Abstract
Purpose
To construct a decision-making app for pre-hospital damage control resuscitation (PHDCR) for severely injured patients, and to make a preliminary trial test on the effectiveness and usability aspects of the constructed app.
Methods
Decision-making algorithms were first established by a thorough literature review, and were then used to be learned by computer with 3 kinds of text segmentation algorithms, i.e., dictionary-based segmentation, machine learning algorithms based on labeling, and deep learning algorithms based on understanding. B/S architecture mode and Spring Boot were used as a framework to construct the app. A total of 16 Grade-5 medical students were recruited to test the effectiveness and usability aspects of the app by using an animal model-based test on simulated PHDCR. Twelve adult Bama miniature pigs were subjected to penetrating abdominal injuries and were randomly assigned to the 16 students, who were randomly divided into 2 groups (n = 8 each): group A (decided on PHDCR by themselves) and group B (decided on PHDCR with the aid of the app). The students were asked to complete the PHDCR within 1 h, and then blood samples were taken and thromboelastography, routine coagulation test, blood cell count, and blood gas analysis were examined. The lab examination results along with the value of mean arterial pressure were used to compare the resuscitation effects between the 2 groups. Furthermore, a 4-statement-based post-test survey on a 5-point Likert scale was performed in group B students to test the usability aspects of the constructed app.
Results
With the above 3 kinds of text segmentation algorithm, B/S architecture mode, and Spring Boot as the development framework, the decision-making app for PHDCR was successfully constructed. The time to decide PHDCR was (28.8 ± 3.41) sec in group B, much shorter than that in group A (87.5 ± 8.53) sec (p < 0.001). The outcomes of animals treated by group B students were much better than that by group A students as indicated by higher mean arterial pressure, oxygen saturation and fibrinogen concentration and maximum amplitude, and lower R values in group B than those in group A. The post-test survey revealed that group B students gave a mean score of no less than 4 for all 4 statements.
Conclusion
A decision-making app for PHDCR was constructed in the present study and the preliminary trial test revealed that it could help to improve the resuscitation effect in animal models of penetrating abdominal injury.
Keywords: Pre-hospital damage control resuscitation, Decision-making, App, Trauma, Severely injured patients
1. Introduction
Pre-hospital care is an important part of the emergency medical system, which is inseparable and crucial for the successful treatment of the injured and sick.1 Nowadays, many measures have been utilized to improve pre-hospital care to reduce the mortality rates, life-long disabilities, and costs, such as enhancing the education and training of pre-hospital staff,2 improvements of ambulance medical and communication equipment,3, 4, 5, 6 implementation of telemedicine systems,7,8 and optimization of the pre-hospital emergency team.9,10 For example, it was found that having an anesthesiologist on the pre-hospital emergency team for airway management can effectively reduce the complication rates.9,10 In addition, many apps have been developed to facilitate pre-hospital care.11, 12, 13 In Canada, a pre-hospital emergency transport planning decision-making software was developed to optimize the medical care of stroke patients;11 in England, a smartphone app, Major Trauma Triage Tool, was developed and was found to be utilized safely by pre-hospital clinicians in supporting triage decisions relating to potential major trauma;12 and in Netherlands, a pre-hospital triage protocol, the Trauma Triage App, was tested to aid emergency medical services professionals in the identification of patients in need of specialized trauma care.13
Pre-hospital care for trauma patients is extremely important since a considerable percentage of trauma patients die in the phase of pre-hospital care. With the popularization of advanced trauma life support (ATLS) courses, trauma mortality, specifically preventable or potentially preventable mortality among trauma patients, has been reduced greatly.14,15 However, pre-hospital damage control resuscitation (PHDCR), an important part of the care of severely injured patients,16,17 is not covered by ATLS courses. In general, the pre-hospital staff in most countries are not professional trauma surgeons and do not master the skills of PHDCR, putting severely injured patients in danger.18,19 Under this condition, a decision-making app of PHDCR might be helpful. In the present study, we aimed to develop a decision-making app for PHDCR and to conduct preliminary experimental tests on animal models to provide a basis for the next clinical application.
2. Methods
All procedures involving animals were approved by the Ethics Committee of the Army Medical University of China PLA and were performed following the relevant regulations of the Ethics Committee of the Army Medical University of China PLA, China.
