Intelligent medicine treatment process optimization system based on dynamic risk assessment

Abstract

The invention discloses an intelligent drug treatment process optimization system based on dynamic risk assessment, and relates to the technical field of treatment process optimization, and the system comprises an initial risk index acquisition module which carries out the pre-hospital assessment of a target patient to obtain an initial risk index; the early warning limit value judgment module is used for judging whether the initial risk index reaches a preset early warning limit value or not; the patient shunting module is used for sending out an early warning signal; the iconography feature acquisition module is used for carrying out feature collection to obtain iconography features; the risk index acquisition module is used for adjusting the initial risk index to obtain a target risk index; and the treatment execution module is used for performing treatment execution on the target patient. According to the method and the device, the technical problem that treatment shunting is not timely and risk judgment is inaccurate due to lack of full-process dynamic risk assessment for patient medicine treatment in the prior art is solved, and the technical effects of realizing dynamic optimization of the medicine treatment process of the target patient and improving treatment timeliness and accuracy are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rescue process optimization, in particular to an intelligent drug rescue process optimization system based on dynamic risk assessment. BACKGROUND

[0002] In the field of medical rescue, especially for diseases such as stroke and acute cerebrovascular disease, which have high requirements for rescue timeliness and accuracy, the traditional drug rescue process often faces problems such as pre-hospital and in-hospital information discontinuity, static risk assessment, and rescue resource allocation lag: pre-hospital stage relies on subjective experience of medical staff to judge patient condition risk, lacks standardized quantitative assessment means, resulting in difficulty in timely identification of high-risk patients; there is a delay in information transmission link, pre-hospital assessment data cannot be quickly synchronized to the in-hospital management platform, and in-hospital preparation for rescue is difficult to be made in advance; after admission of patients, the conventional diagnosis and treatment process needs to go through queuing, imaging examination, multi-disciplinary consultation and other links, lacks a priority shunting mechanism for high-risk patients, and is prone to cause rescue delay; at the same time, risk assessment is based on single time point disease data, without dynamic adjustment combined with subsequent imaging examination results, which may cause risk judgment deviation, and further affect the adaptability of drug rescue plan.

[0003] The prior art has the technical problem of lacking full-process dynamic risk assessment for patient drug rescue, resulting in untimely rescue shunting and inaccurate risk judgment. SUMMARY

[0004] The present application provides an intelligent drug rescue process optimization system based on dynamic risk assessment, which is used to solve the technical problem of lacking full-process dynamic risk assessment for patient drug rescue in the prior art, resulting in untimely rescue shunting and inaccurate risk judgment.

[0005] In view of the above problems, the present application provides an intelligent drug rescue process optimization system based on dynamic risk assessment.

[0006] The present application provides an intelligent drug rescue process optimization system based on dynamic risk assessment, which comprises:

[0007] An initial risk index acquisition module is configured to obtain an initial risk index by performing pre-hospital assessment on a target patient via a mobile terminal; a warning limit value judgment module is configured to synchronously transmit the initial risk index to an in-hospital management platform, and judge whether the initial risk index reaches a predetermined warning limit value via the in-hospital management platform; a patient shunting module is configured to issue a warning signal if the initial risk index reaches the predetermined warning limit value, and shunt the target patient to a green rescue channel based on the warning signal; an imaging feature acquisition module is configured to collect brain images of the target patient based on the green rescue channel, and collect imaging features of the brain images; a risk index acquisition module is configured to adjust the initial risk index based on the imaging features to obtain a target risk index; and a rescue execution module is configured to perform rescue on the target patient according to a drug rescue plan corresponding to the matched target risk index.

[0008] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] An initial risk index acquisition module is configured to obtain an initial risk index by performing pre-hospital assessment on a target patient via a mobile terminal; a warning limit value judgment module is configured to synchronously transmit the initial risk index to an in-hospital management platform, and judge whether the initial risk index reaches a predetermined warning limit value via the in-hospital management platform; a patient shunting module is configured to issue a warning signal if the initial risk index reaches the predetermined warning limit value, and shunt the target patient to a green rescue channel based on the warning signal; an imaging feature acquisition module is configured to collect brain images of the target patient based on the green rescue channel, and collect imaging features of the brain images; a risk index acquisition module is configured to adjust the initial risk index based on the imaging features to obtain a target risk index; and a rescue execution module is configured to perform rescue on the target patient according to a drug rescue plan corresponding to the matched target risk index. The technical effect of dynamically optimizing the drug rescue process of the target patient is achieved, and the timeliness and accuracy of rescue are improved. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0011] Figure 1 A structural schematic diagram of an intelligent drug rescue process optimization system based on dynamic risk assessment provided by the embodiments of the present application;

[0012] Figure 2A flow chart of the treatment execution record acquisition unit in the intelligent drug treatment process optimization system based on dynamic risk assessment provided in an embodiment of the present application.

