Integrated emergency stroke unit having regions

Abstract

The present application relates to the field of medicine, and discloses an integrated emergency stroke unit having regions. The integrated emergency stroke unit having regions comprises an information interaction area, an imaging examination area, and a thrombolysis treatment area. The information interaction area is provided with a doctor workstation, a comprehensive information monitoring display, and a process-support AI assistant. The imaging examination area is provided with a detection device, a first television display screen, a low-field movable magnetic resonance imager, and an AI-assisted decision-making system. The process-support AI assistant performs automated consultation and assists in nervous system scale scoring by means of voice interaction or touch screen input interaction, and also provides retrieval support for a real-time knowledge base and clinical guidelines. The AI-assisted decision-making system assists in decision-making on the basis of an automated consultation result, a scale scoring result, a detection result, and a head and neck magnetic resonance imaging result, generates a treatment plan, and displays the treatment plan. According to the present application, all diagnosis and treatment activities involved upon the arrival of a patient suffering from acute ischemic stroke at a hospital are integrated into a single space, which can significantly shorten diagnosis and treatment time.

Description

A regional integrated emergency stroke unit

[0001] This application claims priority to Chinese Patent Application No. 202410993951.7, filed on July 24, 2024, entitled "A Regionally Integrated Emergency Stroke Unit", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of medical technology, and in particular to a regionally integrated emergency stroke unit. Background Technology

[0003] Currently, approximately 70% of known stroke cases are ischemic strokes, primarily caused by thrombosis or embolism within the responsible blood vessel. Numerous studies have shown that the earlier the thrombus is dissolved or removed, and the more reperfusion of ischemic brain tissue is restored, the more brain tissue can be saved, leading to a better clinical prognosis. However, only 2.17% of patients with acute ischemic stroke receive endovascular treatment within 6 hours. In response, some scholars have proposed significantly reducing pre-hospital and in-hospital delays in stroke patients by establishing stroke centers and promoting mobile stroke units, among other organized stroke care systems.

[0004] Previous studies suggest that hospitals treating stroke patients should establish stroke units as much as possible, and all patients with acute ischemic stroke should be admitted to stroke units for treatment as early and as far as possible. For in-hospital delays in acute ischemic stroke, organized management is emphasized. However, in stroke units, there is still a common situation where patients need to travel back and forth between the examination room, laboratory, imaging room and treatment room, which delays valuable treatment time and there is room for further optimization. Summary of the Invention

[0005] The purpose of this application is to provide a regionally integrated emergency stroke unit that concentrates all diagnostic and treatment activities involved in acute ischemic stroke patients after arrival at the hospital in one space, so as to significantly shorten the time for diagnosis and treatment.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] This application provides a regional integrated emergency stroke unit, including: an information interaction area, an imaging examination area, and a thrombolytic therapy area;

[0008] The information interaction area is equipped with a doctor's workstation, a comprehensive information monitoring display, and a workflow support AI assistant. The doctor's workstation includes a computer, a file rack, and a folding table. The computer is used for doctors to input information; the file rack is used to store documents; and the folding table is used for temporary placement and storage of medical equipment and documents. The comprehensive information monitoring display shows the patient's basic information, arrival time, arrival time at the consultation room, and vital signs. The workflow support AI assistant is used for automatic consultation and assisting in neurological scale scoring through voice interaction or touchscreen input, and also provides real-time knowledge base and clinical guideline retrieval support.

[0009] The imaging examination area is equipped with testing equipment, a first television display screen, a low-field portable magnetic resonance imaging (MRI) scanner, and an AI-assisted decision-making system. The testing equipment is used for electrocardiogram (ECG) examinations, pharmacogenomics testing, and monitoring of the patient's blood oxygen saturation, blood pressure, and peripheral blood glucose levels to obtain test results. The first television display screen is used to magnify and display the content of the comprehensive information monitoring display. The low-field portable MRI scanner is used to perform head and neck MRI on the patient. The AI-assisted decision-making system is communicatively connected to the comprehensive information monitoring display, the process support AI assistant, the testing equipment, the first television display screen, and the low-field portable MRI scanner. The AI-assisted decision-making system is used to assist in decision-making based on automatic consultation results, scale scoring results, test results, and head and neck MRI results, generating and displaying a treatment plan.

