Structured data processing system and method for infectious shock diagnosis and treatment image information

Abstract

The invention relates to the technical field of medical information processing, and discloses an infectious shock diagnosis and treatment image information structured data processing system and method, and the system comprises a processor, an image collection module, a multi-modal data fusion module and other modules. The image acquisition module acquires a CT image, an ultrasonic image and hemodynamic time series data, the multi-modal data fusion module processes the data to generate multi-modal fusion data, the dynamic feature extraction module generates a dynamic feature vector, and the structured processing engine generates a structured diagnosis and treatment record. The abnormity detection module monitors and records parameters, the diagnosis and treatment decision module generates diagnosis and treatment suggestions according to abnormal signals, the early warning feedback module provides visual early warning, and the performance monitoring module evaluates system performance. The multi-modal data fusion processing is realized, the structured diagnosis and treatment record is generated, the accurate diagnosis and treatment decision support is provided, the real-time monitoring and system optimization functions are realized, and the diagnosis and treatment level and efficiency of the infectious shock are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical information processing, and in particular to a system and method for processing structured data of septic shock diagnosis and treatment image information. Background Art

[0002] Septic shock is a serious clinical syndrome with rapid disease progression and high mortality rate. Timely and accurate diagnosis and treatment are crucial to improving the prognosis of patients. However, in the diagnosis and treatment of septic shock, there are many difficulties in data processing and decision support.

[0003] In terms of medical imaging data processing, the diagnosis of septic shock patients often relies on a variety of medical imaging data, such as CT images, ultrasound images, and hemodynamic time series data. These data have different modalities and characteristics, and traditional processing methods are difficult to effectively integrate and analyze them. CT images can clearly present the morphological structure of organs, but cannot directly reflect the dynamic changes of hemodynamics; ultrasound images can observe the blood flow of some tissues and organs in real time, but the information is relatively limited; although hemodynamic time series data can reflect the changes in cardiac function and circulatory status over time, it lacks intuitive anatomical structure information. It has become a major challenge to align these multimodal data in time and space and normalize the features in order to comprehensively and accurately analyze the patient's condition.

[0004] From the perspective of clinical data management, most of the existing medical records are in unstructured or semi-structured form, containing a large amount of text descriptions and scattered data. When doctors view and analyze these data, they need to spend a lot of time and energy to screen and integrate key information, which is inefficient and easy to miss important details. For example, when recording patients' physiological parameters, hemodynamic indicators, and imaging characteristics, there is a lack of a unified and standardized data format, and the recording methods of different doctors vary greatly, which brings great inconvenience to subsequent data analysis and clinical decision-making. In addition, unstructured data is not conducive to the information management and big data analysis of medical data, and it is impossible to fully explore the potential value behind the data, making it difficult to achieve precision medicine based on big data.

[0005] In terms of diagnosis and treatment decision support, septic shock is a complex and changeable disease, and the conditions of each patient are different, which places high demands on the doctor's diagnosis and treatment decision-making ability. At present, doctors mainly formulate diagnosis and treatment plans based on their own experience and limited clinical guidelines, and lack personalized and precise decision support tools. Due to the lack of full consideration of individual differences among patients, traditional diagnosis and treatment plans may not achieve the best treatment effect and may even delay the disease. Moreover, when faced with a variety of possible diagnosis and treatment plans, how to quickly and accurately select the optimal plan and coordinate resource conflicts and timing conflicts between different plans are also urgent problems to be solved in the clinical diagnosis and treatment process.

[0006] In addition, there is a lack of effective real-time monitoring and early warning mechanisms throughout the diagnosis and treatment process. It is impossible to detect subtle changes in the patient's condition in time, resulting in the inability to take appropriate intervention measures in the early stages of the disease's deterioration. At the same time, the performance evaluation of the diagnosis and treatment system is not perfect enough, and problems in the system in data processing, decision support, etc. cannot be discovered in time, making it difficult to optimize and improve the system in a targeted manner, limiting the improvement of the diagnosis and treatment level of septic shock. In summary, it is urgent to develop a system and method that can effectively process septic shock diagnosis and treatment image information, realize data structured management, provide accurate diagnosis and treatment decision support, and have real-time monitoring and system optimization functions. Summary of the invention

