CN111768818A - Artificial intelligence-based emergency aid decision-making system and method for pelvic fractures - Google Patents

Abstract

The invention relates to an artificial intelligence-based pelvis fracture emergency aid assistant decision-making system and a method, which comprises the following steps: the data layer comprises an intelligent assistant decision-making knowledge base for first aid of pelvic fracture, the intelligent assistant decision-making knowledge base comprises historical case data and pelvic fracture diagnosis and treatment literature data, the data layer imports basic information of a patient with pelvic fracture emergency treatment, creates data information of the patient with pelvic fracture emergency treatment, collects vital signs, tests and examination results and clinical treatment measures so as to track disease changes of the patient in real time; the processing layer intelligently learns related knowledge and cases of the pelvis fracture emergency treatment, extracts related literature data to expand an intelligent assistant decision-making knowledge base, performs image recognition on the imaging of a patient with the pelvis fracture emergency treatment, and performs semantic analysis and matching on the data analyzed by the patient with the pelvis fracture emergency treatment and the intelligent assistant decision-making knowledge base to extract decision-making suggestions; the application layer displays relevant illness state data of the pelvic fracture emergency treatment patient, emergency treatment intelligent aid decision suggestions and actual treatment measures, and the summarized data are summarized in the data layer.

Description

Artificial intelligence-based emergency aid decision-making system and method for pelvic fractures

technical field

The invention relates to the technical field of trauma emergency aid, in particular to an artificial intelligence-based emergency aid decision-making system and method for pelvic fractures.

Background technique

The incidence of pelvic fractures is 0.3% to 8% of all fractures, but the mortality rate of pelvic fractures is close to 16%, and the mortality rate is 27% in patients with closed pelvic fractures that are hemodynamically unstable, and even more in patients with open pelvic fractures. As high as more than 55%, it was dubbed the horror title of "The Killing Fracture". Despite the improvement of trauma emergency procedures and measures, pelvic fractures still have the highest mortality rate of all skeletal traumas, and bleeding is the main reversible cause of death in approximately 42% of patients with pelvic fractures. The sources of bleeding mainly come from four aspects: oozing blood from pelvic fractures, combined thoracic and abdominal organ damage, pelvic venous hemorrhage, and arterial injury blood loss. The most important aspect of first aid for a pelvic fracture is early diagnosis of bleeding. Due to the unclear main complaints of some patients, the difficulty of physical examination, the risk of handling during imaging examination, and the insufficient understanding of the injury mechanism by physicians, the classification of pelvic fractures is unclear, and the diagnosis of combined injuries is delayed or even missed. Different fracture types are prone to different vascular injuries. For example, posterior dislocation of the pelvic ring (open book fracture) usually causes damage to the internal iliac artery and its branches; Butterfly fractures tend to involve the lower pudendal vessels, and forward pressure may damage the external iliac or femoral vessels. artery. For bleeding caused by venous injury at the fracture site, external fixation and pelvic tamponade can be effectively intervened to reduce pelvic volume and stabilize the fracture. However, tamponade is not sufficient to block arterial hemorrhage. Arterial injuries, however, tend to be the most lethal, usually in deeper small vessels rather than well-known arteries. Angiographic embolization is an effective means of controlling arterial bleeding, but this treatment relies on a professional vascular team and requires time-consuming preparation. The World Society of Emergency Surgery recommends pelvic tamponade as a first-line rescue method, but many guidelines disagree with this. Different emergency strategies are proposed for hemodynamically unstable pelvic fractures, and new updates are made every year. Although strategies are not uniform, timing is believed to be the key to successful intervention and improved survival. However, emergency physicians often rely on their own experience to judge the condition, how to comprehensively assess the condition of patients with pelvic fractures, seize the opportunity to determine the source of bleeding as soon as possible, and treat scientifically, and adopt correct treatment strategies at the right time based on evidence-based medicine to improve the survival rate of patients , is a huge challenge for trauma emergency physicians to be solved urgently.

Therefore, it is necessary to invent an artificial intelligence-based emergency aid decision-making platform for pelvic fractures to assist physicians in quickly making scientific predictions, precise treatment, and improving prognosis, and to solve the above-mentioned technical problems in emergency treatment of patients with pelvic fractures.

SUMMARY OF THE INVENTION

Aiming at the problems and deficiencies existing in the prior art, the present invention provides an artificial intelligence-based emergency aid decision-making system and method for pelvic fractures.

