Medical examination report cloud storage management system

Abstract

The invention discloses a medical examination report cloud storage management system, and relates to the technical field of physical training monitoring, and the service system comprises a basic physiological data collection module, a historical record module, a heart rate algorithm model module, a heart rate response characteristic analysis module, and a personalized heart rate interval setting module. Through high-precision heart rate monitoring equipment and an advanced data analysis technology, real-time and accurate monitoring of heart rate changes of children is achieved, the system can capture heart rate data of the children in the training process in real time, the data are analyzed through a machine learning algorithm, the heart rate response mode and the training adaptability of the children are recognized, and the training accuracy of the children is improved. On the basis of the analysis results, timely and effective training guidance and suggestions are provided for coaches and parents, the coaches and parents are helped to better understand and master the training state of children, and therefore the overall training effect is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to a cloud storage management system for medical examination reports. Background Art

[0002] With the rapid advancement of medical technology and the popularization of digital medicine, the number of medical examination reports, as an important basis for patient diagnosis, treatment and prognosis assessment, has exploded. These reports are not only diverse in types, including blood tests, imaging tests, pathological examinations and other categories, but also in various data forms, including both precise structured data and detailed unstructured data. Therefore, how to effectively classify, manage and store these reports has become an increasingly prominent issue.

[0003] However, the existing medical report management system lacks the function of automatically classifying reports according to examination type, urgency and importance. As a result, in emergency situations, medical staff are unable to quickly find the required reports, delaying treatment decisions. The lack of intelligent classification also makes report retrieval cumbersome and inefficient, requiring manual screening of large amounts of information, increasing the workload of medical staff and the risk of medical errors. In addition, the existing medical system is unable to deeply explore the complex relationship between examination results and disease diagnosis, treatment plans and prognosis assessments, making it difficult to distinguish truly important reports.

[0004] In summary, the current management of medical examination reports faces challenges such as inaccurate classification, inaccurate urgency assessment, and shallow importance rating. These problems not only affect the efficiency and quality of medical services, but also pose a potential threat to the diagnosis and treatment safety of patients. Therefore, the development of an intelligent data classification system that can automatically classify and store reports according to examination type, urgency, and importance is of great significance to improving the level of medical report management and ensuring the safety of patient diagnosis and treatment. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a medical examination report cloud storage management system, which can automatically classify and store according to the examination type, urgency and importance, which can not only improve the efficiency and accuracy of data management, but also ensure that urgent and important reports can be processed in a timely manner, thereby helping to improve the quality and timeliness of medical services. In addition, by adopting a distributed cloud storage architecture, the system can also ensure the security and reliability of data and reduce the risk of data loss.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a medical examination report cloud storage management system, the components of which include: an examination report acquisition module, an examination type identification module, an urgency assessment module, an importance rating module, and a cloud storage module;

[0007] The examination report acquisition module is connected to the medical examination equipment and information system within the hospital to collect the original medical data of various medical examination reports, including structured data and unstructured data, and has a built-in data verification mechanism to perform preliminary verification and cleaning on the collected data;

[0008] The examination type recognition module recognizes and classifies the examination type from the report text, and classifies the report into different types of blood examination, imaging examination, pathological examination, and electrocardiogram examination by learning and pattern recognition of a large number of historical reports;

[0009] The urgency assessment module comprehensively considers the patient's symptoms, the critical values ​​of the examination indicators, and the clinician's annotations, and divides the report into three categories: urgent, general, and non-urgent;

[0010] The importance rating module analyzes the impact of the examination results on disease diagnosis, treatment plan formulation and prognosis assessment, and then distinguishes between important and ordinary reports;

[0011] The cloud storage module adopts a distributed storage architecture, and adopts differentiated storage strategies for reports of different categories according to their importance and access frequency, including short-term storage area, long-term storage area, and archive storage area.