2.1. Construction of the decision-making app for PHDCR
2.1.1. Design concept
This app was developed to aid the pre-hospital staff who are not usually professional trauma surgeons to make decisions of PHDCR. Decision-making algorithms were firstly established based on the published literature related to PHDCR, then were incorporated into software. The decision was made based on the physiological indicators (e.g. body temperature, respiratory rate, blood pressure, heart rate, etc.), and point-of-care lab testing indicators (e.g. hemoglobin concentration and international standardized ratio (INR), etc.).
2.1.2. Establishment of decision-making algorithms for PHDCR
A thorough literature review was made using various combinations of the following keywords: “damage control resuscitation”, “pre-hospital care”, “pre-hospital damage control resuscitation”, “decision-making”, “prediction”, “transfusion”, “severe trauma”, and “severe bleeding”, etc. Then the most often used key physiological indicators and point-of-care lab examination indicators were selected, and stratifications of each indicator were established. The corresponding diagnosis and treatment measures were designated to the stratifications of each indicator.
2.1.3. Construction of decision-making app for PHDCR
B/S architecture mode was used to develop the app. Three kinds of Chinese text segmentation algorithm, i.e., dictionary-based segmentation, machine learning algorithms based on labeling, and deep learning algorithms based on understanding,20,21 were used to learn the above-mentioned decision-making algorithms for PHDCR. Among them, dictionary-based is also known as mechanical segmentation, and machine learning algorithms and deep learning algorithms are collectively referred to as statistical segmentation methods. This project uses the “Jieba” Chinese word segmentation component based on statistical segmentation methods to perform word segmentation, word labeling, and keyword extraction on the Chinese text of the injury description, and also supports custom dictionaries.22,23 “Jieba” word segmentation mainly uses the same dictionary for word segmentation and word labeling. Based on the patient's injury history, injury site, and other disease description text information, combined with the diagnosis and treatment content of trauma surgery, the decision module of "disease analysis" first divides the input disease text into Chinese words through the Chinese text segmentation algorithm. Then, keyword extraction was used to match the corresponding injury site, diagnosis, and treatment suggestions.
According to the principle of software layering, the back-end system is divided into 3 parts: the view layer, the business layer, and the data layer, which are technically implemented in Java language, and Spring Boot is used as the development framework.24,25 The view layer is mainly composed of the Thymeleaf template engine on the backend and the Asynchronous JavaScript and XML communication module on the frontend. The business layer consists of a user login module, a casualty management module, a case management module, an injury diagnosis engine, and a treatment plan management module. It serves as a bridge between user access and the actual database, housing important business logic. The data layer is mainly composed of Structured Query Language statements and related code that directly access the database,26 and is technically completed using the Spring Java Database Connectivity Template framework.
2.2. Trial test of the decision-making app for PHDCR
2.2.1. Participants and groups
A total of 16 Grade-5 medical students from our university were recruited to test the effectiveness of helping decision-making of the app and the usability aspects of the app by animal model-based resuscitation. There were 14 boys and 2 girls, with an average age of (23.1 ± 0.6) years. The students were randomly divided into 2 groups (n = 8 each): group A students decided on PHDCR by their own judgment, and group B students decided on PHDCR with the aid of the app. Group B students accepted 10-min pre-testing training to use the app.
2.2.2. Testing of the effectiveness of animal model-based resuscitation
Sixteen adult Bama miniature pigs of either sex were provided by the Laboratory Animal Center of the Army Medical University. The animals were anesthetized intramuscularly with 1.5 μg/kg of Sufentanil Citrate (for inducing anesthesia) and 6 mg/kg/h Propofol Emulsion (for maintaining anesthesia). The right femoral artery was bluntly separated, and an arterial catheter was placed as a blood sampling channel and connected to an invasive blood pressure monitor. Mean arterial pressure was continuously monitored and was recorded before injury (0 h) and 1 h after injury.
Penetrating abdominal injuries were then produced with a homemade custom-made machine.27,28 Briefly, the pig was fixed in a supine position on a multifunctional animal injury platform, and the injury mode was set to projectile impact injury using a pressure of 2000 psi. The laser sighting point was set 2 cm to the left of the pig's third nipple, with the projectile exit point 28 cm away from the pig and at a vertical angle to the horizontal line (Fig. 1). Once the injury conditions were set, the projectile was fired.