The present invention solves the above-mentioned technical problems through the following technical solutions:

The invention provides an artificial intelligence-based emergency aid decision-making system for pelvic fractures, which is characterized in that it includes a data layer, a processing layer and an application layer;

The data layer includes a pelvic fracture first aid intelligent decision-making knowledge base, and the intelligent decision-making knowledge base includes historical case data and intelligently retrieved pelvic fracture diagnosis and treatment literature data. The data layer is used to import basic information of pelvic fracture first aid patients and create a pelvic fracture. Data information of fracture emergency patients, collection of vital signs, inspection and examination results, and clinical treatment measures to track patient condition changes in real time;

The processing layer is used to intelligently learn the knowledge and cases related to pelvic fracture first aid, extract relevant literature data to expand the intelligent auxiliary decision-making knowledge base, perform image recognition on the imaging of pelvic fracture first aid patients, and compare the data identified and analyzed by the pelvic fracture first aid patients with the data. The intelligent auxiliary decision-making knowledge base performs semantic analysis and matching, and extracts decision-making suggestions and displays them in the application layer;

The application layer is used to display relevant condition data of emergency patients with pelvic fractures, emergency intelligent auxiliary decision-making suggestions and actual treatment measures, and the aggregated data is summarized in the data layer.

Preferably, the data layer is used to import the basic information of emergency patients with post-traumatic hip pain considering pelvic fractures, and collect the chief complaints, current medical history, past history, personal history, dynamic changes of vital signs, and inspections of emergency patients with pelvic fractures. results and clinical measures.

Preferably, the processing layer is used to perform accurate Tile classification, Young-Burgess classification, and AO classification for pelvic fractures by performing image recognition on the first-aid patients with pelvic fractures based on the collected information of first-aid patients with pelvic fractures. The trauma score, trauma severity score, brief injury score, modified Glasgow coma score, and shock index score were obtained by analyzing the current condition data of emergency patients with pelvic fracture;

Based on the basic information of emergency patients with pelvic fracture, fracture classification, trauma score, trauma severity score, brief injury score, modified Glasgow coma score, shock index score, and diagnosis and treatment intelligent auxiliary decision-making knowledge base Semantic analysis and matching are performed to obtain emergency intelligent decision-making suggestions for the emergency patient with pelvic fracture.

Preferably, the processing layer is used to perform trauma severity scoring using a trauma severity scoring algorithm, using abbreviated injury scoring algorithm to perform abbreviated injury scoring, using an improved Glasgow coma scoring algorithm to perform an improved Glasgow coma scoring algorithm, and using a shock index algorithm to perform shock. Index score.

The present invention also provides an artificial intelligence-based emergency aid decision-making method for pelvic fractures, which is characterized in that it includes the following steps:

S1. Import the basic information of emergency patients with pelvic fractures, and collect the chief complaints, current medical history, past history, personal history, dynamic changes of vital signs, inspection results and clinical treatment measures of emergency patients with pelvic fractures;

S2. Based on the collected information of emergency patients with pelvic fractures, by performing image recognition on the imaging of emergency patients with pelvic fractures, accurate Tile classification, Young-Burgess classification, and AO classification are performed for pelvic fractures. By analyzing the first aid for pelvic fractures The patient's current condition data was used for trauma score, trauma severity score, brief injury score, modified Glasgow coma score, and shock index score;

S3. Based on the basic information, fracture classification, trauma score, trauma severity score, brief injury score, modified Glasgow coma score, shock index score of emergency patients with pelvic fracture, and intelligent diagnosis and treatment assistance The decision knowledge base is used for semantic analysis and matching, so as to obtain the first aid intelligent auxiliary decision-making suggestions for the emergency patient with pelvic fracture;

S4. Display the first aid intelligent auxiliary decision-making suggestion and collected patient information for the first aid patient with pelvic fracture.

Preferably, the trauma severity score algorithm is used for the trauma severity score, the abbreviated injury score algorithm is used for the abbreviated injury score, the modified Glasgow coma score algorithm is used for the modified Glasgow coma score, and the shock index algorithm is used for the shock index score.

On the basis of conforming to common knowledge in the art, the above preferred conditions can be combined arbitrarily to obtain preferred examples of the present invention.

The positive progressive effect of the present invention is:

1. Quickly collect patient information on the same platform, greatly shorten the emergency time for patients with pelvic fractures, and improve the success rate of emergency treatment;

2. Instruct clinicians to carry out first aid for patients with pelvic fractures, avoid missed diagnosis to the greatest extent, accurately determine the source of bleeding, and grasp the timing of treatment to select appropriate first aid measures;

3. Continuous learning of artificial intelligence, providing advanced and evidence-based medical first aid measures with sufficient evidence, significantly improving the survival rate of patients;

4. Provide a wealth of case materials for clinical research, optimize the emergency procedures for pelvic fractures, and significantly improve the experience of clinicians.