[0010] The thrombolysis treatment area is equipped with a small refrigerator, an emergency medicine cabinet, and a storage cabinet for storing medicines, reagents, and emergency medicines that require different temperatures. The thrombolysis treatment area is also equipped with an ambulance and a vital signs monitoring system. The ambulance is used to rescue patients in case of emergency medical conditions. The vital signs monitoring system is used to monitor and display the patient's vital signs in real time.

[0011] In an exemplary embodiment, the process supports the inclusion of an intelligent interaction module within the AI ​​assistant; the intelligent interaction module, based on the emergency structured medical record, asks structured questions about the patient's onset time, symptoms, accompanying symptoms, past medical history, and allergy history, and performs speech recognition on the patient's and / or their family's answers and performs structured input of key medical information.

[0012] In one exemplary embodiment, the intelligent interaction module is further configured to perform NIHSS (National Institutes of Health and Human Services) scores, ABCD2 scores, and pre-illness mRS scores through voice interaction or touchscreen typing interaction with the patient and / or their family members, thereby obtaining scale score results.

[0013] In an exemplary embodiment, the intelligent interaction module also stores a large language model; the large language model is equipped with a knowledge base related to stroke clinical diagnosis and treatment, including clinical scales, guidelines for the acute phase of cerebral infarction, guidelines for secondary prevention of cerebral infarction, guidelines for the diagnosis and treatment of cerebral hemorrhage, and guidelines for the clinical diagnosis and treatment of subarachnoid hemorrhage; during the diagnosis and treatment process, the knowledge base and clinical guidelines can be retrieved through voice interaction with the doctor or touch screen typing interaction.

[0014] In one exemplary embodiment, the intelligent interaction module is also used to provide explanations of medical procedures and treatment plans, as well as to obtain informed consent through voice interaction or touchscreen typing interaction with the patient and / or their family members.

[0015] In one exemplary embodiment, the detection device includes an electrocardiograph, a wearable electrocardiogram monitor, a fingertip pulse oximeter, a blood pressure monitor, a blood glucose meter, and a non-invasive rapid genotyping device.

[0016] In one exemplary embodiment, the low-field movable magnetic resonance imaging system integrates a mobile magnetic resonance imaging system; the mobile magnetic resonance imaging system uses a deep learning-based image reconstruction method to achieve image super-resolution reconstruction using the patient's low-field and high-field MRI datasets to obtain head and neck magnetic resonance imaging results.

[0017] In one exemplary embodiment, the AI-assisted decision-making system is used to identify the location of acute cerebral infarction lesions in the head and neck magnetic resonance imaging results using deep learning methods and automatically calculate the infarct volume; the AI-assisted decision-making system is also used to determine whether the patient has intracranial large artery occlusion / stenosis by using the MRA sequence of moving magnetic resonance imaging; the AI-assisted decision-making system is also used to register different sequences to a standard space using automatic registration technology to achieve automated identification of DWI / FLAIR mismatch of acute cerebral infarction lesions, determine the approximate onset time of stroke after awakening and guide reperfusion therapy.

[0018] In an exemplary embodiment, the AI-assisted decision-making system is further used to combine image interpretation information with scale scoring results to determine whether the patient has a clinical-image mismatch, and, in conjunction with the infarct volume, to assist in deciding whether the patient needs endovascular treatment to achieve intracranial reperfusion.

[0019] In one exemplary embodiment, the thrombolysis treatment area is further provided with a second television display screen and a trash can; the second television display screen is used to display the content of the integrated information monitoring display screen; the trash can is used to temporarily place medical or domestic waste.

[0020] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0021] This application proposes a regionally integrated emergency stroke unit by streamlining the acute ischemic stroke treatment process. This unit integrates the information exchange area, imaging examination area, and thrombolysis treatment area into a single consultation room. In other words, it centralizes all diagnostic and treatment activities for acute ischemic stroke patients upon arrival at the hospital into one space. This spatial integration and organized system management of patient visits, examinations, and thrombolysis provides a one-stop solution for ischemic stroke treatment. It addresses the current problem of acute ischemic stroke patients facing dispersed triage, consultation, physical examination, laboratory tests, and other procedures across different emergency departments, involving multiple disciplines and significantly extending the time from arrival to treatment. This approach significantly shortens the patient's treatment time.