[0013] Explanation of the accompanying symbols: initial risk index acquisition module 10, warning limit judgment module 20, patient triage module 30, imaging feature acquisition module 40, risk index acquisition module 50, treatment execution module 60. DETAILED DESCRIPTION

[0014] This application provides an intelligent drug treatment process optimization system based on dynamic risk assessment to solve the technical problems in the existing technology of lack of full-process dynamic risk assessment of patient drug treatment, resulting in untimely treatment diversion and inaccurate risk judgment.

[0015] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work are within the scope of protection of this application.

[0016] Examples, such as Figure 1 As shown, the present application provides an intelligent drug treatment process optimization system based on dynamic risk assessment, the system comprising:

[0017] The initial risk index acquisition module 10 is used to perform a pre-hospital assessment on the target patient through a mobile terminal to obtain an initial risk index.

[0018] Specifically, multi-dimensional status monitoring of the target patient is carried out through a mobile terminal to generate an initial risk index: first, the physical status acquisition unit retrieves the predetermined status indicators including vital signs, including body temperature, heart rate, respiratory rate, blood pressure, blood oxygen saturation, brain tissue ischemic damage indicators, including cerebral blood flow, cerebral blood volume, etc., neurological function impairment indicators, including consciousness, muscle strength, etc. and onset duration, and monitors the real-time physical status; then the state index acquisition unit eliminates the dimension of the real-time physical status and normalizes and weights it to calculate the real-time state index; at the same time, the influence coefficient acquisition unit analyzes the target patient's previous history of atrial fibrillation to determine the real-time influence coefficient; finally, the risk index acquisition unit takes the product of the real-time influence coefficient and the real-time state index as the initial risk index, which provides a basic basis for subsequent risk judgment.

[0019] The warning limit judgment module 20 is used to synchronously transmit the initial risk index to the in-hospital management platform, and judge whether the initial risk index reaches a predetermined warning limit through the in-hospital management platform.

[0020] Specifically, after the initial risk index is synchronously transmitted to the in-hospital management platform, the platform determines whether the initial risk index reaches the predetermined early warning limit value according to the preset risk grading standard, so as to quickly identify the emergency degree of the patient and provide data support for subsequent diversion decision.

[0021] The patient diversion module 30 is configured to, if the initial risk index reaches the predetermined early warning limit value, send an early warning signal and divert the target patient to a green rescue channel based on the early warning signal.

[0022] Specifically, if the initial risk index reaches the predetermined early warning limit value, an early warning signal is immediately sent to trigger a green rescue channel starting mechanism, so that the target patient is preferentially diverted to the green rescue channel, thereby skipping the regular diagnosis and treatment queuing process, shortening the pre-hospital to in-hospital rescue connection time, and ensuring that the high-risk patient is quickly admitted to the hospital.

[0023] The imaging feature acquisition module 40 is configured to acquire brain images of the target patient based on the green rescue channel, and collect features of the brain images to obtain imaging features.

[0024] Specifically, based on key image data acquisition and feature extraction, and relying on the preferential resource allocation of the green rescue channel, brain image acquisition (such as CT, etc.) of the target patient is quickly completed, and feature recognition and collection are performed on the acquired brain images to extract imaging features including key information such as blood vessels and brain tissue morphology, thereby providing imaging basis for dynamic adjustment of the risk index.

[0025] The risk index acquisition module 50 is configured to adjust the initial risk index based on the imaging features to obtain a target risk index.

[0026] Specifically, the risk sign acquisition unit reads the predetermined large vessel occlusion risk indicators such as arterial high density sign, early ischemic change, and eyeball deviation, and iteratively screens target large vessel occlusion risk signs in the imaging features; if the sign meets the predetermined condition constraint, the examination signal sending unit sends a multi-mode examination signal to start target multi-mode image examination including CT angiography and CT perfusion imaging, and the verification result acquisition unit obtains a verification result through multi-mode image verification; then, the index adjustment unit adjusts the initial risk index based on the verification result to obtain a target risk index that is more consistent with the actual condition of the patient; at the same time, the brain image, the multi-mode image, and the target risk index are synchronously transmitted to the neurology department of the in-hospital management platform by the index synchronous transmission subunit, triggering remote consultation, determining an appropriate drug rescue plan by the drug rescue plan acquisition subunit, and if the target risk index reaches a predetermined threshold, a drug preparation signal is also sent to improve the priority of anti-platelet drugs through the preparation priority improvement micro-unit, so as to prepare in advance for rescue execution.

[0027] The rescue execution module 60 is configured to perform rescue on the target patient according to the matched drug rescue plan corresponding to the target risk index.

[0028] Specifically, on one hand, rescue is performed on the target patient according to the matched drug rescue plan, and the rescue execution record acquisition unit records the whole rescue process to generate a rescue execution record; on the other hand, the admission drug administration time length calculation unit calculates the admission drug administration time length according to the admission time and the drug administration time in the record, and if the time length exceeds a first time length threshold, the process optimization signal sending unit sends a process optimization signal immediately to trigger the designated person accompanying mechanism starting unit to start the designated person accompanying mechanism, and the patient is directly guided to the emergency department by the designated nursing staff, skipping the triage process; at the same time, the rescue visual view acquisition sub-unit visualizes the rescue execution record to generate a rescue visual view, and if it is monitored that the time consumption of any rescue link exceeds the corresponding predetermined threshold, the early warning signal sending sub-unit sends an early warning signal, and the early warning information pushing sub-unit synchronously pushes the early warning information to the target rescue terminal such as the attending physician and the emergency nurse, so as to ensure timely correction of process deviation and guarantee the rescue efficiency and quality.