[0007] The purpose of the present invention is to provide a system and method for processing structured data of septic shock diagnosis and treatment image information to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above-mentioned object, the present invention provides the following technical solutions: a structured data processing system for septic shock diagnosis and treatment image information, the system comprising: a processor, an image acquisition module, a multimodal data fusion module, a dynamic feature extraction module, a structured processing engine, a diagnosis and treatment decision module, an abnormality detection module, an early warning feedback module and a performance monitoring module; The image acquisition module is used to obtain the patient's medical imaging data, including CT images, ultrasound images and hemodynamic time series data; the multimodal data fusion module performs time-space alignment and feature normalization processing on the medical imaging data to generate multimodal fusion data; the dynamic feature extraction module performs time series analysis and spatial feature extraction on the multimodal fusion data through a preset deep learning model to generate a dynamic feature vector; the structured processing engine maps the dynamic feature vector to a predefined clinical data template to generate a structured diagnosis and treatment record; The abnormality detection module monitors the physiological parameters, hemodynamic indicators and imaging features in the structured diagnosis and treatment records in real time. If it is detected that the parameters deviate from the preset threshold range, an abnormal signal is generated and sent to the processor; the diagnosis and treatment decision module calls the preset clinical pathway algorithm according to the abnormal signal to generate adaptive diagnosis and treatment recommendations; the early warning feedback module receives the abnormal signal and triggers the visual early warning interface; the performance monitoring module records the response time of the abnormal signal and the number of revisions of the diagnosis and treatment decision, and generates a system performance log.

[0009] Preferably, the specific operation process of the dynamic feature extraction module is as follows: The long short-term memory network is used to model the time dependence of the hemodynamic time series data and output the time series feature vector. The multi-scale features of CT images and ultrasound images are extracted through a three-dimensional convolutional neural network, and the spatial feature vector is output. The time series feature vector and the spatial feature vector are tensor-concatenated, and the dynamic feature vector is generated after dimensionality reduction through a fully connected layer.

[0010] Preferably, the structured processing engine includes a template matching submodule and a semantic parsing submodule; The template matching submodule maps the physiological parameters in the dynamic feature vector to cardiac function indicators, microcirculation status and organ perfusion labels based on the data fields defined in the clinical guidelines; the semantic parsing submodule performs entity recognition and relationship extraction on the unstructured text report through natural language processing technology to generate standardized semantic data; The structured medical record is generated by merging standardized semantic data with mapped data fields.

[0011] Preferably, the specific operation process of the anomaly detection module is as follows: Dynamic threshold intervals are set for cardiac function indicators, microcirculation status and organ perfusion labels respectively, and the dynamic threshold intervals are adaptively adjusted based on patient historical data and group statistical data; The deviation between the current parameter value and the dynamic threshold interval is calculated in real time. If the deviation of any parameter exceeds the preset tolerance coefficient, an abnormal signal is generated and the abnormal type is marked.

[0012] Preferably, the diagnosis and treatment decision module includes a path optimization submodule and a conflict resolution submodule; The path optimization submodule adopts a reinforcement learning algorithm to generate a priority ranking of diagnosis and treatment plans according to the abnormality type and individual differences of patients; the conflict resolution submodule coordinates the resource conflicts and timing conflicts among multiple plans through a fuzzy logic algorithm and outputs the final diagnosis and treatment recommendations.

[0013] Preferably, the specific analysis process of the performance monitoring module is as follows: Record the response time from the generation of abnormal signals to the output of diagnosis and treatment recommendations. If the response time exceeds the preset response threshold, it is marked as a delay event. Count the number of times the diagnosis and treatment recommendations are clinically revised. If the number of revisions exceeds the preset revision threshold, it will be marked as an inefficient decision-making event. The frequencies of delay events and inefficient decision-making events are weighted and calculated to generate a system performance score.

[0014] Preferably, the processor is in communication with a data tracing module, and the data tracing module is used to: Label each data item in the structured medical record with its source, including imaging equipment model, acquisition time, and operator information; When the system performance score is lower than the preset qualified threshold, data traceability analysis is triggered to locate abnormal nodes of the data source of delayed events or inefficient decision-making events.

[0015] Preferably, the early warning feedback module includes a multi-level early warning submodule; The multi-level warning submodule divides the warning level according to the severity of the abnormal signal. The first-level warning triggers a pop-up window prompt on the interface, the second-level warning is sent to the mobile terminal simultaneously, and the third-level warning is directly connected to the emergency system.