Description of drawings

FIG. 1 is a structural block diagram of an artificial intelligence-based emergency aid decision-making system for pelvic fractures according to a preferred embodiment of the present invention.

FIG. 2 is a flowchart of an artificial intelligence-based first aid decision-making method for pelvic fractures according to a preferred embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

Example 1

As shown in FIG. 1 , this embodiment provides an artificial intelligence-based pelvic fracture emergency aid decision-making system, including a data layer 1, a processing layer 2, and an application layer 3 of the artificial intelligence-based pelvic fracture emergency aid decision-making platform (see FIG. 1 ). ).

In the present invention, data layer 1 includes patient data collection, pelvic fracture first aid intelligent auxiliary decision-making knowledge base, and the knowledge base includes historical case data and intelligently retrieved pelvic fracture diagnosis and treatment literature data. Data collection: Import in-hospital clinical data such as 120 Emergency Center, HIS/LIS/RIS, etc. through manual input or through API interface, import basic information of admitted patients, create patient data information, collect vital signs, inspection and inspection results, clinical treatment measures, etc. Real-time tracking of patient condition changes, modularized and electronic input to improve efficiency and avoid missed diagnosis. The processed trauma patient information data is stored in the intelligent auxiliary decision-making knowledge base in the form of historical cases; the pelvic fracture emergency-related documents intelligently screened by the processing layer are stored in the knowledge base for reference by clinicians.

In the present invention, the intelligent learning of processing layer 2 has machine learning and deep learning capabilities. Through intelligent learning of pelvic fracture first aid related knowledge and cases, relevant literature data is extracted to expand the intelligent auxiliary decision-making knowledge base, and the knowledge base is regularly updated to provide advanced evidence-based medical evidence. Sufficient first aid strategy suggestions; through image recognition of patient imaging, accurate Tile classification, Young-Burgess classification, AO classification, etc. for pelvic fractures, trauma score and trauma severity score by analyzing the patient's current condition data , Concise Injury Score, Improved Glasgow Coma Score, Shock Index, and semantically analyze and match patient data with the diagnosis and treatment intelligent decision-making knowledge base to provide decision-making suggestions for judging prognosis, inspection, examination, treatment, consultation, and transfer, and display them in Application layer 3, accurately determine the source of bleeding, grasp the timing of treatment and choose appropriate first aid measures.

The processing layer is used to perform trauma severity scoring using the trauma severity scoring algorithm, using the concise injury scoring algorithm to perform the concise injury scoring, using the modified Glasgow coma scoring algorithm to perform the modified Glasgow coma scoring, and using the shock index algorithm to perform the shock index scoring.

In the present invention, the application layer 3 is used to display the patient's real-time condition, intelligently assist decision-making suggestions and actual treatment measures, process, summarize, and store the data in the data layer to form historical cases.

Example 2

As shown in FIG. 2 , this embodiment provides an artificial intelligence-based first aid decision-making method for pelvic fractures, including the following steps:

Step 101 , import the basic information of the emergency pelvic fracture patient, and collect the chief complaint, current illness history, past history, personal history, dynamic changes of vital signs, inspection results and clinical treatment measures of the emergency pelvic fracture patient.

Step 102 , based on the collected information of emergency patients with pelvic fractures, perform image recognition on the imaging of emergency patients with pelvic fractures, perform accurate Tile classification, Young-Burgess classification, and AO classification for pelvic fractures, and analyze the pelvic fractures by analyzing the pelvic fractures. The current condition data of emergency patients were evaluated for trauma score, trauma severity score, brief injury score, modified Glasgow coma score, and shock index score.

Step 103, based on the basic information, fracture classification, trauma score, trauma severity score, brief injury score, modified Glasgow coma score, shock index score of the emergency patient with pelvic fracture, and diagnosis and treatment intelligence The auxiliary decision-making knowledge base is used for semantic analysis and matching to obtain emergency intelligent auxiliary decision-making suggestions for the emergency patient with pelvic fracture.

Based on decision-making suggestions, clinicians make clinical treatment in a timely manner. After treatment, the vital signs and inspection results of patients with pelvic fractures change dynamically. The processing layer of the artificial intelligence-based pelvic fracture emergency aid decision-making platform continues to assist clinical decision-making until the disease is resolved. : improvement - general ward admission; deterioration - intensive care treatment, transfer, death; operating room.

Step 104 , displaying the first aid intelligent decision-making advice and collected patient information for the first aid patient with the pelvic fracture.