[0029] In one possible implementation manner, the initial risk index acquisition module 10 further includes:

[0030] The body state acquisition unit is configured to monitor the state of the target patient according to the predetermined state indicators to obtain a real-time body state.

[0031] The state index acquisition unit is configured to perform dimensionless processing on the real-time body state, and obtain a real-time state index through normalization and weighting.

[0032] The influence coefficient acquisition unit is configured to obtain the past atrial fibrillation history of the target patient, and analyze the past atrial fibrillation history to obtain a real-time influence coefficient.

[0033] The risk index acquisition unit is configured to take the product of the real-time influence coefficient and the real-time state index as the initial risk index.

[0034] Specifically, the body state acquisition unit, as the data collection core, is responsible for calling the preset multi-dimensional predetermined state indicators, conducting comprehensive state monitoring on the target patient to obtain the real-time body state. Among them, the predetermined state indicators cover four categories of key information: first, vital sign indicators, including body temperature, heart rate, respiratory rate, blood pressure and oxygen saturation, which directly reflect the patient's basic physiological function; second, brain tissue ischemia and injury indicators, including cerebral blood flow, cerebral blood volume, mean transit time and peak time, which are used to assess the blood supply and injury of the brain; third, neurological deficit indicators, including consciousness, horizontal gaze, facial paralysis, upper and lower limb muscle strength, walking ability, language and sensation, which accurately capture abnormal manifestations of neurological function; fourth, the length of onset, which is an important time dimension basis for judging the progression of the disease. Through the synchronous monitoring of these indicators, a complete set of real-time body state data is formed.

[0035] The state index acquisition unit first eliminates the dimension of different types of indicators in the real-time body state (such as degrees Celsius of body temperature, beats per minute of heart rate, etc.), unifies the data measurement standard, and avoids the influence of unit differences on subsequent calculations; then according to the importance of each indicator in patient risk assessment, different weights are given and normalized weighted operation is performed, the multi-dimensional body state data is converted into a single, quantifiable real-time state index, and comprehensive quantitative assessment of the current physical condition of the patient is realized.

[0036] The influence coefficient acquisition unit obtains the patient's past atrial fibrillation history by calling the target patient's medical records or entering information on the mobile terminal by medical staff; then analyzes the details such as the length of the history, the frequency of onset, and the treatment, judges the influence degree of the history on the current disease risk, and finally generates the corresponding real-time influence coefficient. For example, for patients with long history, frequent onset and poor control, the real-time influence coefficient will be relatively higher to reflect the additional risk brought by the history.

[0037] The risk index acquisition multiplies the real-time state index obtained and the determined real-time influence coefficient, and the result is the initial risk index of the target patient. This calculation method not only integrates the quantitative assessment of the patient's current body state, but also considers the risk addition of the past atrial fibrillation history, ensuring that the initial risk index can comprehensively and objectively reflect the patient's pre-hospital risk level, providing accurate data support for subsequent early warning judgment and diversion decision-making.

[0038] In one possible implementation manner, the body state acquisition unit further includes:

[0039] The predetermined state indicators include vital sign indicators, brain tissue ischemic injury indicators, neurological deficit indicators, and onset duration. The vital sign indicators include body temperature, heart rate, respiratory rate, blood pressure, and blood oxygen saturation. The brain tissue ischemic injury indicators include cerebral blood flow, cerebral blood volume, mean transit time, and peak time. The neurological deficit indicators include consciousness, horizontal gaze, facial paralysis, upper and lower limb muscle strength, walking ability, language, and sensation.

[0040] Specifically, the predetermined state indicators include four dimensions of vital sign indicators, brain tissue ischemic injury indicators, neurological deficit indicators, and onset duration, and each dimension corresponds to specific monitoring content, forming a complete monitoring system covering the multiple aspects of the patient's state. The vital sign indicators specifically include body temperature, heart rate, respiratory rate, blood pressure, and blood oxygen saturation, which are used to monitor the basic physiological activity state of the patient. The brain tissue ischemic injury indicators specifically include cerebral blood flow, cerebral blood volume, mean transit time, and peak time, which are used to monitor the brain blood supply and ischemic injury-related state of the patient. The neurological deficit indicators specifically include consciousness, horizontal gaze, facial paralysis, upper and lower limb muscle strength, walking ability, language, and sensation, which are used to monitor the integrity and normal operation of the nervous system function of the patient. The onset duration is an independent monitoring dimension, which records the time interval from the occurrence of related symptoms to the start of monitoring, and provides a key reference for the time dimension of subsequent risk assessment.