[0016] Preferably, the system further comprises a model iteration module, wherein the model iteration module is used for: Regularly collect system performance logs and clinical feedback data, update deep learning models and reinforcement learning strategies through online learning algorithms, and optimize the accuracy of dynamic feature extraction and diagnosis and treatment decisions.

[0017] Preferably, the present invention also includes a method for processing structured data of septic shock diagnosis and treatment image information, the method comprising the following steps: Acquire multimodal medical imaging data through an image acquisition module; Perform spatiotemporal alignment and feature normalization on multimodal data to generate fused data; Use deep learning models to extract spatiotemporal dynamic features of fused data and generate feature vectors; Map feature vectors to clinical data templates to generate structured diagnosis and treatment records; Real-time monitoring of parameter deviations in diagnosis and treatment records, triggering abnormal signals and adaptive diagnosis and treatment recommendations; Record system response events and generate performance logs, and locate abnormal nodes through data tracing; A multi-level feedback mechanism is triggered according to the warning level, and the algorithm performance is optimized through model iteration.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention acquires a variety of medical imaging data through an image acquisition module, and uses a multimodal data fusion module to perform spatiotemporal alignment and feature normalization processing to generate multimodal fusion data. This process realizes the complementary advantages of different modal data, so that the data can more comprehensively reflect the patient's condition. For example, the anatomical structure information of CT images, the real-time blood flow information of ultrasound images, and the dynamic change information of hemodynamic time series data are combined to provide a rich and accurate data basis for subsequent analysis. The dynamic feature extraction module further performs time series analysis and spatial feature extraction on the fused data to generate a dynamic feature vector. This deep data mining method can more accurately capture the trend of disease changes. Compared with the traditional single data processing method, it greatly improves the accuracy and effectiveness of data processing, which helps doctors understand the patient's condition more comprehensively and deeply.

[0019] The structured processing engine maps dynamic feature vectors to predefined clinical data templates to generate structured medical records. This makes the originally complex and disordered medical data standardized and orderly, making it convenient for doctors to quickly access and analyze. Doctors no longer need to spend a lot of time looking for key information in unstructured text records. They can directly obtain patients' physiological parameters, hemodynamic indicators, imaging characteristics, and diagnostic results from structured medical records, which significantly improves the efficiency of diagnosis and treatment. At the same time, structured data is more convenient for the information management and big data analysis of medical data, providing strong support for medical research and clinical decision-making, and is conducive to the realization of precision medicine based on big data. Accurate anomaly detection and timely warning: The anomaly detection module monitors the various parameters in the structured medical records in real time, and can more accurately judge whether the parameters are abnormal through the dynamic threshold interval adaptively adjusted based on the patient's historical data and group statistical data. When it is detected that the parameter deviates from the preset threshold range, an abnormal signal is immediately generated and the abnormal type is marked. This function helps doctors to detect subtle changes in the patient's condition in a timely manner. The early warning feedback module divides the warning into multiple levels according to the severity of the abnormal signal, from pop-up prompts on the interface to connecting to the emergency system, ensuring that doctors can take appropriate intervention measures at the first time, effectively avoiding the deterioration of the disease and improving the success rate of patient treatment.

[0020] The diagnosis and treatment decision module calls the preset clinical pathway algorithm according to the abnormal signal, and generates adaptive diagnosis and treatment recommendations based on the individual differences of the patients. The pathway optimization submodule uses a reinforcement learning algorithm to prioritize the diagnosis and treatment plans, fully considering the disease characteristics and individual differences of different patients, making the diagnosis and treatment plans more targeted. The conflict resolution submodule coordinates the resource conflicts and timing conflicts between multiple plans through a fuzzy logic algorithm to ensure that the final output diagnosis and treatment recommendations are feasible. This personalized and precise diagnosis and treatment decision support method can help doctors develop more scientific and reasonable diagnosis and treatment plans, improve treatment effects, and improve patient prognosis.