[0012] Furthermore, the examination report acquisition module communicates with the medical examination equipment within the hospital to obtain examination data, integrates with the hospital information system, obtains electronic medical records and report information, and comprehensively collects the original medical data of various medical examination reports, including structured data, i.e., the numerical values ​​and date and time information of various examination indicators, and unstructured data, i.e., the text description, images, and videos of the pathology report. The collected data will immediately enter the built-in data verification mechanism for preliminary verification and cleaning. The verification mechanism includes:

[0013] Duplicate data processing: For duplicate data, keep the latest valid records and delete and mark the old records;

[0014] Outlier processing: For obviously abnormal values, they will be rechecked with the data source, and judged and corrected based on the distribution range of historical data.

[0015] Furthermore, the examination type identification module extracts text data from the original medical examination report received by the examination report acquisition module, preprocesses the text data, including removing irrelevant symbols, special characters, word segmentation, and part-of-speech tagging, and converts the input examination report text into a series of feature vector representations, namely {v1, v2, …, v T}, where T is the length of the inspection report text, and the query vector q, key vector k and value vector v are defined. The query vector q is regarded as an abstract representation of the current feature of interest, the key vector k is used to measure the relevance of each input feature vector to the query vector, and the value vector v is the vector that actually carries the information. The attention score e is calculated. ij , that is, e ij =f(q i ,k j ), function f(q i ,k j ) is the dot product operation, that is Measure the reported data similarity between the query vector q and the key vector k, normalize the attention score, and get the attention weight α ij ,Right now The feature vectors are weighted and summed according to the attention weights to obtain the final attention output y, that is, y = ∑ j α ij v j ,According to the calculation results, we can focus on the key parts in the input text in a targeted manner, so as to more accurately identify and classify the inspection type.

[0016] Furthermore, the examination report acquisition module identifies the medical examination report by paying attention to the final result output, emphasizing the key words and phrases in the report related to the specific examination type, namely, blood test, imaging examination, pathological examination, electrocardiogram examination, and its classification results include:

[0017] Blood tests: Automatically identify and weight blood parameters mentioned in reports, i.e. red blood cell count, platelets;

[0018] Imaging examination: Assign higher weights to imaging terms that appear in reports, i.e., CT, MRI, and X-ray, to help distinguish different types of imaging reports, and capture descriptive statements for descriptions of imaging results, i.e., lung nodules and fracture lines are clearly visible;

[0019] Pathological examination: highlight the description of pathological changes in the report, i.e., cancer cells and obvious inflammatory response found in tissue samples, extract the characteristics of pathological changes in the report, i.e., cancer cell infiltration and inflammatory response, and give higher weight to the diagnostic criteria contained in the pathological report, i.e., histological grade and tumor stage, for accurate classification;

[0020] Electrocardiogram examination: Pay more attention to the key information in the electrocardiogram report, namely ST segment elevation, abnormal QRS complex morphology, and arrhythmia.

[0021] Furthermore, the specific steps of the emergency assessment module are:

[0022] Data preparation: Receive raw data from the examination report acquisition module and evaluate the urgency of symptoms in the examination report based on the classification results of the examination type identification module;

[0023] Feature extraction: Extract the description of the patient's symptoms from the report text, i.e., dyspnea and persistent high fever, extract the values ​​of key examination indicators, i.e., white blood cell count and blood sugar value in blood tests, and extract the emergency instructions marked by the doctor in the report;

[0024] Urgency assessment: symptom severity score, i.e. scoring the severity of symptoms based on the symptom manifestation characteristics; indicator criticality score, i.e. assessing whether the indicator exceeds the normal range and its criticality based on the characteristics of the examination indicator; and doctor's weight, i.e. the emergency description marked by the doctor, which will be given additional weight;

[0025] Comprehensive score: Calculates a comprehensive score for the urgency assessment and classifies reports into urgent, general, and non-urgent based on the comprehensive score.