[0041] In one possible implementation manner, the risk index acquisition module 50 further includes:

[0042] A risk sign acquisition unit is configured to read predetermined large vessel occlusion risk indicators and traverse in the imaging features based on the predetermined large vessel occlusion risk indicators to obtain target large vessel occlusion risk signs.

[0043] An examination signal sending unit is configured to send a multi-mode examination signal if the target large vessel occlusion risk signs meet predetermined condition constraints.

[0044] A verification result acquisition unit is configured to start multi-mode image examination based on the multi-mode examination signal to obtain target multi-mode images, and perform verification based on the target multi-mode images to obtain a verification result.

[0045] An index adjustment unit is configured to analyze the verification result and adjust the initial risk index to obtain the target risk index.

[0046] Specifically, the risk sign acquisition unit first reads a set of indicators including arterial hyperdensity sign, early ischemic change and eyeball deviation, and then conducts a comprehensive traversal search on the collected patient brain image features based on these indicators. Through image recognition algorithms, the image manifestations corresponding to each risk indicator are matched one by one, such as identifying whether there is an arterial hyperdensity shadow in the brain image, whether there is early ischemic density change in the brain tissue, whether there is abnormal deviation in the eyeball position, etc. Finally, the image information matched with the predetermined large vessel occlusion risk indicators is screened out to form the target large vessel occlusion risk sign.

[0047] The examination signal sending unit receives the target large vessel occlusion risk sign output by the risk sign acquisition unit, i.e. the image information matched with the predetermined large vessel occlusion risk indicators such as arterial hyperdensity sign, early ischemic change and eyeball deviation, which is screened out from the patient brain image features. Then, the built-in predetermined condition constraints of the system are called for judgment. The predetermined condition constraints include multi-dimensional judgment criteria for risk signs, such as the number of risk signs, e.g. the simultaneous occurrence of arterial hyperdensity sign and early ischemic change, the severity, e.g. the range of brain tissue area involved in early ischemic change, the clarity and distribution characteristics of arterial hyperdensity sign, and whether the direction and amplitude of eyeball deviation conform to the typical lesion manifestations, etc. When the target large vessel occlusion risk sign meets the above predetermined condition constraints, it means that the patient has a possibility of large vessel occlusion reaching the threshold value that needs further accurate investigation. At this time, a multi-modal examination signal is immediately generated and sent, which will be transmitted to the device control terminal of the in-hospital image examination department and the in-hospital management platform, ensuring that the image department receives the examination instruction in time and starts the multi-modal image examination including CT angiography and CT perfusion imaging for the patient, providing more accurate image basis for subsequent disease verification and risk index adjustment.

[0048] After the check result acquisition unit receives the multimodal examination signal, it immediately synchronously pushes the signal to the device control terminal of the in-hospital image examination department and the in-hospital management platform. The device control terminal is associated with special image examination instruments such as CT angiography equipment and CT perfusion imaging equipment. After the signal triggers, the instruments quickly enter the ready-to-work state. At the same time, the in-hospital management platform automatically associates the basic information of the target patient, the pre-hospital assessment data, and the previous brain image data, to ensure that the examination process accurately corresponds to the patient. Subsequently, under the priority resource allocation support of the green treatment channel, the target patient quickly enters the image examination link and sequentially completes the CT angiography and CT perfusion imaging examination: CT angiography clearly presents the shape, running and lumen patency of the cerebral blood vessels by intravenous injection of contrast medium, and can directly judge whether there is large vessel stenosis or occlusion; CT perfusion imaging quantitatively analyzes parameters such as cerebral blood flow and cerebral blood volume by dynamically monitoring the perfusion process of contrast medium in brain tissue, and evaluates the ischemic area and damage degree of brain tissue. The two together constitute the target multimodal image. Finally, the target multimodal image is subjected to data analysis and cross-checking: on the one hand, the CT angiography image confirms whether there is actual vascular lesion corresponding to the previous target large vessel occlusion risk sign (such as arterial high density sign), and determines the occlusion site and degree; on the other hand, combined with the CT perfusion imaging data, it is judged whether the brain tissue corresponding to the risk sign has perfusion abnormalities, the relevance of the risk sign and the actual condition is verified, false positive sign interference is excluded, and finally the information of vascular lesion, brain tissue perfusion state and the like is integrated to form a complete check result containing "whether there is large vessel occlusion, occlusion details, and brain tissue damage correlation", which provides accurate basis for subsequent initial risk index adjustment.

