM1.