[0026] Furthermore, in the specific steps of the urgency assessment module, the urgency assessment combines the frequency of occurrence, severity and correlation factors of the symptoms to create a scoring model, assigns a severity level to each symptom based on medical knowledge, i.e., mild, moderate, severe, and assigns a severity level value to each severity level. For a single symptom S i , that is, S i =Freq(s i )×Severity(s i ), where Freq(s i ) is a symptom i The frequency of occurrence, Severity(s i ) is a symptom i The severity level values ​​for the two symptoms s i and j , P(s i ,s j ) are symptoms i and j The probability of simultaneous occurrence, P(s i ) and P(s j ) are symptoms i and j The probability of a single symptom is calculated, a symptom co-occurrence matrix is ​​constructed, the probability of different symptoms occurring at the same time is recorded, the score of each symptom is calculated, and the comprehensive symptom severity score is calculated according to the correlation between the symptoms. That is, the comprehensive symptom severity score S is: n is the number of symptoms in the symptom list, S i Is a single symptom iRating, PMI(s i ,s j ) are symptoms i and j The point mutual information between i ) and Severity(s j ) are symptoms i and j The severity level value.

[0027] Furthermore, the urgency assessment module combines the symptom severity score, the indicator criticality score and the doctor's weighted score to calculate a comprehensive score, that is, C = w S ·S+w I I+w D D, C are comprehensive scores, w S 、w I and w D are the weight coefficients of symptom severity score, indicator criticality score and doctor annotation weight, S, I and D are the symptom severity score, indicator criticality score and doctor annotation score, respectively. Corresponding severity scoring standards are formulated for each type of examination result, including blood test, imaging test, pathological examination and electrocardiogram examination. For each type of examination result c i , we define its severity score C i ,Right now m is the number of abnormal indicators, a j is the abnormal index j, Freq(a j ) is the abnormal indicator a j The frequency of occurrence in this type of inspection, Severity (a j ) is the abnormal indicator a j The severity level value of the comprehensive inspection results is expressed as: n is the number of check types, C i is the severity score of the i-th inspection result. According to the comprehensive score T, a threshold is set to divide the urgency category, that is, for T>T 紧急 , it is considered urgent. 非紧急 <T≤T 紧急 , then it is judged as general, for T≤T 非紧急 , it is judged as non-urgent.

[0028] Furthermore, the importance rating module collects the annotated and verified historical medical examination report data and the data identified by the examination type identification module and the urgency assessment module. These data contain detailed information on examination results, disease diagnosis, treatment plans and prognosis assessment. Features are extracted from the training set, including the numerical value and description of the examination results, the correlation with the disease diagnosis, the influence weight on the treatment plan selection, and the predictive value for the prognosis assessment. For each feature x i , calculate its impact on disease diagnosis, treatment selection and prognosis assessment, that is, N is the number of features, w j is the weight of feature j, Impact ij is feature x i The degree of influence of feature j is used to define the objective function of the comprehensive influence on disease diagnosis, treatment plan selection and prognosis evaluation, that is, Z = α·f1+β·f2+γ·f3, where f1 represents the influence on disease diagnosis, f2 represents the influence on treatment plan selection, and f3 represents the influence on prognosis evaluation. α, β and γ are the weights for disease diagnosis, treatment plan selection and prognosis evaluation, respectively, and α+β+γ=1. The objective function Z is calculated using the feature importance score and weight to distinguish between important and ordinary examination reports.

[0029] Furthermore, the specific steps of the cloud storage module are:

[0030] Architecture design: Determine the distributed storage architecture, including the layout of server nodes, data redundancy scheme, etc., and pre-set the classification standards for short-term storage area, long-term storage area and archive storage area based on factors such as urgency, importance and access frequency;

[0031] Data receiving: receiving medical examination report data after being processed and classified by the examination type identification module, urgency assessment module and importance rating module;

[0032] Initial storage: According to the classification rules, the newly received data is stored in the corresponding initial storage area, that is, important and urgent data is stored in the short-term storage area;

[0033] Monitoring access: Continuously monitor the access frequency of each report, and regularly evaluate whether the report meets the migration conditions based on time thresholds, access frequency thresholds, and factors such as importance changes;

[0034] Data migration: When a report meets the migration conditions, it is automatically migrated to the corresponding storage area. That is, reports with reduced access frequency are migrated from the short-term storage area to the long-term storage area, and the index and access path of the data after migration are updated to make the data accessible.