[0049] The analysis of the check result and the adjustment process of the initial risk index by the index adjustment unit need to be carried out in combination with the multi-modal image check information and the disease risk association rule, and the specific implementation is as follows: first, the index adjustment unit receives the check result output by the check result acquisition unit, which is based on the target multi-modal image, at least includes CT angiography and CT perfusion imaging, covers the actual determination conclusion of large vessel occlusion, including whether there is occlusion, the specific blood vessel site of occlusion (such as internal carotid artery, middle cerebral artery, etc.), the occlusion degree (such as complete occlusion, partial stenosis), and the brain tissue perfusion abnormal range (such as ischemic penumbra volume, core infarction size) and other key information. Then, the built-in "check result-risk adjustment coefficient" corresponding model of the system is called, and the above check information is quantified according to the preset dimension: for example, when there is complete occlusion of the internal carotid artery and the core infarction volume is large, a high risk adjustment coefficient is matched; when there is partial stenosis of the middle cerebral artery and there is no obvious brain tissue perfusion abnormality, a low risk adjustment coefficient is matched; when there is no explicit large vessel occlusion and only slight imaging signs, a basic adjustment coefficient is matched. Then, based on the quantified risk adjustment coefficient, the initial risk index is calculated and adjusted. If the check result indicates that the disease risk is higher than the initial assessment, such as the existence of an unexpected large vessel occlusion, the initial risk index is multiplied by an adjustment coefficient greater than 1 to increase the risk index value; if the check result indicates that the disease risk is lower than the initial assessment, such as only slight risk signs without actual occlusion, the initial risk index is multiplied by an adjustment coefficient less than 1 to reduce the risk index value; if the check result is basically consistent with the initial assessment, the initial risk index is maintained or slightly corrected. Finally, through the above analysis and calculation, the target risk index that can accurately reflect the current actual disease risk level of the patient is obtained.

[0050] In a possible implementation manner, the risk sign acquisition unit further includes:

[0051] The predetermined large vessel occlusion risk indicators include arterial high density sign, early ischemic change, and eyeball deviation.

[0052] Specifically, the predetermined large vessel occlusion risk indicators specifically include three key image manifestations of arterial high density sign, early ischemic change and eyeball deviation. The three indicators jointly constitute the preliminary basis for determining the risk of large vessel occlusion from the dimensions of blood vessel morphology, brain tissue state and eye signs. Among them, the arterial high density sign mainly manifests as the appearance of an abnormal high density shadow in the intracranial large vessel lumen area in the brain image, which is originally of equal density or low density, such as the internal carotid artery and the middle cerebral artery. This sign usually indicates the presence of thrombus or embolism in the blood vessel and is a typical early image signal of large vessel occlusion. The early ischemic change reflects the image manifestation of the local brain tissue appearing different in density from the normal brain tissue, such as the blurring of the gray-white matter boundary of the brain tissue, the shallowing or disappearance of the brain sulcus, and the swelling of the local brain tissue, which reflects the early ischemic change of the brain tissue in the blood supply area after large vessel occlusion. The eyeball deviation refers to the abnormality of the horizontal gaze direction of the patient's eyes in the brain image, such as the deviation of the eyes to one side (commonly to the side of large vessel occlusion), which is related to the abnormality of eye movement control caused by the ischemic influence on the brain neural pathway and can be used as an auxiliary basis for determining the risk of large vessel occlusion. The three indicators jointly constitute the reference standard for the risk of large vessel occlusion.

[0053] In a possible implementation manner, the verification result acquisition unit further includes:

[0054] The target multi-modal image at least includes CT angiography and CT perfusion imaging.

[0055] Specifically, the target multi-modal image at least includes CT angiography and CT perfusion imaging. The CT angiography can clearly present the lumen morphology, running path and wall condition of the intracranial large vessel (such as the internal carotid artery, the middle cerebral artery and the anterior cerebral artery) by acquiring the image data of the flow of the iodine-containing contrast agent in the brain blood vessel, reconstructing the three-dimensional image of the brain blood vessel by the post-processing technology, and directly providing the morphological evidence for judging whether the large vessel is occluded or not. The CT perfusion imaging can reflect the local blood perfusion state of the brain tissue, help to judge whether the brain tissue in the blood supply area after large vessel occlusion is ischemic, the size of the ischemic area, and distinguish the ischemic penumbra and the core infarction area, and provide the image support at the brain tissue function level for evaluating the severity of the disease and the subsequent treatment decision. The two cooperate with each other to not only determine the vascular lesion from the structure, but also evaluate the brain tissue damage risk from the function, and jointly constitute the comprehensive and accurate target multi-modal image, which provides a reliable image data basis for verifying the large vessel occlusion risk sign and adjusting the initial risk index.

[0056] In a possible implementation manner, the index adjustment unit further includes:

[0057] An index synchronization transmission subunit, configured to synchronize the brain image, the target multi-modal image, and the target risk index to a neurology department of the in-hospital management platform.

[0058] A drug rescue plan acquisition subunit, configured to trigger remote consultation of the neurology department, and obtain the drug rescue plan.

[0059] Specifically, the index synchronization transmission subunit is a data interaction core, and receives the generated key medical data, including the brain image collected and processed by the image feature acquisition module, the target multi-modal image generated by the verification result acquisition unit, and the target risk index calculated by the index adjustment unit. The subunit integrates and packages the three types of data through an encryption data transmission protocol built in the system, to ensure the integrity and security of the data in the transmission process. Then, the subunit accurately locates a dedicated data receiving port of the neurology department in the in-hospital management platform, synchronously pushes the integrated brain image, target multi-modal image, and target risk index to the port, and realizes efficient flow of the patient core diagnosis and treatment data to the neurology department.