[0021] The performance monitoring module records the response time of abnormal signals and the number of revisions to diagnosis and treatment decisions, and generates a system performance log. By analyzing these data, the system's performance can be comprehensively evaluated and problems in the system's operation can be discovered in a timely manner. When the system performance score is lower than the preset qualified threshold, the data traceability module traces the data and locates the abnormal nodes that are the data sources of delayed events or inefficient decision-making events, providing a clear direction for system optimization. The model iteration module regularly collects system performance logs and clinical feedback data, updates deep learning models and reinforcement learning strategies through online learning algorithms, continuously optimizes dynamic feature extraction and the accuracy of diagnosis and treatment decisions, realizes continuous optimization and self-improvement of the system, and ensures that the system always maintains an efficient and reliable operating state. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a working principle diagram of the septic shock diagnosis and treatment image information structured data processing system of the present invention; Figure 2 Flowchart for conflict resolution and plan optimization in diagnosis and treatment decisions; Figure 3 Flowchart for iterative optimization of system models; Figure 4 Flow chart of the method for processing data for diagnosis and treatment of septic shock. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] See also Figure 1-4 The present invention provides a technical solution: a septic shock diagnosis and treatment image information structured data processing system, the system comprising: Image acquisition module: responsible for acquiring the patient's medical imaging data, including CT images, ultrasound images and hemodynamic time series data. These data are collected from the hospital's imaging equipment and monitoring instruments to provide basic information for subsequent analysis.

[0025] Multimodal data fusion module: performs spatiotemporal alignment and feature normalization on the collected medical imaging data. Spatiotemporal alignment is to ensure the consistency of different modal data in time and space, and feature normalization is to unify the value ranges of different features to facilitate subsequent analysis and processing, and finally generate multimodal fusion data.

[0026] Dynamic feature extraction module: Use the preset deep learning model to perform time series analysis and spatial feature extraction on multimodal fusion data. Time series analysis is used to mine the changing patterns of data over time, while spatial feature extraction focuses on the spatial information in the image, and finally generates a dynamic feature vector.

[0027] Structured processing engine: Map dynamic feature vectors to predefined clinical data templates to generate structured diagnosis and treatment records. The clinical data template is designed based on clinical guidelines and actual diagnosis and treatment needs, and can present complex medical data in a structured form, making it easier for doctors to view and analyze.

[0028] Abnormality detection module: Real-time monitoring of physiological parameters, hemodynamic indicators and imaging features in structured diagnosis and treatment records. Once it is detected that the parameters deviate from the preset threshold range, an abnormal signal is generated and sent to the processor to detect the patient's abnormal condition in time.

[0029] Diagnosis and treatment decision module: Based on abnormal signals, the preset clinical pathway algorithm is called to generate adaptive diagnosis and treatment suggestions. These algorithms are developed based on a large amount of clinical experience and research results, and can provide doctors with scientific and reasonable diagnosis and treatment references.

[0030] Early warning feedback module: Receives abnormal signals and triggers a visual early warning interface. Through intuitive visualization, doctors can quickly understand the patient's abnormal condition and take appropriate measures in a timely manner.

[0031] Performance monitoring module: records the response time of abnormal signals and the number of revisions of diagnosis and treatment decisions, and generates system performance logs. By recording and analyzing these data, the performance of the system can be evaluated and a basis for system optimization can be provided.

[0032] The present invention will be further described below in conjunction with Examples 1 to 5: Embodiment 1: The purpose of this embodiment is to elaborate on how the dynamic feature extraction module extracts features from data of different modalities and generates dynamic feature vectors, so as to provide accurate data support for subsequent structured processing and diagnosis and treatment decisions.

[0033] The operation process of the dynamic feature extraction module is as follows: Long short-term memory network (LSTM) is used to model the time dependency of hemodynamic time series data. Hemodynamic time series data contains important information about the patient's cardiac function, vascular resistance, etc. that changes over time. Long short-term memory network can effectively capture the long-term dependency in the data and overcome the gradient vanishing and gradient exploding problems that traditional neural networks are prone to when processing time series data. Assume that the hemodynamic time series data is a sequence , the input of the LSTM network is , through a series of gating mechanisms, such as the forget gate , Input Gate and output gate , for cell status Update and finally output the time series feature vector The core formula is: in, is the Sigmoid function, represents element-wise multiplication, and are the weight matrix and bias vector respectively.

[0034] The CT images and ultrasound images are respectively subjected to multi-scale feature extraction by three-dimensional convolutional neural networks (3D-CNN). CT images and ultrasound images contain rich information about the structure and pathology of human organs. The three-dimensional convolutional neural network can perform convolution operations on images in three-dimensional space and extract spatial features of different scales. Taking the CT image as an example, after processing by multiple convolutional layers and pooling layers, the spatial feature vector is finally output. Assume that the CT image is , after the convolutional layer and pooling layer The output space feature vector .