[0022] Instruction manual illustrations

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 is a schematic diagram of the three-dimensional model of the regional integrated emergency stroke unit of this application;

[0025] Figure 2 is a schematic diagram of the three-dimensional model of the information interaction area;

[0026] Figure 3 is a schematic diagram of the three-dimensional model of the imaging examination area;

[0027] Figure 4 is a schematic diagram of the three-dimensional model of the thrombolytic therapy area;

[0028] Figure 5 shows a list of items stored inside a small ambulance. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] The purpose of this application is to provide a regionally integrated emergency stroke unit that concentrates all diagnostic and treatment activities involved in acute ischemic stroke patients after arrival at the hospital in one space, so as to significantly shorten the time for diagnosis and treatment.

[0031] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] Referring to Figure 1, this application provides a regional integrated Emergency Stroke Unit (ESU), including: an information interaction area, an imaging examination area, a thrombolysis treatment area, and an entrance / exit. The overall size of the unit can be 8.3m × 4m.

[0033] Referring to Figure 2, the information interaction area is equipped with a doctor's workstation, a comprehensive information monitoring display 1, and a process support AI assistant 2.

[0034] The doctor's workstation specifically includes a computer, a file rack 3, and a folding table 4. The computer includes a host, dual-screen displays 5, and a keyboard 6, allowing doctors to input and view patient information. In some embodiments, each display in the dual-screen displays 5 can be 27 inches in size. The file rack 3 can be used to hold paper documents. The folding table 4 can be a wall-mounted / wall-mounted folding table for temporary placement and storage of small medical equipment, instruments, documents, etc. In this application, the symbol " / " represents "or".

[0035] The integrated information monitoring display 1 is used to display the patient's basic information, arrival time, arrival time in the examination room, and vital signs, and its size can be 50 inches. The process supports the AI ​​assistant 2 for automatic consultation and auxiliary neurological scale scoring through voice interaction or touch screen input interaction, and also provides real-time knowledge base and clinical guideline retrieval support.

[0036] Specifically, the process supports AI Assistant 2 as an AI intelligent robot. AI Assistant 2 is equipped with an intelligent interaction module. The automatic consultation function of the intelligent interaction module is based on the emergency room's structured medical records. It asks structured questions about the patient's onset time, symptoms, accompanying symptoms, past medical history, allergies, and other key information. The system then uses speech recognition to record the patient's and / or their family's answers, or the doctor records the information provided by the patient and / or their family, storing the results as a structured consultation. The subsequent AI-assisted decision-making system can then provide suggestions for further treatment based on the consultation results and imaging.

[0037] The intelligent interactive module's auxiliary neurological scale scoring function involves asking questions about key auxiliary neurological scales, such as the National Institutes of Health Stroke Scale (NIHSS), the ABCD2 (age, blood pressure, clinical features, duration, diabetes) score, and the pre-onset modified Rankin Scale (mRS) score. The system obtains scores by recording the patient's and / or their family's responses via voice recognition or by having the doctor input information provided by the patient and / or their family. The NIHSS score can assess the severity of neurological deficits and is a crucial basis for reperfusion therapy in ischemic cerebrovascular diseases (such as stroke). The ABCD2 score can stratify the risk of stroke in patients with transient ischemic attacks and serves as an important reference for antiplatelet aggregation therapy, helping doctors make further treatment decisions. The pre-onset mRS score assesses the patient's functional status before the onset of the disease and is an important basis for endovascular reperfusion therapy and other treatments for ischemic cerebrovascular diseases.

[0038] The retrieval support function of the intelligent interaction module is implemented by a large language model. This large language model has a knowledge base related to stroke clinical diagnosis and treatment, including but not limited to the aforementioned clinical scales, international / domestic guidelines for the acute phase of cerebral infarction, guidelines for secondary prevention of cerebral infarction, guidelines for the diagnosis and treatment of cerebral hemorrhage, and clinical guidelines for the diagnosis and treatment of subarachnoid hemorrhage. During the diagnosis and treatment process, when doctors need to retrieve relevant information, they can directly ask via voice, and the intelligent interaction module will provide real-time feedback. For example, when a doctor encounters a patient with cerebral infarction within the intravenous thrombolysis time window but has suspected contraindications to thrombolysis, they can ask the intelligent interaction module about the indications and contraindications for alteplase intravenous thrombolysis to quickly verify whether the patient meets the treatment indications and rapidly assist in decision-making.