[0060] The drug rescue plan acquisition subunit starts to operate after completing data pushing. First, the subunit automatically triggers a remote consultation mechanism of the neurology department, sends a consultation request signal to a preset neurologist terminal (such as a doctor's office computer or a mobile diagnosis and treatment APP) of the department, and attaches patient basic information and a data digest of the synchronized image and risk index to the signal, to remind the doctor to participate in the consultation in a timely manner. During the remote consultation, the specialist judges the patient's large vessel occlusion and brain tissue damage degree based on the brain image and the target multi-modal image, determines the patient's disease risk level in combination with the target risk index, and jointly discusses an adaptive treatment scheme. After the consultation forms a unified diagnosis and treatment opinion, the subunit automatically converts the treatment scheme determined in the discussion into a standardized drug rescue plan (including the type, dose, administration opportunity, and matters needing attention of the drug, and the like), and completes collection and storage of the plan, to provide an explicit scheme basis for subsequent rescue execution modules to develop precise rescue.

[0061] In a possible implementation manner, the drug rescue plan acquisition subunit further includes:

[0062] A drug preparation signal sending micro-unit, configured to send a drug preparation signal when the target risk index reaches a predetermined threshold.

[0063] A preparation priority promotion micro-unit, configured to promote a preparation priority of an anti-platelet drug based on the drug preparation signal.

[0064] Specifically, the drug preparation signal sending micro-unit continuously receives the target risk index output by the index adjustment unit, and compares it with a predetermined threshold value built in the system in real time. The threshold value is set based on clinical diagnosis and treatment guidelines and corresponds to the critical value of the drug preparation required in advance under different risk levels of the patient's condition. When it is monitored that the target risk index reaches or exceeds the predetermined threshold value, it indicates that the patient's condition risk is high, and the drug preparation needs to be started in advance to shorten the rescue delay time. At this time, a drug preparation signal is immediately generated, which carries the patient's basic information, the target risk index value and the corresponding risk level label, to ensure that the subsequent drug preparation work is accurately associated with the patient and meets the needs of the patient's condition. Then, the signal is pushed to the hospital pharmacy management terminal, emergency nursing station terminal and hospital management platform simultaneously to realize the synchronized awareness of drug preparation demand in multiple terminals.

[0065] After receiving the drug preparation signal, the preparation priority promotion micro-unit quickly starts the priority adjustment mechanism: on the one hand, it identifies the anti-platelet drugs suitable for the patient's condition through the hospital pharmacy management system to call the current drug preparation queue; on the other hand, based on the risk level label in the drug preparation signal, the identified anti-platelet drugs are removed from the regular preparation queue and promoted to the highest priority queue of the pharmacy drug preparation task. At the same time, a priority reminder notice is sent to the pharmacy staff terminal, clearly marking "high-risk patients urgently need anti-platelet drugs", to ensure that the pharmacy prioritizes the allocation of manpower and drug resources to complete the identification, packaging and preparation of such drugs. In addition, the drug preparation progress is fed back to the hospital management platform and emergency nursing station terminal in real time, so that medical staff can master the drug preparation situation and make good connection for subsequent rescue execution link drug administration, further shortening the interval time from the plan determination to the drug administration.

[0066] In one possible implementation manner, the rescue execution module 60 further includes:

[0067] A rescue execution record acquisition unit is configured to acquire a rescue execution record, wherein the rescue execution record refers to a process record of rescue execution on the target patient based on the drug rescue plan.

[0068] An admission drug administration duration calculation unit is configured to calculate an admission drug administration duration according to an admission time and a drug administration time in the rescue execution record.

[0069] A process optimization signal sending unit is configured to send a process optimization signal when the admission drug administration duration exceeds a first duration threshold.

[0070] An accompanying examination mechanism starting unit is configured to start a special person accompanying examination mechanism based on the process optimization signal, and guide the target patient to an emergency department according to the special person accompanying examination mechanism and skip a triage process.

[0071] Specifically, the rescue execution record acquisition unit is responsible for acquiring the rescue execution record of the target patient throughout the entire process. The record covers all details of rescue operations based on the drug rescue plan, including the inspection items received by the patient after admission, the specific time of drug administration, the administration method, the drug dosage, the real-time changes of the patient's vital signs during the rescue process, the operation notes of medical staff, and other information. These information are collected through real-time input by medical staff on the diagnosis terminal, automatic collection by equipment, and other means, and ultimately form a complete and traceable rescue execution record, providing basic data support for subsequent time calculation and process optimization.

[0072] The admission drug administration time calculation unit accurately extracts the admission time and the first drug administration time of the patient from the rescue execution record. Then, through a time difference calculation algorithm, the first drug administration time is subtracted from the admission time to obtain the specific admission drug administration time. This time directly reflects the time interval from the patient's admission to the start of core drug rescue, and is a key quantitative indicator for evaluating the efficiency of the rescue process.