[0035] The temporal feature vector and the spatial feature vector are concatenated and the dynamic feature vector is generated after dimensionality reduction through the fully connected layer. And the spatial feature vector obtained from CT images and ultrasound images , Tensor splicing is performed to obtain a high-dimensional vector. Then, the fully connected layer is used to reduce the dimensionality of the vector, remove redundant information, and generate the final dynamic feature vector. .

[0036] Embodiment 2: The purpose of this embodiment is to explain in detail how the structured processing engine converts dynamic feature vectors into structured medical records, making medical data easier for doctors to view and analyze, thereby improving the efficiency of diagnosis and treatment.

[0037] The structured processing engine includes a template matching submodule and a semantic parsing submodule.

[0038] The template matching submodule classifies the physiological parameters in the dynamic feature vector into cardiac function indicators, microcirculation status, and organ perfusion labels based on the data fields defined in the clinical guidelines. Clinical guidelines are specifications and standards summarized after long-term practice and research in the medical field, which define a series of data fields related to the diagnosis and treatment of septic shock. The template matching submodule classifies the physiological parameters in the dynamic feature vector based on these data fields. For example, the parameters reflecting the systolic and diastolic functions of the heart are mapped to the cardiac function indicator label, the parameters related to the microcirculation blood flow status are mapped to the microcirculation status label, and the parameters reflecting the organ blood perfusion status are mapped to the organ perfusion label.

[0039] The semantic parsing submodule uses natural language processing technology to perform entity recognition and relationship extraction on unstructured text reports to generate standardized semantic data. In medical records, there are a large number of unstructured text reports, such as doctors' diagnosis descriptions, examination result descriptions, etc. The semantic parsing submodule uses natural language processing technology to analyze these texts. First, entity recognition is performed to identify key entities in the text, such as disease names, symptoms, examination items, etc.; then relationship extraction is performed to determine the relationships between these entities, such as causal relationships, concomitant relationships, etc. Through these operations, unstructured text is converted into standardized semantic data. For example, for the text "The patient has fever and hypotension, septic shock is considered, and cardiac ultrasound shows weakened left ventricular systolic function", the semantic parsing submodule can identify entities such as "fever", "hypotension", "septic shock", "weakened left ventricular systolic function", and extract the relationships between them to generate standardized semantic data.

[0040] The structured medical record is generated by combining the standardized semantic data with the mapped data fields. The standardized semantic data generated by the semantic parsing submodule and the data fields mapped by the template matching submodule are integrated to form a complete structured medical record. In this way, doctors can quickly obtain the patient's condition information through the structured medical record, including various physiological parameters, diagnosis results and the relationship between them.

[0041] Embodiment 3: The operation process of the abnormal detection module is as follows: dynamic threshold intervals are set for cardiac function indicators, microcirculation status, and organ perfusion labels. These dynamic threshold intervals are not fixed, but are adaptively adjusted based on the patient's historical data and group statistical data. The patient's condition is dynamically changing, and there are individual differences between different patients, so fixed thresholds cannot accurately reflect the patient's true condition. By collecting the patient's historical data, including cardiac function indicators, microcirculation status, and organ perfusion labels at different time points, combined with group statistical data, such as the mean and standard deviation of various indicators of patients with the same condition, the dynamic threshold interval of each patient is determined. For example, for the ejection fraction in a patient's cardiac function indicators, a dynamic threshold interval that changes over time is determined based on its historical ejection fraction data and the ejection fraction statistics of patients with the same type of septic shock.

[0042] The deviation between the current parameter value and the dynamic threshold interval is calculated in real time. If the deviation of any parameter exceeds the preset tolerance coefficient, an abnormal signal is generated and the abnormal type is marked. The deviation can be calculated in a variety of ways, such as calculating the ratio of the difference between the current parameter value and the boundary value of the dynamic threshold interval to the width of the threshold interval. Assume that the current parameter value is , the dynamic threshold interval is , the preset tolerance factor is , the deviation calculation formula is: When the deviation is greater than When the parameter is abnormal, it indicates that the current parameter is abnormal. The abnormality detection module generates an abnormal signal and marks the abnormality type according to the label category to which the parameter belongs, such as "abnormal cardiac function index", "abnormal microcirculation state" or "abnormal organ perfusion".