Fig. 1.
Preparation of animal model. (A) Machine used to produce animal model. (B) The pig was fixed in the supine position, preparing to be injured.
Ten-min after the injury, the animals were randomly assigned to the 16 students. Lab examination results including blood cell count, blood gas analysis, and INR were given to the students. The students were asked to perform a simulated PHDCR based on the physiological indicators and lab examination results. Group A students decided on PHDCR by themselves, and group B students decided on PHDCR with the aid of the app. Resuscitation fluid consisted of hydroxyethyl starch, lactate Ringer's liquid, prothrombin complex concentrate (PCC), fibrinogen concentrate (FC), and tranexamic acid (TXA). The students were allowed to select the kinds and dosages of resuscitation fluid and reagents by themselves.
The students were asked to complete the PHDCR within 1 h, and then blood samples were taken to test the effects of resuscitation. Thromboelastography, routine coagulation test, and blood gas analysis were examined as previously described,27,28 and the R value, maximum amplitude (MA), pH value, fibrinogen concentration, and oxygen saturation were recorded.
2.2.3. Test of the usability aspects of the app by post-test survey
After the training, the trainees were asked to indicate their agreement with a series of survey items related to usability aspects of the app on a 5-point Likert scale (1 = fully disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = fully agree).29 The post-test survey consisted of 4 statements: (1) The user interface is friendly; (2) It's easy to input various types of data; (3) The treatment instructions given by the app are easy to understand and follow; (4) I feel confident in applying PHDCR with the help of the app.
2.3. Statistical analysis
SPSS statistics software, version 25.0 (SPSS Inc, Chicago, IL, USA), was used to analyze the data, and the quantitative data such as routine coagulation test and thromboelastography were expressed as the mean ± standard deviation (SD). The t-test was used for comparison between the 2 groups. A p ≤ 0.05 was considered significant.
3. Result
3.1. The decision-making app for PHDCR was successfully constructed
Based on the published literature,30, 31, 32, 33, 34, 35, 36, 37 decision-making algorithms for PHDCR were established. Five physiological indicators (i.e., body temperature, systolic blood pressure, respiratory rate, blood pressure, heart rate, and oxygen saturation) and 5 point-of-care lab testing indicators (i.e., hemoglobin concentration, base excess, platelet count, fibrinogen concentration and INR) were selected. For each indicator, there were 3–4 stratifications; and related diagnosis and treatment measures were designated to each stratification (Table 1).
Table 1.
Decision-making algorithms for pre-hospital damage control resuscitation.
| Parameters | Stratification | Diagnosis | Suggested treatment |
|---|---|---|---|
| Heart rate (beats/min) | 101–160 | Mild to moderate hemorrhagic shock | Infusion of 500 mL hydroxyethyl starch and 500 mL lactate Ringer's liquid |
| >160 or 40–59 | Severe hemorrhagic shock | Infusion of 500 mL hydroxyethyl starch and 1000 mL lactate Ringer's liquid | |
| <40 | Cardiac respiratory arrest | Cardiopulmonary resuscitation | |
| Systolic blood pressure (mmHg) | 60–89 | Mild to moderate hemorrhagic shock | 1. Infuse 200 mL–400 mL of red blood cell suspension; 2. Infuse 500 mL of hydroxyethyl starch; 3. Infuse 800 mL of lactate Ringer's liquid; 4. Decide the type and amount of fluid resuscitation in the next step according to the evolution of the disease |
| 40–59 | Severe hemorrhagic shock | 1. Infuse 600 mL of red blood cell suspension; 2. Infuse 1000 mL of hydroxyethyl starch; 3. Infuse 1000 mL of lactate Ringer's liquid; 4. Decide the type and amount of fluid resuscitation in the next step according to the evolution of the disease |
|
| <40 | Cardiac respiratory arrest | Cardiopulmonary resuscitation | |
| Temperature (°C) | 32.0–35.9 | Mild hypothermia | Heat preservation |