[0073] The process optimization signal sending unit takes the admission drug administration time as the basis for judgment, and pre-stores a first time threshold set by the system. The threshold is set based on the clinical rescue time efficiency standard, aiming to ensure that high-risk patients can quickly receive drug treatment. The result obtained by the admission drug administration time calculation unit is compared with the threshold in real time. If the admission drug administration time exceeds the first time threshold, it indicates that there is a bottleneck in the current rescue process, such as delay in inspection connection, excessive time consumption in drug dispensing, etc. At this time, a process optimization signal is immediately generated, which includes the specific value of the patient's admission drug administration time and the preliminary judgment result of the overtime link. The signal is simultaneously pushed to the hospital management platform, the emergency department terminal, and the pharmacy terminal, ensuring that relevant departments are aware of the process problems in a timely manner.

[0074] After receiving the process optimization signal, the accompanying inspection mechanism starting unit quickly starts the dedicated accompanying inspection mechanism. On the one hand, it automatically matches designated nursing personnel with the corresponding qualifications through the hospital management platform and sends an accompanying inspection task notification to their terminal, specifying the accompanying object (target patient), accompanying target, and instruction to skip the triage process. On the other hand, it simultaneously pushes the accompanying coordination signal to various relevant departments of the hospital (such as the imaging department and the laboratory), informing that the target patient will be guided by the designated nursing personnel to complete the necessary inspection without queuing. In actual execution, the designated nursing personnel receive the notification and immediately interface with the target patient, guiding them to skip the triage process and quickly go to the emergency department or the required inspection department. By reducing intermediate links and optimizing personnel coordination, the overall time from admission to receiving rescue is shortened, the overtime problem of admission drug administration time is addressed, and the efficiency and timeliness of the subsequent rescue process are improved.

[0075] In one possible implementation manner, as shown in Figure 2 the rescue execution record acquisition unit further includes:

[0076] The rescue visual view acquisition subunit is configured to visualize the rescue execution record to obtain a rescue visual view.

[0077] The early warning signal issuing subunit is configured to issue an early warning signal according to the rescue visual view if any rescue link exceeds a corresponding predetermined threshold.

[0078] The early warning information pushing subunit is configured to push early warning information to a target rescue terminal based on the early warning signal, wherein the target rescue terminal includes a primary physician and an emergency nurse.

[0079] Specifically, the rescue visual view acquisition subunit extracts all key information in the rescue execution record, including patient admission time, start and end time of each rescue link, operation performer of each link, drug type and dosage in the drug rescue plan, patient vital sign value and target risk index change data at different time nodes, and operation note information input by medical staff during the rescue process. Then, a horizontal time axis is constructed with the patient admission time as the starting point, and the above-mentioned rescue links are arranged in the time axis in the actual occurrence order. Different colored rectangular modules are used to distinguish link types (for example, a blue module represents an image examination link, a green module represents a drug preparation link, and a red module represents a drug administration link), and the length of the module corresponds to the actual time length of each link. At the same time, key data labels are marked at the corresponding positions of the time axis. Finally, the textual rescue execution record is converted into an intuitive and clear rescue visual view, which presents the time progress, key data and operation details of the whole rescue process.

[0080] The early warning signal issuing subunit compares the actual time consumption of each rescue link in the visual view with the corresponding predetermined threshold. If the actual time consumption of any rescue link exceeds the corresponding predetermined threshold (for example, the drug preparation time consumption of 25 minutes exceeds the predetermined threshold of 20 minutes, or the image examination time consumption of 40 minutes exceeds the predetermined threshold of 30 minutes), it is determined that there is a risk of efficiency delay in this link. At this time, an early warning signal is immediately generated, which includes the name of the overtime link, the actual time consumption, the predetermined threshold and the overtime duration, etc. The signal provides accurate content for subsequent early warning information pushing.

[0081] After receiving the early warning signal, the early warning information pushing subunit quickly starts the information distribution mechanism, first identifies the target treatment terminal preset by the system, and clearly includes the attending physician terminal (such as the doctor's exclusive diagnosis and treatment APP, office computer) and the emergency nurse terminal (such as the nurse station terminal); then the key information (timeout link, time-consuming data, etc.) in the early warning signal is converted into standardized early warning information and pushed to the above target treatment terminal; the terminal reminder function will also be triggered during the pushing process to ensure that the attending physician and emergency nurse can receive the early warning information at the first time and intervene in the timeout link to check the problem (such as coordinating resources to speed up drug allocation, giving priority to examination), avoiding further delay and ensuring the efficient progress of the treatment process.

[0082] It should be noted that the above sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. Moreover, the above describes specific embodiments of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0083] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0084] The specification and drawings are merely exemplary of the present application, and any and all modifications, variations, combinations or equivalents that fall within the scope of the present application are considered to be covered by the present application. Obviously, those skilled in the art can make various modifications and changes to the present application without departing from the scope of the present application. Thus, if these modifications and changes of the present application belong to the scope of the present application and its equivalents, the present application is intended to include these modifications and changes.