[0043] Embodiment 4: The purpose of this embodiment is to explain in detail how the diagnosis and treatment decision module generates reasonable diagnosis and treatment plans based on abnormality types and individual differences of patients, coordinates conflicts between multiple plans, provides final diagnosis and treatment recommendations to doctors, and assists doctors in making scientific diagnosis and treatment decisions.

[0044] The diagnosis and treatment decision module includes a path optimization submodule and a conflict resolution submodule.

[0045] The path optimization submodule uses a reinforcement learning algorithm to generate a priority ranking of diagnosis and treatment plans based on the abnormality type and individual differences of patients. Reinforcement learning is a method of learning the optimal strategy by interacting with the environment and obtaining rewards. In the present invention, the path optimization submodule takes the abnormality type and individual differences of patients as input, explores different diagnosis and treatment plans through a reinforcement learning algorithm, and obtains rewards based on the effect of these plans on improving the patient's condition. For example, for patients with different types of abnormalities (such as abnormal cardiac function, abnormal microcirculation, etc.) and different physical conditions (age, underlying diseases, etc.), try different drug treatment plans, fluid resuscitation plans, etc., and give corresponding rewards based on changes in the patient's condition indicators (such as blood pressure recovery, improved organ perfusion, etc.). Through continuous learning and optimization, a priority ranking of diagnosis and treatment plans for different situations is generated to provide a reference for doctors.

[0046] The conflict resolution submodule coordinates resource conflicts and timing conflicts among multiple plans through fuzzy logic algorithms, and outputs the final diagnosis and treatment recommendations. In the actual diagnosis and treatment process, there may be multiple diagnosis and treatment plans that are applicable at the same time, but there are resource conflicts (such as insufficient drug supply, limited medical equipment, etc.) or timing conflicts (such as conflicts in the order of implementation of different treatment measures). Fuzzy logic algorithms can process fuzzy and uncertain information, fuzzify the conflicting factors of different plans, and make inferences and decisions based on expert experience and preset rules. For example, for the conflict between drug treatment plans and surgical treatment plans in terms of resources and time, the fuzzy logic algorithm comprehensively considers various factors, such as the urgency of the patient's condition, resource availability, etc., to determine the final diagnosis and treatment recommendations and ensure the feasibility and effectiveness of the diagnosis and treatment plans.

[0047] Embodiment 5: The analysis process of the performance monitoring module is as follows: record the response time from the generation of abnormal signals to the output of diagnosis and treatment recommendations. If the response time exceeds the preset response threshold, it is marked as a delayed event. The preset response threshold is set based on actual clinical needs and experience. Generally speaking, patients with septic shock are in critical condition and need to make diagnosis and treatment decisions quickly. If the system takes too long to give diagnosis and treatment recommendations after detecting an abnormal signal, it may affect the patient's treatment effect. For example, set the preset response threshold to 5 minutes. When the abnormal signal is generated, if the system takes more than 5 minutes to output diagnosis and treatment recommendations, the event will be marked as a delayed event.

[0048] The number of times the diagnosis and treatment recommendations are clinically revised is counted. If the number of revisions exceeds the preset revision threshold, it is marked as an inefficient decision-making event. The diagnosis and treatment recommendations are generated by the system based on the algorithm, but in actual clinical applications, doctors may revise the diagnosis and treatment recommendations based on their own experience and further observation of the patient. If the number of revisions is too many, it means that the diagnosis and treatment recommendations generated by the system may be inaccurate or unreasonable. The preset revision threshold can be adjusted according to the actual situation and clinical data of different hospitals. For example, after statistical analysis, in a certain hospital, if the number of revisions of the diagnosis and treatment recommendations exceeds 10 times within a month, it is marked as an inefficient decision-making event.

[0049] The frequency of delay events and inefficient decision-making events is weighted to generate a system performance score. Different types of events have different impacts on system performance, so it is necessary to weight the frequency of delay events and inefficient decision-making events. For example, if the weight of delay events is set to 0.6 and the weight of inefficient decision-making events is set to 0.4, the system performance score calculation formula is: The processor is communicatively connected to the data traceability module, which is used to label the source of each data item in the structured diagnosis and treatment record, including the imaging device model, acquisition time, and operator information. When the system performance score is lower than the preset qualified threshold, the data traceability analysis is triggered to locate the abnormal node of the data source of the delayed event or inefficient decision-making event. By marking the data source label, the data collection process and related information can be traced. When there is a problem with system performance, the data traceability module can help find the root cause of the problem. For example, if it is found that a delay event is caused by a specific imaging device collecting data too slowly, or an inefficient decision-making event is caused by an operator entering data incorrectly, improvements can be made to these problems.