| 30.0–32.0 | Moderate hypothermia | Heat preservation + Infusion of warmed fluids | |
| <30.0 | Severe hypothermia | Heat preservation + Infusion of warmed fluids | |
| Respiration rate (breathes/min) | 21–35 | Mild dyspnea | 1. Inhaled oxygen concentration 30%–50%; 2. Find and treat the cause of dyspnea |
| 36–50 or 6–11 | Moderate dyspnea | 1. Breathe pure oxygen; 2. Find and treat the cause of dyspnea |
|
| >50 or <6 | Severe dyspnea | 1. Breathe pure oxygen; 2. Find and treat the cause of dyspnea |
|
| Oxygen saturation (%) | 80–92 | Mild hypoxemia | High concentration oxygen and look for the cause |
| 60–79 | Moderate hypoxemia | Breathe pure oxygen and look for the cause | |
| ≤59 | Severe hypoxemia | Breathe pure oxygen and look for the cause | |
| Hemoglobin concentration (g/L) | 110–159 | Mild hypohemoglobinemia | Infuse 200 mL of red blood cell suspension |
| 60–109 | Moderate hypohemoglobinemia | Infuse 400 mL of red blood cell suspension | |
| <60 | Severe hypohemoglobinemia | Infuse 800 mL of red blood cell suspension | |
| Base excess (mmol/L) | -3–-6 | Mild acidosis | None |
| -6.1–-9 | Moderate acidosis | None | |
| ≤9.1 | Severe acidosis | Infuse 1.25% of sodium bicarbonate | |
| INR | 1.31–2.5 | Mild coagulopathy | Infuse 1 g of TXA as soon as possible within 1 h after injury and then another 1 g of TXA within 8 h. |
| 2.5–3.5 | Moderate coagulopathy | 1. Infuse 1 unit of PCC 2. Infuse 1 g of TXA as soon as possible within 1 h after injury and then another 1 g of TXA within 8 h. |
|
| ≥3.5 | Severe coagulopathy | 1. Infuse 2 units of PCC 2. Infuse 1 g of TXA as soon as possible within 1 h after injury and then another 1 g of TXA within 8 h. |
|
| Platelet count (10^9/L) | 100–300 | Mild thrombocytopenia | None |
| 50–99 | Moderate thrombocytopenia | Infuse 2 units of platelets | |
| ≤49 | Severe thrombocytopenia | Infuse 2−4 units of platelets | |
| Fibrinogen concentration (g/L) | 2–4 | Mild hypofibrinogenemia | None |
| 0.5–2 | Moderate hypofibrinogenemia | Infuse 2 g of FC | |
| <0.5 | Severe hypofibrinogenemia | Infuse 4 g of FC |
INR: international standardized ratio; TXA: tranexamic acid; PCC: prothrombin complex concentrate; FC: fibrinogen concentrate.
With the above-mentioned 3 kinds of Chinese text segmentation algorithms, B/S architecture mode, and Spring Boot as the development framework, the decision-making app for PHDCR was successfully constructed. Users input their account and password through the login interface and click the "login" button to enter the system. On the case management page, injury information could be added by clicking the "Add Case" button, or selecting the "Details" button of the existing injury to view the case details (Fig. 2). After editing the injury description and detection indicators of the patient, click the "analysis" button to match the corresponding diagnosis and treatment plan. As the patient's condition evolves, the text and detection indicators can be edited again in the first medical record to generate a second treatment result. It took about 30 sec to enter the data of each single injured patient, and 5 sec for calculation and for the app to demonstrate the diagnosis and treatment recommendations.
Fig. 2.
The app of decision-making for pre-hospital damage control resuscitation. (A) Edit the casualty text. (B) Input the data of physiological indicators and lab examination results. (C) Case management page.
3.2. The app helped to improve the outcomes of animal model-based trial test
After clarifying the animal injury situation, group B edited the injury text and inputted the physiological indicators and lab examination results into the software. Click "Analyze" to obtain diagnostic information and treatment details. The time to decide PHDCR was (28.8 ± 3.4) sec in group B, much shorter than that in group A (87.5 ± 8.5) sec.
An average of 500 mL hydroxyethyl starch, 300 mL lactate Ringer's liquid, 1.33 units of FC, 1 unit of PCC, and 1 bolus of TXA were used in each swine in group B, and an average of 500 mL hydroxyethyl starch, 300 mL lactate Ringer's liquid and 1 bolus of TXA were used in each pig in group A.