[0039] The AI ​​assistant 2, supporting the aforementioned process, is also used for efficient interaction with medical staff, patients, and / or their families, providing functions such as explaining medical procedures, describing treatment plans, and signing informed consent forms. The medical procedures and treatment plans include indications, contraindications, potential risks, and benefits for treatments such as reperfusion therapy (intravenous thrombolysis and endovascular thrombectomy) and secondary prevention therapy (antiplatelet aggregation and lipid-lowering therapy for ischemic cerebrovascular disease). It can assist doctors in providing relevant data and guidelines to address patients' and their families' concerns regarding treatment plans, and help doctors with explanations. When doctors obtain informed consent from patients and their families, the assistant can provide voice reminders about the signing location and required information, and assist in explaining current medical insurance policies for relevant medications.

[0040] Referring to Figure 3, the imaging examination area is equipped with a detection device 7, a first television display screen 8, a low-field movable magnetic resonance imaging instrument 9, and an AI-assisted decision-making system, etc.

[0041] The testing equipment 7 includes an electrocardiograph, a wearable electrocardiogram monitor, a fingertip pulse oximeter, a blood pressure monitor, and a blood glucose meter, used for electrocardiogram examinations and monitoring patients' blood oxygen saturation, blood pressure, and peripheral blood glucose levels. The testing equipment also includes a non-invasive rapid genotyping instrument for genes such as CYP2C19, used for pharmacogenomics testing to obtain results.

[0042] The first television display screen 8 is a large-size television display screen of 60 inches or more, used to magnify the display content of the integrated information monitoring display 1 in the information interaction area.

[0043] The imaging examination area is equipped with a low-field portable magnetic resonance imaging (MRI) scanner, which can perform MRI scans on patients and generate head and neck MRI results. This application utilizes a portable, non-magnetically shielded, and easy-to-use low-field (0.23T) MRI scanner for rapid imaging evaluation of acute stroke patients, differentiating between cerebral infarction and cerebral hemorrhage within 10 minutes, while simultaneously assessing the condition of major blood vessels. This addresses the problems of small aperture, relatively high incidence of claustrophobia, and inability of large-sized patients to undergo MRI examinations with mobile head and neck MRI systems. It also overcomes the shortcomings of high-field MRI, such as poor accessibility and long scan times, which prevent its application in neuroimaging assessment of acute stroke. Furthermore, it breaks through the bottleneck of requiring stroke patients to undergo head CT scans in a separate CT room, improving diagnostic accuracy and significantly shortening the time from arrival at the hospital to treatment.

[0044] This application's low-field mobile magnetic resonance imaging system (MRI) integrates a mobile MRI imaging system. Utilizing artificial intelligence image analysis technology, the mobile MRI imaging system significantly reduces detection time while improving image quality. It assists physicians in differentiating between ischemia and hemorrhage, identifying the location and volume of the core infarct, recognizing large artery occlusion and stenosis, and determining the duration of stroke in awake patients, enabling rapid treatment decisions. Specifically, the mobile MRI imaging system achieves super-resolution image reconstruction using a deep learning-based image reconstruction method and a large dataset of low-field and high-field MRI data from the same patient. This significantly improves the detail and contrast of 0.23T MRI images, enhancing image quality and shortening scan time.

[0045] The AI-assisted decision-making system is communicatively connected to the integrated information monitoring display 1, the process support AI assistant 2, the detection device 7, the first television display screen 8, and the low-field portable magnetic resonance imaging (MRI) unit 9. The AI-assisted decision-making system is used to assist in decision-making based on automatic consultation results, scale scoring results, detection results, and head and neck MRI results, generating and displaying treatment plans.

[0046] Specifically, the AI-assisted decision-making system utilizes deep learning methods and a large amount of manually labeled datasets to quickly and accurately identify the location of acute ischemic stroke lesions and automatically calculate their volume. The system also assists doctors in interpreting images using moving magnetic resonance imaging (MRA) sequences, quickly determining whether the patient has intracranial large artery occlusion / stenosis. Furthermore, the system employs automatic registration technology to register different sequences (DWI and FLAIR) to a standard space, achieving automated identification of DWI / FLAIR mismatches in acute ischemic stroke lesions, determining the approximate onset time of stroke after awakening, and guiding reperfusion therapy. By combining the above image interpretation information with the NIHSS score obtained from the intelligent interaction module, the system assists doctors in determining whether the patient has a clinical-image mismatch. Combined with the infarct volume, this further assists doctors in deciding whether the patient needs endovascular treatment to achieve intracranial reperfusion.