[0035] Furthermore, the differentiated storage strategies of the cloud storage module specifically include:

[0036] Short-term storage area: used to store urgent and important reports that are frequently accessed recently, using high-speed, high-capacity storage media for fast reading and response;

[0037] Long-term storage area: Use storage media with larger capacity to store various reports that are retained for a period of time, optimizing the balance between storage cost and access speed;

[0038] Archive storage area: used for cold storage in the cloud, storing reports that are rarely accessed but still need to be retained, saving storage costs, but with slower access speeds.

[0039] Compared with the prior art, this medical examination report cloud storage management system has the following beneficial effects:

[0040] 1. By comprehensively considering the patient's symptoms, the critical values ​​of the examination indicators and the multiple factors marked by the clinician, the system of the present invention can accurately evaluate the urgency and importance of the report, and perform differentiated storage accordingly, ensuring that urgent and important reports can be processed and obtained with priority, providing timely and accurate information support for clinical decision-making. At the same time, the reports are classified according to different types of blood tests, imaging tests, pathological examinations, and electrocardiograms, which reduces the workload of medical staff and enables them to devote more energy to the treatment and care of patients.

[0041] 2. The present invention uses an urgency assessment module and an importance rating module to classify reports into three categories: emergency, general, and non-emergency, and distinguish important and ordinary reports based on the patient's symptoms, critical values ​​of examination indicators, and factors marked by clinicians. This automated classification and storage mechanism ensures that emergency and important reports can be processed and obtained with priority, which helps to improve the quality and timeliness of medical services. In particular, in emergency situations, this mechanism can quickly identify and process emergency reports, providing strong support for timely treatment of patients, thereby improving the overall level of medical services.

[0042] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 The present invention is an operation flow chart of a cloud storage management system for medical examination reports.

[0045] Figure 2 The present invention is a flowchart of a cloud storage management system for medical examination reports. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.

[0047] Embodiment 1

[0048] This embodiment describes in detail the specific implementation process of a medical examination report cloud storage management system. The system achieves optimal allocation of medical resources through automatic evaluation of urgency and importance, so that urgent and important reports can be processed quickly, thereby improving the quality and response speed of medical services. The distributed cloud storage architecture and differentiated storage strategies not only ensure the security and reliability of data, but also greatly facilitate medical staff and patients to quickly obtain the required information through diversified data retrieval methods, thereby improving the overall data management efficiency and user experience.

[0049] In the specific implementation, the deployment of the medical examination report cloud storage management system starts from the hospital's internal network. The system design and control unit realizes communication with the medical examination equipment through integration with the hospital information system. The examination report acquisition module is responsible for collecting the original medical data of various medical examination reports, including structured data and unstructured data. These data are initially verified and cleaned through the built-in data verification mechanism to ensure the accuracy and completeness of the data. Duplicate data processing and outlier processing ensure the consistency and reliability of the data. Subsequently, the data is transmitted to the cloud storage module, which adopts a distributed storage architecture and pre-sets the classification standards of short-term storage area, long-term storage area and archive storage area according to the urgency, importance and access frequency of the data.

[0050] Then, the examination type identification module extracts and preprocesses the text data of the collected medical examination reports, and uses pattern recognition technology to classify the reports into different types such as blood tests, imaging examinations, pathological examinations, and electrocardiograms. The urgency assessment module comprehensively considers factors such as the patient's symptoms, the critical values ​​of the examination indicators, and the clinician's annotations, and divides the reports into three categories: urgent, general, and non-urgent. The importance rating module further analyzes the impact of the examination results on disease diagnosis, treatment plan formulation, and prognosis evaluation, and distinguishes between important and ordinary reports. The cloud storage module adopts differentiated storage strategies based on these classification and rating results, stores the data in the corresponding storage area, and optimizes the utilization of storage resources and data accessibility.