There was no animal die at 1 h after injury. The mean arterial pressure and oxygen saturation were significantly higher in group B than those in group A, indicating a better physiological condition after PHDCR with the aid of the app (Table 2). This might be caused by better coagulation conditions in group B than that in group A, as indicated by higher MA values and fibrinogen concentration, and lower R values in group B than those in group A (Table 2).
Table 2.
Physiological outcomes and lab examination results at 1 h after injury.
| Items | 0 h | 1 h after injury | p1 value | p2 value |
|---|---|---|---|---|
| Mean arterial pressure (mmHg) | ||||
| Group A | 86.25 ± 5.42 | 58.33 ± 6.12 | <0.001 | <0.001 |
| Group B | 84.92 ± 4.23 | 70.13 ± 3.49 | <0.001 | |
| Oxygen saturation (%) | ||||
| Group A | 98.63 ± 1.19 | 74.00 ± 4.21 | <0.001 | <0.001 |
| Group B | 98.38 ± 1.06 | 84.13 ± 5.08 | <0.001 | |
| Fibrinogen concentration (g/L) | ||||
| Group A | 2.16 ± 0.15 | 1.45 ± 0.24 | <0.001 | 0.011 |
| Group B | 2.06 ± 0.13 | 1.72 ± 0.10 | <0.001 | |
| Maximum amplitude | ||||
| Group A | 78.53 ± 2.69 | 62.58 ± 4.20 | <0.001 | 0.006 |
| Group B | 79.02 ± 2.92 | 69.73 ± 4.54 | <0.001 | |
| R values | ||||
| Group A | 1.66 ± 0.33 | 5.19 ± 1.02 | <0.001 | 0.012 |
| Group B | 1.59 ± 0.34 | 3.90 ± 0.73 | <0.001 | |
| pH value | ||||
| Group A | 7.674 ± 0.05 | 7.523 ± 0.11 | 0.003 | 0.069 |
| Group B | 7.687 ± 0.03 | 7.615 ± 0.07 | 0.020 |
p1: Comparison between 1 h after injury and baseline value (0 h).
p2: Comparison between group A and group B at 1 h after injury.
Data presented as mean ± SD, excepted p values.
3.3. The usability aspects of the app
Group B students gave a mean score no less than 4 for all 4 statements (Fig. 3), indicating that the students are satisfied with the user experience of the app, including the user interface, the convenience of inputting the data, readability of the results of output; and they were confident in applying HDCR with the help of the app.
Fig. 3.
Results of post-test survey rating. S1 to S4 referred to statement 1 to statement 5, respectively.
4. Discussion
A decision-making app for PHDCR was successfully constructed in the present study and demonstrated in animal experiments. The experiment was designed in parallel with the actual sample size of 8 cases in each group. The statistical analysis efficacy under the sample size was calculated using pass software, and the average MA value of the main evaluation index in group A was 62.58 according to α = 0.05. The SD is 4.20; the mean value and standard deviation of group B are 69.73 and 4.54, and the statistical test efficacy of the main analysis indicators in this study is above 85%. It was found to help aid grade-5 students to improve the resuscitation effect in animal models of penetrating abdominal injury. And post-test survey revealed that the students are satisfied with the user experience of the app.
Damage control resuscitation is a standard protocol for severely injured patients, and it has evolved significantly in the past decade, from simple permissive hypotensive resuscitation to balanced resuscitation with a fixed transfusion ratio of blood product, and to hemostatic resuscitation with FC, PCC, and TXA under the guide of viscoelastic test.28,32 A considerable number of injured patients develop traumatic coagulopathy upon arrival at the hospitals in civilian trauma care,38 and this proportion significantly increases during wartime. It was found that 38% of the wounded had already developed traumatic coagulopathy when they were sent to field hospitals in Operation Iraqi Freedom,39,40 and a higher proportion of the injured have experienced hypothermia,39,40 which is an important part of lethal triad. These facts highlight the importance of implementing PHDCR in the care of severely injured patients. However, the judgment of the injury state and how to implement PHDCR is rather difficult for pre-hospital staff without profound professional knowledge of trauma care and PHDCR. For the first time, a decision-making app for PHDCR was constructed in the present study and it was found it could help to improve the resuscitation effect in animal models of penetrating abdominal injury. In addition, this app has a significant prospect in combat casualty care, where PHDCR is needed and professional trauma surgeon is lacking during the stage of tactical care.39,40
Several measures were taken to make the app practical and easy to use. First, only 5 key physiological indicators and 5 key point-of-care lab testing indicators were selected to construct decision-making algorithms for PHDCR, making it easy to understand and enter the related data into the app. Second, the “Jieba” Chinese word segmentation component based on statistical segmentation methods,22 was used to establish the calculation algorithm of the app, making the software run very fast and smoothly. Third, the view layer is mainly composed of the Thymeleaf template engine on the backend and the AJAX communication module on the front-end. The Thymeleaf template engine encapsulated the front-end user interface page, making the user interface very friendly. These were confirmed by post-test survey results that group B students gave a mean score of no less than 4 for all 4 statements.