[0047] Referring to Figure 4, the thrombolytic therapy area is equipped with a small refrigerator 10, an emergency medicine cabinet 11, and a storage cabinet 12 for storing medicines, reagents, and emergency medications requiring different temperatures. The small refrigerator 10 is equipped with a temperature display and setting device to monitor the storage temperature of medicines in real time; its dimensions can be designed to be 0.5m × 0.5m × 0.8m. The emergency medicine cabinet 11 can be designed to be 1.4m × 0.6m × 1.7m. For example, tenecteplase needs to be stored below 25°C and protected from light, and should be used immediately after dissolution; if not used immediately, it should be refrigerated at 2–8°C and used within 24 hours. Alteplase also needs to be stored below 25°C and protected from light, and should be used immediately after dissolution; if not used immediately, it should be refrigerated at 2–8°C and used within 24 hours.

[0048] Furthermore, the thrombolysis treatment area is equipped with a small ambulance 13 for emergency medical care in case of a patient's medical emergency. The small ambulance 13 can be designed to be 0.7m × 0.5m × 1.5m in size. The items stored inside the small ambulance 13 are shown in Figure 5; the medications contained therein should be stored below 20°C, protected from light, and in a sealed container. The thrombolysis treatment area is also equipped with a 42-inch second television display screen 14, which can display the content of the integrated information monitoring display screen 1 in the information interaction area. The thrombolysis treatment area is also equipped with medical and domestic waste bins 15 for temporary placement of medical or domestic waste. The waste bins 15 can be designed to be 0.45m × 0.3m × 0.3m in size. The thrombolysis treatment area is also equipped with vital sign monitoring systems such as electrocardiogram and blood pressure monitors, as well as gene testing equipment; the corresponding parameters can be displayed on the television displays / monitors in each area. A camera can also be installed in the thrombolysis treatment area to work with the second television display screen to achieve online doctor-patient interaction and remote monitoring functions.

[0049] The thrombolytic therapy area is also equipped with next-generation thrombolytic drugs and other essential medications, allowing for rapid confirmation of medication regimens through pharmacogenomics testing. Based on a non-invasive rapid genotyping instrument using drug metabolism genes such as CYP2C19, doctors or nurses collect exfoliated oral mucosal cells from patients using oral swabs to detect the genotypes of genes such as CYP2C19 (a key gene for clopidogrel metabolism), CYP2C9 (a key gene for warfarin metabolism), and VKORC (a key gene for warfarin metabolism). The results are automatically transmitted to an AI-assisted decision-making system, which provides preliminary assessments of the effectiveness and safety of various medications based on the genotyping of key drug metabolism genes, and offers treatment recommendations. For example, patients carrying the *2 or *3 alleles of the CYP2C19 gene who require dual antiplatelet therapy may have clopidogrel metabolized as a medium or slow metabolizer, affecting its effectiveness. In the absence of contraindications to ticagrelor, ticagrelor may need to be substituted.

[0050] This application presents a unique integrated emergency stroke unit (ESU) for acute stroke treatment. It highly centralizes traditional organized management of acute stroke into a single space, incorporating innovative solutions to create three main areas: an information interaction area, an imaging examination area, and a thrombolysis treatment area. This is expected to reduce the time from patient arrival to reperfusion therapy to less than 20 minutes, providing a superior medical experience for medical staff, patients, and their families. Upon arrival, patients first enter the imaging examination area of ​​the integrated ESU. This area is equipped with intelligent wearable devices that can collect critical information such as vital signs in real time. It also features a new low-field mobile magnetic resonance imaging (MRI) scanner, which is small, lightweight, portable, and requires minimal metal implants, making it safer. The MRI interface is simple, with a modular scanning setup that is easy to learn and requires minimal space, eliminating the need for patients to travel between the examination room, scanning room, laboratory, and treatment room – a significant improvement over traditional CT scans. The AI-assisted decision-making system incorporates an innovative identification sequence that can rapidly differentiate between cerebral hemorrhage and cerebral ischemia within one and a half minutes. This allows patients with acute ischemic stroke to receive reperfusion therapy in the shortest possible time, saving brain function, reducing disability rates, and improving long-term quality of life. Through AI image high-definition processing technology, the resolution of brain images can be further improved, comparable to high-field MRI images. The information interaction area uses an integrated information system to achieve rapid input of fully structured data. Based on vital signs and imaging findings, AI-assisted decision-making is implemented, displaying key data in real time throughout the process to assist the medical team in making rapid decisions. The information interaction area is also equipped with an AI intelligent robot that provides explanations of medical procedures, treatment plans, and informed consent signing, improving communication efficiency for medical staff. The thrombolysis treatment area supports online doctor-patient interaction and remote monitoring, and is equipped with next-generation thrombolytic drugs and advanced thrombectomy devices. Bedside gene testing equipment can quickly confirm antiplatelet therapy plans, maximizing clinical benefits for patients.