[0051] Secondly, during the operation of the system, the cloud storage module continuously monitors the access frequency and importance changes of each report. The data reception and initial storage steps ensure that the newly received data is stored in the corresponding initial storage area according to the classification rules. The access monitoring step evaluates the migration conditions of the report through the time threshold and access frequency threshold. When the access frequency of the report decreases or the importance changes, the data migration step automatically migrates the report to the corresponding storage area, such as from the short-term storage area to the long-term storage area, to optimize storage resources and maintain data accessibility. In addition, the system design and control unit support remote operation and monitoring, which is convenient for maintenance personnel to perform system maintenance and troubleshooting.

[0052] The urgency assessment module and the importance rating module not only consider the symptom severity score and the indicator criticality score, but also combine the doctor's annotation weights to calculate a comprehensive score, so as to more accurately divide the urgency of the report. The importance rating module analyzes the impact of the examination results on disease diagnosis, treatment plan selection and prognosis evaluation, defines the objective function of the comprehensive disease diagnosis, treatment plan selection and prognosis evaluation, and uses the feature importance score and weight to calculate the objective function, so as to distinguish between important and ordinary examination reports. This refined rating mechanism helps medical staff quickly identify and process key medical examination reports, and improve the quality and efficiency of medical services.

[0053] Finally, the needs and suggestions of medical staff during use are analyzed, and the system is continuously optimized. The automated control strategy is adjusted according to actual usage and user feedback to improve the system's ease of use and efficiency. The algorithms for urgency assessment and importance rating are continuously updated based on the latest medical knowledge and clinical practice to ensure that the system can adapt to the ever-changing medical environment. In addition, the storage strategy of the cloud storage module is dynamically adjusted according to changes in data access patterns and the development of storage technology to maintain the efficiency and reliability of the system in long-term operation.

[0054] Through the above implementation process, the system has greatly improved the efficiency and accuracy of report management through intelligent data classification and automated processing, reduced human errors, ensured the integrity and consistency of data, and achieved optimal allocation of medical resources through automatic assessment of urgency and importance, so that urgent and important reports can be processed quickly, thereby improving the quality and response speed of medical services.

[0055] Embodiment 2

[0056] This embodiment provides a cloud storage management system for medical examination reports to achieve effective classification, urgency assessment, importance rating and efficient storage of medical examination reports, and particularly emphasizes that the examination type identification module performs in-depth classification of the processed examination report texts. The urgency assessment module comprehensively considers the patient's symptoms, the critical values ​​of the examination indicators and the clinician's annotation information, and calculates the urgency of each report through a complex scoring model, so that the system can not only improve the efficiency and accuracy of data management, but also ensure that urgent and important reports can be processed in a timely manner, which helps to improve the quality and timeliness of medical services.

[0057] In the specific implementation, the medical examination report cloud storage management system is fully connected to various medical examination equipment and information systems within the hospital, including laboratory testing equipment, imaging examination equipment and the hospital's electronic medical record system. The examination report acquisition module can collect the original data of various medical examination reports in real time, whether it is the various indicator values ​​of blood tests, images and report texts of imaging examinations, or detailed descriptions of pathological examinations, they can all be obtained accurately. At the same time, the built-in data verification mechanism will strictly verify and clean the collected data. For example, for repeated indicator data in blood tests, the system will automatically retain the latest valid records and conduct secondary verification of abnormal values ​​and data sources to ensure the accuracy of the data.

[0058] The inspection type recognition module uses advanced natural language processing technology to conduct in-depth analysis of the report text. For example, for the input inspection report text, it converts it into a series of feature vectors {v1, v2, …, v T}, where T is the length of the inspection report text, defines the query vector q, key vector k and value vector v, and calculates the attention score e ij , that is, e ij =f(q i ,k j ), function f(q i ,k j ) is the dot product operation, that is Normalize the attention score to get the attention weight α ij ,Right now The feature vectors are weighted and summed according to the attention weights to obtain the final attention output y, that is, y = ∑ j α ij v j When a blood test report mentions keywords such as "red blood cell count" and "platelets", the system can accurately identify them as blood test types. For terms such as "CT", "MRI", and "X-ray" that appear in imaging examination reports, higher weights are assigned to help distinguish different types of imaging reports. For descriptions of imaging results, such as "lung nodules" and "fracture lines are clearly visible", the system can capture descriptive sentences for accurate classification. Through learning and pattern recognition of a large number of historical reports, the system can accurately classify reports into different types, providing a basis for subsequent urgency assessment and importance rating.