Although specific embodiments of the present invention have been described above, those skilled in the art will understand that these are merely illustrative and the scope of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principle and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims (6)

1. An artificial intelligence-based pelvis fracture emergency aid assistant decision-making system is characterized by comprising a data layer, a processing layer and an application layer;

the data layer comprises a pelvis fracture emergency treatment intelligent aid decision-making knowledge base, the intelligent aid decision-making knowledge base comprises historical case data and intelligently retrieved pelvis fracture diagnosis and treatment literature data, and the data layer is used for importing basic information of a pelvis fracture emergency treatment patient, creating data information of the pelvis fracture emergency treatment patient, collecting vital signs, testing and checking results and carrying out clinical treatment measures to track disease state changes of the patient in real time;

the processing layer is used for intelligently learning related knowledge and cases of the pelvis fracture emergency treatment, extracting related literature data to expand an intelligent aid decision-making knowledge base, carrying out image recognition on the imaging of a patient with the pelvis fracture emergency treatment, carrying out semantic analysis and matching on the data analyzed by the patient with the pelvis fracture emergency treatment and the intelligent aid decision-making knowledge base, and extracting decision-making suggestions to be displayed on the application layer;

the application layer is used for displaying relevant illness state data, emergency treatment intelligent aid decision suggestions and actual treatment measures of the pelvic fracture emergency treatment patient, and the summarized data are summarized to the data layer.

2. The artificial intelligence based pelvic fracture aid decision making system according to claim 1, wherein the data layer is used for introducing post-traumatic hip pain and collecting chief complaints, current medical history, past medical history, personal history, dynamically changing vital signs, examination results and clinical treatment measures of the pelvic fracture aid patient in consideration of basic information of the pelvic fracture aid patient.

3. The artificial intelligence based pelvis fracture emergency aid decision making system according to claim 1, wherein the processing layer is used for carrying out accurate Tile typing, Young-Burgess typing and AO typing on pelvis fracture by carrying out image recognition on the imaging of the pelvis fracture emergency aid patient based on the collected information of the pelvis fracture emergency aid patient, carrying out trauma scoring, trauma severity scoring, concise damage scoring, improved Glasgow coma scoring and shock index scoring by analyzing the current disease data of the pelvis fracture emergency aid patient;

based on the basic information, the fracture type, the trauma score, the trauma severity score, the concise damage score, the improved Glasgow coma score and the shock index score of the pelvis fracture emergency patient, semantic analysis and matching are carried out on the pelvis fracture emergency patient and the diagnosis and treatment intelligent auxiliary decision knowledge base, so that the first-aid intelligent auxiliary decision suggestion for the pelvis fracture emergency patient is obtained.

4. The artificial intelligence based pelvic fracture first aid decision making system of claim 3, wherein the processing layer is configured to perform the trauma severity scoring using a trauma severity scoring algorithm, the concise damage scoring using a concise damage scoring algorithm, the modified Glasgow coma scoring using a modified Glasgow coma scoring algorithm, and the shock index scoring using a shock index algorithm.

5. An artificial intelligence-based pelvis fracture emergency aid assistant decision-making method is characterized by comprising the following steps:

s1, importing basic information of the pelvis fracture emergency treatment patient, and collecting chief complaints, current medical history, past medical history, personal history, dynamically-changed vital signs, inspection results and clinical treatment measures of the pelvis fracture emergency treatment patient;

s2, based on the collected information of the pelvis fracture emergency patients, carrying out image recognition on the iconography of the pelvis fracture emergency patients, carrying out accurate Tile typing, Young-Burgess typing and AO typing on the pelvis fracture, and carrying out trauma scoring, trauma severity scoring, concise damage scoring, improved Glasgow coma scoring and shock index scoring by analyzing the current disease data of the pelvis fracture emergency patients;

s3, performing semantic analysis and matching with a diagnosis and treatment intelligent auxiliary decision knowledge base based on the basic information, the fracture type, the trauma score, the trauma severity score, the brief damage score, the improved Glasgow coma score and the shock index score of the pelvis fracture emergency patient to obtain an emergency intelligent auxiliary decision suggestion for the pelvis fracture emergency patient;

and S4, displaying the emergency intelligent aid decision-making suggestion and the collected patient information for the patient with the pelvis fracture emergency treatment.

6. The artificial intelligence-based pelvis fracture emergency aid decision-making method according to claim 5, wherein a wound severity scoring algorithm is adopted for scoring the wound severity, a concise damage scoring algorithm is adopted for scoring the concise damage, an improved Glasgow coma scoring algorithm is adopted for scoring the improved Glasgow coma, and a shock index algorithm is adopted for scoring the shock index.

Images