[0051] This application's integrated emergency stroke unit utilizes an intelligent interactive module for rapid input of fully structured information / data. It also employs an AI-assisted decision-making system based on vital signs and imaging findings to support decision-making. Furthermore, it displays key medical data in real-time on television screens / monitors throughout the entire process. Through seamless integration of the AI-assisted decision-making system with medical equipment, key medical data is automatically collected and updated, reflecting the patient's condition in real time. This key medical data includes: test results from wearable devices or point-of-care devices, such as electrocardiographs, rapid blood glucose meters, blood pressure monitors, and non-invasive rapid genotyping instruments for genes like CYP2C19. The process involves: doctors or nurses connecting chest and limb leads to the patient to collect an electrocardiogram; doctors or nurses collecting capillary blood to rapidly test blood glucose; doctors or nurses using a cuff to collect blood pressure; and doctors or nurses collecting oral mucosal exfoliated cells for rapid CYP2C19 and other gene genotyping.

[0052] The fully structured information / data is collected through structured questioning (multiple-choice questions) using detection equipment and intelligent interactive modules. This includes: test results from wearable devices or point-of-care devices, such as electrocardiographs, rapid blood glucose meters, blood pressure monitors, and CYP2C19 rapid genotyping analyzers; the patient's current symptoms (limb weakness, slurred speech, blurred vision, etc.); physical signs (whether there is muscle weakness, visual field defects, dysarthria, aphasia, etc.); and past medical history (whether there is a history of hypertension, diabetes, etc.). Doctors, with the assistance of the intelligent interactive module, collect patient medical history and record the patient's voice descriptions during physical examinations, enabling rapid input of fully structured information / data.

[0053] Physical signs refer to information obtained through physical examination. Imaging findings include the presence of intracranial ischemic lesions, intracranial hemorrhage lesions, DWI / FLAIR image mismatch, and intracranial large artery occlusion. An AI-assisted decision-making system is used for automatic etiological classification of cerebrovascular diseases in clinical diagnosis and treatment, providing suggestions for further examination and treatment to assist doctors in decision-making.

[0054] This application utilizes artificial intelligence technology to construct an integrated medical AI-assisted decision-making system, enabling doctors to obtain analysis results within minutes and quickly determine treatment plans. The analysis results include: the patient's preliminary diagnosis (based on symptoms, signs, and imaging findings), etiological classification (based on signs, medical history, and imaging findings), presence of imaging mismatch (based on imaging findings), presence of clinical-imaging mismatch (based on signs, scale scores, and imaging findings), and presence of intracranial major artery occlusion (based on imaging findings).

[0055] The AI-assisted decision-making system is also used to provide treatment recommendations for various acute-phase treatments in the clinical diagnosis and treatment of cerebrovascular diseases, especially the matching of indications and contraindications for reperfusion therapy.

[0056] The AI-assisted decision-making system can automatically remind doctors of patients who meet the criteria for thrombolysis, immediately administering rtPA or TNK treatment and continuously monitoring vital signs. Patients meeting the criteria for intravenous thrombolysis should immediately enter the informed consent process. After obtaining informed consent from the patient and their family, intravenous thrombolysis treatment under vital sign monitoring should be administered immediately. Patients meeting the criteria for endovascular treatment should immediately enter the endovascular treatment procedure description process. Simultaneously, the intelligent interactive module will immediately call the interventional radiology physician on duty. After the physician arrives, it will assist the physician in obtaining informed consent from the patient and their family before transferring the patient to the catheterization lab for endovascular treatment.