The current study possesses several limitations. First, only one animal model, i.e., penetrating abdominal injury, was used to test the effectiveness of the app. More types of animal models and injured human beings should be used to test the effectiveness of the app. Second, point-of-care lab examinations used in the app may not be available in some remote rural areas. Under this circumstance, simpler indicators might be needed to construct the app for decision-making app for PHDCR. Third, although the students we recruited were 5th graders who lacked clinical experience, they might have different familiarity with the operation of damage control resuscitation. The heterogeneity of the study population was not analyzed and needs to be improved in the next step.
In conclusion, a decision-making app for PHDCR was constructed in the present study and the preliminary trial test revealed that it could help to improve the resuscitation effect in animal models of penetrating abdominal injury. And tests on more types of animal models and injured human beings are needed to further validate the effectiveness of the app.
CRediT authorship contribution statement
Haoyang Yang: Writing – original draft, Software, Methodology. Wenqiong Du: Data curation. Zhaowen Zong: Writing – original draft, Software, Funding acquisition. Xin Zhong: Methodology. Yijun Jia: Validation. Renqing Jiang: Validation. Chenglin Dai: Resources. Zhao Ye: Data curation.
Ethical statement
All procedures involving animals were approved by the Ethics Committee of the Army Medical University of China PLA and were performed in accordance withfollowing the relevant regulations of the Ethics Committee of the Army Medical University of China PLA, China.
Funding
This work was supported by Key Project of State Key Laboratory of Trauma, Burn and Combined Injury (SKLTS202213).
Declaration of competing interest
The authors declare no conflicts of interest.
Footnotes
Peer review under responsibility of Chinese Medical Association.
References
- 1.Wilson M.H., Habig K., Wright C., et al. Pre-hospital emergency medicine. Lancet. 2015;386:2526–2534. doi: 10.1016/S0140-6736(15)00985-X. [DOI] [PubMed] [Google Scholar]
- 2.Hani E., Saeedi M., Hekmatpou D. The impacts of head trauma management education on the clinical decision-making of pre-hospital emergency staff. J Educ Health Promot. 2023;12:113. doi: 10.4103/jehp.jehp_376_22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wydo S.M., Seamon M.J., Melanson S.W., et al. Portable ultrasound in disaster triage: a focused review. Eur J Trauma Emerg Surg. 2016;42:151–159. doi: 10.1007/s00068-015-0498-8. [DOI] [PubMed] [Google Scholar]
- 4.Ogedegbe C., Morchel H., Hazelwood V., et al. Development and evaluation of a novel, real time mobile telesonography system in management of patients with abdominal trauma: study protocol. BMC Emerg Med. 2012;12:19. doi: 10.1186/1471-227X-12-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhang K., Liu W.L., Locatis C., et al. Mobile videoconferencing apps for telemedicine. Telemed J e Health. 2016;22:56–62. doi: 10.1089/tmj.2015.0027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tokunaga K., Inoue S., Suruga Y., et al. Practical use of a communication application on mobile devices by our stroke team. J Neuroendovasc Ther. 2020;14:339–344. doi: 10.5797/jnet.oa.2020-0030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wake E., Atkins H., Willock A., et al. Telehealth in trauma: a scoping review. J Telemed Telecare. 2022;28:412–422. doi: 10.1177/1357633X20940868. [DOI] [PubMed] [Google Scholar]
- 8.Langabeer J.R., Champagne-Langabeer T., Alqusairi D., et al. Cost-benefit analysis of telehealth in pre-hospital care. J Telemed Telecare. 2017;23:747–751. doi: 10.1177/1357633X16680541. [DOI] [PubMed] [Google Scholar]