[0050] The early warning feedback module includes a multi-level early warning sub-module. The multi-level early warning sub-module divides the early warning level according to the severity of the abnormal signal. The first-level early warning triggers a pop-up window prompt on the interface, the second-level early warning is sent to the mobile terminal simultaneously, and the third-level early warning is directly connected to the emergency system. The severity of the abnormal signal can be evaluated based on factors such as the size of the deviation and the importance of the abnormal parameters. The first-level early warning is suitable for some relatively minor abnormal situations, and prompts the doctor through a pop-up window on the system interface; the second-level early warning is for more serious abnormalities. In addition to the prompt on the interface, the early warning information will also be sent to the doctor's mobile terminal to ensure that the doctor can know it in time; the third-level early warning is for emergency situations that endanger the patient's life, directly connecting to the emergency system and starting the emergency rescue procedure.

[0051] The system also includes a model iteration module, which is used to regularly collect system performance logs and clinical feedback data, update deep learning models and reinforcement learning strategies through online learning algorithms, and optimize the accuracy of dynamic feature extraction and diagnosis and treatment decisions. With the continuous accumulation of clinical data and the operation of the system, it is necessary to continuously optimize models and algorithms to improve the performance of the system. The model iteration module collects data from system performance logs, such as information on delayed events, inefficient decision-making events, and clinical feedback data, such as doctors' evaluation of diagnosis and treatment recommendations and patients' treatment effects, and uses online learning algorithms to update deep learning models and reinforcement learning strategies. For example, online learning algorithms such as stochastic gradient descent are used to adjust the parameters of the deep learning model so that the model can extract dynamic features more accurately; optimize reinforcement learning strategies to improve the accuracy and rationality of diagnosis and treatment decisions.

[0052] The present invention also includes a method for processing structured data of septic shock diagnosis and treatment image information, the method comprising: Acquire multimodal medical imaging data through an image acquisition module; Perform spatiotemporal alignment and feature normalization on multimodal data to generate fused data; Use deep learning models to extract spatiotemporal dynamic features of fused data and generate feature vectors; Map feature vectors to clinical data templates to generate structured diagnosis and treatment records; Real-time monitoring of parameter deviations in diagnosis and treatment records, triggering abnormal signals and adaptive diagnosis and treatment recommendations; Record system response events and generate performance logs, and locate abnormal nodes through data tracing; A multi-level feedback mechanism is triggered according to the warning level, and the algorithm performance is optimized through model iteration.

[0053] The implementation of this method refers to the above embodiment and will not be described in detail in the specification.

[0054] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0055] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A septic shock diagnosis and treatment image information structured data processing system, characterized in that: It includes a processor, an image acquisition module, a multimodal data fusion module, a dynamic feature extraction module, a structured processing engine, a diagnosis and treatment decision module, an anomaly detection module, a warning feedback module and a performance monitoring module; The image acquisition module is used to obtain the patient's medical imaging data, including CT images, ultrasound images and hemodynamic time series data; the multimodal data fusion module performs time-space alignment and feature normalization processing on the medical imaging data to generate multimodal fusion data; the dynamic feature extraction module performs time series analysis and spatial feature extraction on the multimodal fusion data through a preset deep learning model to generate a dynamic feature vector; the structured processing engine maps the dynamic feature vector to a predefined clinical data template to generate a structured diagnosis and treatment record; The abnormality detection module monitors the physiological parameters, hemodynamic indicators and imaging features in the structured diagnosis and treatment records in real time, and generates an abnormality signal and sends it to the processor if it detects that the parameters deviate from the preset threshold range; The diagnosis and treatment decision module calls a preset clinical pathway algorithm according to the abnormal signal to generate adaptive diagnosis and treatment suggestions; The early warning feedback module receives the abnormal signal and triggers the visual early warning interface; The performance monitoring module records the response time of abnormal signals and the number of revisions of diagnosis and treatment decisions, and generates a system performance log.