[0057] This application innovatively integrates the emergency room, laboratory, imaging room, and pharmacy into a single space, proposing an integrated acute stroke unit concept. By combining the information exchange area, imaging examination area, and thrombolysis treatment area, the treatment route for acute ischemic stroke patients is highly centralized, minimizing delays upon arrival at the hospital. Of course, in practical applications, the regional integration of the acute stroke unit can be expanded or reduced according to needs.

[0058] The regionally integrated emergency stroke unit proposed in this application integrates three major areas: an information interaction area, an imaging examination area, and a thrombolytic therapy area. It integrates functions such as real-time vital sign monitoring, intelligent consultation, neurological examination and scoring, mobile MRI imaging, and rapid pharmacogenetic testing, creating an innovative and optimized in-hospital diagnosis and treatment system for acute ischemic stroke. Specifically, this application's ESU uses an integrated concept of an acute stroke unit, integrating all aspects of acute ischemic stroke patient triage, consultation, physical examination, laboratory tests, examinations, diagnosis, treatment plan formulation, and treatment. Centered on the patient, it significantly reduces the patient's movement after arriving at the emergency room. Suspected acute stroke patients entering the ESU first arrive at the information interaction area, where the process support AI assistant 2 performs information entry, consultation, neurological examination, and scale assessment. They then proceed to the imaging examination area, where doctors or nurses complete electrocardiogram examinations, measure blood pressure, blood oxygen saturation, and peripheral blood glucose. This information is entered into the process support AI assistant 2 in the information interaction area via voice or touchscreen and displayed on the television screens / monitors in each area. After the doctor confirms that the patient has no contraindications for low-field movable MRI, the doctor performs a low-field movable MRI scan, reviewing the images in real time during the scan to make a preliminary assessment of the patient's condition. For patients who meet the indications for intravenous thrombolysis, the doctor obtains preliminary informed consent from the patient's legal representative (family member, etc.). If the patient is conscious, after the scan, the doctor explains the risks and benefits of intravenous thrombolysis to the patient and obtains informed consent. If the patient or their legal representative signs the informed consent form for intravenous thrombolysis, the patient is placed on a treatment bed (provided for immediate use upon arrival; a treatment bed must be provided if the patient enters on foot or in a wheelchair) in the thrombolysis treatment area for intravenous thrombolysis under ECG, blood pressure, and clinical monitoring. Patients who do not meet the indications for intravenous thrombolysis or refuse intravenous thrombolysis proceed with the regular treatment process; the images and reports from the low-field movable MRI scan can be accessed in the medical PACS system.

[0059] The AI-assisted decision-making system used in this application improves the accuracy and speed of doctors' diagnoses, providing a solution for rapidly developing precise treatment plans. The mobile magnetic resonance imaging system utilizes artificial intelligence image analysis technology to significantly shorten examination time while improving image quality. It assists doctors in differentiating between ischemia and hemorrhage, identifying the location and volume of the core infarct, recognizing large artery occlusion and stenosis, and determining the duration of stroke after awakening, enabling rapid treatment decisions. By integrating independent operating units, it achieves regional integration of the entire process of examination, thrombolysis, and further treatment for ischemic stroke patients, minimizing delays after arrival at the hospital, dissolving or removing thrombi as early as possible, restoring reperfusion of ischemic brain tissue, saving more brain tissue, improving clinical prognosis, and demonstrating broad application prospects.