- 9.Gellerfors M., Fevang E., Bäckman A., et al. Pre-hospital advanced airway management by anaesthetist and nurse anaesthetist critical care teams: a prospective observational study of 2028 pre-hospital tracheal intubations. Br J Anaesth. 2018;120:1103–1109. doi: 10.1016/j.bja.2017.12.036. [DOI] [PubMed] [Google Scholar]
- 10.Árnason B., Hertzberg D., Kornhall D., et al. Pre-hospital emergency anaesthesia in trauma patients treated by anaesthesiologist and nurse anaesthetist staffed critical care teams. Acta Anaesthesiol Scand. 2021;65:1329–1336. doi: 10.1111/aas.13946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Holodinsky J.K., Francis M.J., Goyal M., et al. Testing the usability of a software for geospatial and transport modeling in acute stroke service planning. Front Neurol. 2019;10:694. doi: 10.3389/fneur.2019.00694. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Freshwater E.S., Crouch R. Technology for trauma: testing the validity of a smartphone app for pre-hospital clinicians. Int Emerg Nurs. 2015;23:32–37. doi: 10.1016/j.ienj.2014.04.003. [DOI] [PubMed] [Google Scholar]
- 13.van der Sluijs R., Fiddelers A., Waalwijk J.F., et al. The impact of the Trauma Triage App on pre-hospital trauma triage: design and protocol of the stepped-wedge, cluster-randomized TESLA trial. Diagn Progn Res. 2020;4:10. doi: 10.1186/s41512-020-00076-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Navarro S., Montmany S., Rebasa P., et al. Impact of ATLS training on preventable and potentially preventable deaths. World J Surg. 2014;38:2273–2278. doi: 10.1007/s00268-014-2587-y. [DOI] [PubMed] [Google Scholar]
- 15.Mohammad A., Branicki F., Abu-Zidan F.M. Educational and clinical impact of Advanced Trauma Life Support (ATLS) courses: a systematic review. World J Surg. 2014;38:322–329. doi: 10.1007/s00268-013-2294-0. [DOI] [PubMed] [Google Scholar]
- 16.Pereira B., Dorigatti A., Calderon L., et al. Pre-hospital environment bleeding: from history to future prospects. Anaesthesiol Intensive Ther. 2019;51:240–248. doi: 10.5114/ait.2019.86059. [DOI] [PubMed] [Google Scholar]
- 17.van Turenhout E.C., Bossers S.M., Loer S.A., et al. Pre-hospital transfusion of red blood cells. Part 1: a scoping review of current practice and transfusion triggers. Transfus Med. 2020;30:86–105. doi: 10.1111/tme.12667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lukumay G.G., Ndile M.L., Outwater A.H., et al. Provision of post-crash first aid by traffic police in Dar es Salaam, Tanzania: a cross-sectional survey. BMC Emerg Med. 2018;18:45. doi: 10.1186/s12873-018-0199-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Laville V., Lebleu C., Desterke C., et al. Pre-hospital management of pediatric polytrauma during modern conflict: experience and limits of the French military health service. Eur J Trauma Emerg Surg. 2019;45:437–443. doi: 10.1007/s00068-018-0915-x. [DOI] [PubMed] [Google Scholar]
- 20.Xu Q., Wang Z. A data-driven model for Automated Chinese word segmentation and POS tagging. Comput Intell Neurosci. 2022;2022 doi: 10.1155/2022/7622392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Otter D.W., Medina J.R., Kalita J.K. A survey of the usages of deep learning for Natural language processing. IEEE Transact Neural Networks Learn Syst. 2021;32:604–624. doi: 10.1109/TNNLS.2020.2979670. [DOI] [PubMed] [Google Scholar]
- 22.Jieba project. https://github.com/fxsjy/jieba
- 23.Chen N., Su X., Liu T., et al. A benchmark dataset and case study for Chinese medical question intent classification. BMC Med Inf Decis Making. 2020;20:125. doi: 10.1186/s12911-020-1122-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Chen Q., Shi H. Design and implementation of domain knowledge service platform based on spring framework. Proceedings of the Association for Information Science and Technology. 2017;54:635–636. doi: 10.1002/pra2.2017.14505401096. [DOI] [Google Scholar]