Claims

1. Intelligent drug treatment process optimization system based on dynamic risk assessment, characterized by: include: An initial risk index acquisition module is used to obtain an initial risk index by performing a pre-hospital assessment on a target patient through a mobile terminal; An early warning limit judgment module is used to synchronously transmit the initial risk index to the in-hospital management platform, and judge whether the initial risk index reaches a predetermined early warning limit through the in-hospital management platform; A patient diversion module is used to issue an early warning signal if the target patient reaches the target level, and divert the target patient to a green treatment channel based on the early warning signal; an imaging feature acquisition module, configured to acquire a brain image of the target patient based on the green treatment channel, and collect features of the brain image to obtain imaging features; a risk index acquisition module, configured to adjust the initial risk index based on the imaging features to obtain a target risk index; The treatment execution module is used to execute treatment for the target patient according to the drug treatment plan corresponding to the matched target risk index.

2. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 1, characterized in that: The initial risk index acquisition module further includes: A body status acquisition unit, configured to retrieve predetermined status indicators to monitor the status of the target patient and obtain a real-time body status; a state index obtaining unit, configured to perform dimension-eliminating processing on the real-time body state and normalize and weight the real-time state index; an influence coefficient acquisition unit, configured to acquire a previous atrial fibrillation history of the target patient and analyze the previous atrial fibrillation history to obtain a real-time influence coefficient; The risk index obtaining unit is configured to obtain the product of the real-time impact coefficient and the real-time status index as the initial risk index.

3. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 2, characterized in that: The predetermined status indicators include vital sign indicators, brain tissue ischemic damage indicators, neurological function defect indicators and onset duration; The vital signs indicators include body temperature, heart rate, respiratory rate, blood pressure and blood oxygen saturation; Wherein, the brain tissue ischemic injury indicators include cerebral blood flow, cerebral blood volume, mean transit time and peak time; Among them, the neurological function impairment indicators include consciousness, horizontal gaze, facial paralysis, upper and lower limb muscle strength, walking ability, language and sensation.

4. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 1, characterized in that: The risk index acquisition module further includes: a risk sign acquisition unit, configured to read a predetermined large vessel occlusion risk index, and traverse the imaging features based on the predetermined large vessel occlusion risk index to obtain a target large vessel occlusion risk sign; an inspection signal issuing unit, configured to issue a multimodal inspection signal if the target large vessel occlusion risk sign meets a predetermined condition constraint; a verification result acquiring unit, configured to initiate a multimode image inspection based on the multimode inspection signal to obtain a target multimode image, and perform verification using the target multimode image to obtain a verification result; The index adjustment unit is used to analyze the verification result and adjust the initial risk index to obtain the target risk index.

5. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 4, characterized in that: The predetermined large vessel occlusion risk indicators include arterial hyperdensity sign, early ischemic changes, and eye deviation.

6. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 4, characterized in that: The target multimodal imaging includes at least CT angiography and CT perfusion imaging.

7. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 4, characterized in that: The index adjustment unit further includes: An index synchronous transmission subunit, configured to synchronously transmit the brain image, the target multimodal image, and the target risk index to the neurology department of the in-hospital management platform; The drug treatment plan acquisition subunit is used to trigger a remote consultation with the neurology department to obtain the drug treatment plan.

8. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 7, characterized in that: The drug treatment plan acquisition subunit further includes: A medicine preparation signal issuing micro unit, configured to issue a medicine preparation signal when the target risk index reaches a predetermined threshold; The preparation priority increasing micro unit is used to increase the preparation priority of the antiplatelet drug based on the drug preparation signal.

9. The intelligent drug treatment process optimization system based on dynamic risk assessment as claimed in claim 1, characterized in that: The rescue execution module also includes: a treatment execution record acquisition unit, configured to acquire a treatment execution record, wherein the treatment execution record refers to a process record of executing treatment on the target patient based on the drug treatment plan; A hospital admission medication duration calculation unit, configured to calculate the hospital admission medication duration based on the hospital admission time and medication time in the treatment execution record; a process optimization signal issuing unit, configured to issue a process optimization signal when the duration of the medication in hospital exceeds a first duration threshold; The accompanying inspection mechanism starting unit is used to start the dedicated accompanying inspection mechanism based on the process optimization signal, and skip the triage process according to the dedicated accompanying inspection mechanism, and the designated caregiver guides the target patient to the emergency department.

10. The intelligent drug treatment process optimization system based on dynamic risk assessment according to claim 9, characterized in that: The rescue execution record acquisition unit further includes: a treatment visual graph acquisition subunit, configured to perform visualization processing on the treatment execution record to obtain a treatment visual graph; an early warning signal issuing subunit, configured to issue an early warning signal if any treatment link exceeds a corresponding predetermined threshold value according to the visual diagram of the treatment; The early warning information pushing subunit is used to push the early warning information to the target treatment terminal based on the early warning signal, wherein the target treatment terminal includes the attending doctor and the emergency nurse.

Images