[0060] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A regional integrated emergency stroke unit, characterized by, The application relates to an emergency department information system, which comprises an information interaction area, an image examination area and a thrombolytic therapy area. A doctor workstation, a comprehensive information monitoring display and a process support AI assistant are arranged in the information interaction area; the doctor workstation comprises a computer, a file rack and a folding table; the computer is used for information input by doctors; the file rack is used for placing file materials; the folding table is used for temporary placement and storage of medical equipment and files; the comprehensive information monitoring display is used for displaying basic information of a patient, hospital arrival time, clinic arrival time and vital sign conditions; the process support AI assistant is used for automatic inquiry and auxiliary nervous system scale scoring through voice interaction or touch screen key input interaction, and is also used for providing real-time knowledge base and clinical guideline search support. Detection equipment, a first television display screen, a low-field movable magnetic resonance imaging instrument and an AI auxiliary decision system are arranged in the image examination area; the detection equipment is used for electrocardiogram examination, drug gene detection and monitoring of blood oxygen saturation, blood pressure and peripheral blood glucose of a patient to obtain detection results; the first television display screen is used for magnified display of the comprehensive information monitoring display content; the low-field movable magnetic resonance imaging instrument is used for head and neck magnetic resonance imaging of a patient; the AI auxiliary decision system is in communication connection with the comprehensive information monitoring display, the process support AI assistant, the detection equipment, the first television display screen and the low-field movable magnetic resonance imaging instrument; the AI auxiliary decision system is used for auxiliary decision making according to automatic inquiry results, scale scoring results, detection results and head and neck magnetic resonance imaging results, and generates a treatment scheme and displays the treatment scheme. A small refrigerator, an emergency medicine cabinet and a storage cabinet are arranged in the thrombolytic therapy area and are used for storage of medicines, reagents and emergency medicines with different temperature requirements; an emergency vehicle and a vital sign monitoring system are also arranged in the thrombolytic therapy area; the emergency vehicle is used for emergency rescue when a patient has an emergency medical condition; the vital sign monitoring system is used for real-time monitoring and display of vital signs of a patient. An intelligent interaction module is arranged in the process support AI assistant; the intelligent interaction module is based on an emergency structured medical record and performs structured questioning on key medical information such as onset time, onset symptoms, accompanying symptoms, past medical history and allergy history of a patient, performs voice recognition on answers of a patient and / or his / her family members and performs structured input of the key medical information.

2. The regionalized emergency stroke unit of claim 1, wherein, The intelligent interaction module is also used for auxiliary NIHSS score, ABCD2 score and pre-onset mRS score through voice interaction or touch screen key input interaction with a patient and / or his / her family members, and obtains scale scoring results.

3. The regionalized emergency stroke unit of claim 2, wherein, A large language model is also stored in the intelligent interaction module; the large language model is provided with a knowledge base related to stroke clinical diagnosis and treatment, including clinical scales, cerebral infarction acute stage guidelines, cerebral infarction secondary prevention guidelines, cerebral hemorrhage diagnosis and treatment guidelines and subarachnoid hemorrhage clinical diagnosis and treatment guidelines; the knowledge base and clinical guidelines are searched through voice interaction or touch screen key input interaction with a doctor during diagnosis and treatment.

4. The regionalized emergency stroke unit of claim 2, wherein, ​ 5. The regionalized emergency stroke unit of claim 2, wherein, The intelligent interaction module is also used for providing medical procedure explanation and treatment scheme explanation and signing informed consent through voice interaction or touch screen keying interaction with the patient and / or his / her family members.

6. The regional integrated emergency stroke unit of claim 1, wherein, The detection equipment includes an electrocardiograph, a wearable electrocardiograph, a finger clip blood oxygen meter, a sphygmomanometer, a blood glucose meter, and a non-invasive rapid gene typing detector.

7. The regional integrated emergency stroke unit of claim 1, wherein, The low-field mobile magnetic resonance imaging instrument is integrated with a mobile magnetic resonance imaging system; the mobile magnetic resonance imaging system uses deep learning-based image reconstruction methods to realize image super-resolution reconstruction by using low-field and high-field MRI data sets of a patient, and obtains head and neck magnetic resonance imaging results.

8. The regionalized emergency stroke unit of claim 7, wherein, The AI-assisted decision system is used for identifying an acute cerebral infarction lesion site in the head and neck magnetic resonance imaging result and automatically calculating an infarction volume by using a deep learning method; the AI-assisted decision system is also used for judging whether the patient has intracranial large artery occlusion / narrowing by using an MRA sequence of the mobile magnetic resonance; the AI-assisted decision system is also used for using an automatic registration technology to register different sequences to a standard space, realizing automatic identification of DWI / FLAIR mismatch of the acute cerebral infarction lesion, judging an approximate onset time of wake-up stroke, and guiding reperfusion treatment.

9. The regionalized emergency stroke unit of claim 8, wherein, The AI-assisted decision system is also used for combining image interpretation information with scale score results to judge whether the patient has clinical-image mismatch, combining the infarction volume, and assisting in deciding whether the patient needs to be treated by intravascular treatment to realize intracranial reperfusion.

10. The regionalized emergency stroke unit of claim 1, wherein, The thrombolytic treatment area is also provided with a second television display screen and a garbage can; the second television display screen is used for displaying the content of the comprehensive information monitoring display; and the garbage can is used for temporarily placing medical or living garbage.

Images