2. The septic shock diagnosis and treatment image information structured data processing system according to claim 1, characterized in that: The specific operation process of the dynamic feature extraction module is as follows: The long short-term memory network is used to model the time dependence of the hemodynamic time series data and output the time series feature vector. The multi-scale features of CT images and ultrasound images are extracted through a three-dimensional convolutional neural network, and the spatial feature vector is output. The time series feature vector and the spatial feature vector are tensor-concatenated, and the dynamic feature vector is generated after dimensionality reduction through a fully connected layer.

3. The septic shock diagnosis and treatment image information structured data processing system according to claim 1, characterized in that: The structured processing engine includes a template matching submodule and a semantic parsing submodule; The template matching submodule maps the physiological parameters in the dynamic feature vector to cardiac function indicators, microcirculation status and organ perfusion labels based on the data fields defined in the clinical guidelines; the semantic parsing submodule performs entity recognition and relationship extraction on the unstructured text report through natural language processing technology to generate standardized semantic data; The structured medical record is generated by merging standardized semantic data with mapped data fields.

4. The septic shock diagnosis and treatment image information structured data processing system according to claim 3, characterized in that: The specific operation process of the anomaly detection module is as follows: Dynamic threshold intervals are set for cardiac function indicators, microcirculation status and organ perfusion labels respectively, and the dynamic threshold intervals are adaptively adjusted based on patient historical data and group statistical data; The deviation between the current parameter value and the dynamic threshold interval is calculated in real time. If the deviation of any parameter exceeds the preset tolerance coefficient, an abnormal signal is generated and the abnormal type is marked.

5. The septic shock diagnosis and treatment image information structured data processing system according to claim 4, characterized in that: The diagnosis and treatment decision module includes a path optimization submodule and a conflict resolution submodule; The path optimization submodule adopts a reinforcement learning algorithm to generate a priority ranking of diagnosis and treatment plans according to the abnormality type and individual differences of patients; the conflict resolution submodule coordinates the resource conflicts and timing conflicts among multiple plans through a fuzzy logic algorithm and outputs the final diagnosis and treatment recommendations.

6. The septic shock diagnosis and treatment image information structured data processing system according to claim 5, characterized in that: The specific analysis process of the performance monitoring module is as follows: Record the response time from the generation of abnormal signals to the output of diagnosis and treatment recommendations. If the response time exceeds the preset response threshold, it is marked as a delay event. Count the number of times the diagnosis and treatment recommendations are clinically revised. If the number of revisions exceeds the preset revision threshold, it will be marked as an inefficient decision-making event. The frequencies of delay events and inefficient decision-making events are weighted and calculated to generate a system performance score.

7. The septic shock diagnosis and treatment image information structured data processing system according to claim 6, characterized in that: The processor is in communication with a data tracing module, and the data tracing module is used to: Label each data item in the structured medical record with its source, including imaging equipment model, acquisition time, and operator information; When the system performance score is lower than the preset qualified threshold, data traceability analysis is triggered to locate abnormal nodes of the data source of delayed events or inefficient decision-making events.

8. The septic shock diagnosis and treatment image information structured data processing system according to claim 7, characterized in that: The early warning feedback module includes a multi-level early warning submodule; The multi-level warning submodule divides the warning level according to the severity of the abnormal signal. The first-level warning triggers a pop-up window prompt on the interface, the second-level warning is sent to the mobile terminal simultaneously, and the third-level warning is directly connected to the emergency system.

9. The septic shock diagnosis and treatment image information structured data processing system according to claim 8, characterized in that: The system further comprises a model iteration module, wherein the model iteration module is configured to: Regularly collect system performance logs and clinical feedback data, update deep learning models and reinforcement learning strategies through online learning algorithms, and optimize the accuracy of dynamic feature extraction and diagnosis and treatment decisions.

10. A method for processing structured data of septic shock diagnosis and treatment image information, characterized in that: The following steps are involved: Acquire multimodal medical imaging data through an image acquisition module; Perform spatiotemporal alignment and feature normalization on multimodal data to generate fused data; Use deep learning models to extract spatiotemporal dynamic features of fused data and generate feature vectors; Map feature vectors to clinical data templates to generate structured diagnosis and treatment records; Real-time monitoring of parameter deviations in diagnosis and treatment records, triggering abnormal signals and adaptive diagnosis and treatment recommendations; Record system response events and generate performance logs, and locate abnormal nodes through data tracing; A multi-level feedback mechanism is triggered according to the warning level, and the algorithm performance is optimized through model iteration.

Images