- 25.zhang F., Sun G., Zheng B., et al. Design and implementation of energy management system based on spring Boot framework. Info. 2021;12:457. doi: 10.3390/INFO12110457. [DOI] [Google Scholar]
- 26.Ding G., Sun S., Wang G. Schema matching based on SQL statement. Distrib Parallel Databases. 2020;38:193–226. doi: 10.1007/s10619-019-07268-9. [DOI] [Google Scholar]
- 27.Du W.Q., Zhong X., Jiang R.Q., et al. Animal model-based simulation training for three emergent and urgent operations of penetrating thoracic injuries. Chin J Traumatol. 2023;26:41–47. doi: 10.1016/j.cjtee.2022.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ye Z., Zong Z., Zhong X., et al. Characterization of combined blast- and fragment-induced Pelvic injuries and hemostatic resuscitation in rabbits. J Surg Res. 2023;285:158–167. doi: 10.1016/j.jss.2022.12.031. [DOI] [PubMed] [Google Scholar]
- 29.Du W., Zhong X., Jia Y., et al. A novel scenario-based, mixed-reality platform for training Nontechnical skills of battlefield first aid: prospective interventional study. JMIR Serious Games. 2022;10 doi: 10.2196/40727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Brockamp T., Nienaber U., Mutschler M., et al. Predicting on-going hemorrhage and transfusion requirement after severe trauma: a validation of six scoring systems and algorithms on the TraumaRegister DGU. Crit Care. 2012;16:R129. doi: 10.1186/cc11432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Cantle P.M., Cotton B.A. Prediction of massive transfusion in trauma. Crit Care Clin. 2017;33:71–84. doi: 10.1016/j.ccc.2016.08.002. [DOI] [PubMed] [Google Scholar]
- 32.Cap A.P., Pidcoke H.F., Spinella P., et al. Damage control resuscitation. Mil Med. 2018;183:36–43. doi: 10.1093/milmed/usy112. [DOI] [PubMed] [Google Scholar]
- 33.Maegele M., Brockamp T., Nienaber U., et al. Predictive models and algorithms for the need of transfusion including massive transfusion in severely injured patients. Transfus Med Hemotherapy. 2012;39:85–97. doi: 10.1159/000337243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Maegele M., Paffrath T., Bouillon B. Acute traumatic coagulopathy in severe injury: incidence, risk stratification, and treatment options. Dtsch Arztebl Int. 2011;108:827–835. doi: 10.3238/arztebl.2011.0836a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Peltan I.D., Rowhani-Rahbar A., Vande Vusse L.K., et al. Development and validation of a prehospital prediction model for acute traumatic coagulopathy. Crit Care. 2016;20:371. doi: 10.1186/s13054-016-1541-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kozek-Langenecker S.A., Ahmed A.B., Afshari A., et al. Management of severe perioperative bleeding: guidelines from the European Society of Anaesthesiology: first update 2016. Eur J Anaesthesiol. 2017;34:332–395. doi: 10.1097/EJA.0000000000000630. [DOI] [PubMed] [Google Scholar]
- 37.Spahn D.R., Bouillon B., Cerny V., et al. The European guideline on management of major bleeding and coagulopathy following trauma: fifth edition. Crit Care. 2019;23:98. doi: 10.1186/s13054-019-2347-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Chang R., Eastridge B.J., Holcomb J.B. Remote damage control resuscitation in Austere Environments. Wilderness Environ Med. 2017;28:S124–S134. doi: 10.1016/j.wem.2017.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Tobin J.M., Nordmann G.R., Kuncir E.J. Resuscitation during critical care transportation in Afghanistan. J Spec Oper Med. 2015;15:72–75. doi: 10.55460/V3ZO-RG71. [DOI] [PubMed] [Google Scholar]
- 40.Byers R. An upshot of war - damage control resuscitation. Int Emerg Nurs. 2010;18:221–225. doi: 10.1016/j.ienj.2010.01.002. [DOI] [PubMed] [Google Scholar]