[0059] In terms of urgency assessment, we extract descriptions of patient symptoms from the report text, such as dyspnea and persistent high fever, extract key examination index values, such as white blood cell count and blood sugar value in blood tests, and extract emergency instructions marked by doctors in the report. We create a scoring model based on the frequency and severity of symptoms and the correlation between symptoms. We assign a severity level to each symptom based on medical knowledge, such as mild, moderate, and severe. We assign a severity level value to each severity level. For a single symptom, S i =Freq(s i )×Severity(s i ), for two symptoms i and j , Construct a symptom co-occurrence matrix to record the probability of different symptoms occurring at the same time, calculate the score of each symptom, and calculate the comprehensive symptom severity score based on the correlation between the symptoms. That is, the comprehensive symptom severity score S is: The comprehensive score is calculated by combining the symptom severity score, the indicator criticality score and the doctor's weighted score, that is, C = w S ·S+w I I+w D D. Develop a corresponding severity scoring standard for each type of examination result, including blood tests, imaging tests, pathological examinations, and electrocardiograms. i , defining its severity score C i ,Right now The severity score T of all examination results is expressed as According to the comprehensive score T, a threshold is set to classify the urgency categories, that is, for T>T 紧急, it is considered urgent. 非紧急 <T≤T 紧急 , then it is judged as general, for T≤T 非紧急 , it is judged as non-urgent.

[0060] After the urgency assessment is completed, the importance rating module comes into play. This module collects the annotated and verified historical medical examination report data, as well as the data processed by the examination type identification module and the urgency assessment module. By extracting features from these data, including the numerical value and description of the examination results, the correlation with the disease diagnosis, the weight of the impact on the treatment plan selection, and the predictive value of the prognosis assessment, the module can calculate the influence of each feature on the disease diagnosis, treatment plan selection and prognosis assessment, and define an objective function based on this. The feature importance score and weight are used to distinguish between important and ordinary examination reports, and their influence on disease diagnosis, treatment plan selection and prognosis assessment is calculated, that is, Define the objective function of comprehensive disease diagnosis, treatment plan selection and prognosis evaluation, that is, Z = α·f1+β·f2+γ·f3, and use the feature importance score and weight to calculate the objective function Z to distinguish important and ordinary examination reports.

[0061] Finally, the cloud storage module is responsible for differentiated storage according to factors such as the urgency, importance, and access frequency of reports. The system pre-sets classification standards for short-term storage areas, long-term storage areas, and archive storage areas. Newly received data will be stored in the corresponding area according to the classification rules. For example, important and urgent data will be stored in the short-term storage area. In addition, the system will continuously monitor the access frequency of each report and regularly evaluate whether the report meets the migration conditions based on the changes in time thresholds, access frequency thresholds, and importance. When the report meets the migration conditions, the system will automatically migrate it from the short-term storage area to the long-term storage area or archive storage area, and update the index and access path to ensure the accessibility of the data. This differentiated storage strategy can ensure that urgent and important reports are given priority, while optimizing the use of storage resources and improving overall storage efficiency and data security.

[0062] To sum up, through the system provided in this embodiment, medical institutions can significantly improve the management and use efficiency of medical reports, reduce the workload of medical staff, improve the quality of medical services and the safety of patient diagnosis and treatment. The design of the system fully considers the needs of actual application scenarios, and ensures that urgent and important reports can be given priority through an automated classification and storage mechanism, which helps to improve the overall level of medical services.

[0063] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A cloud storage management system for medical examination reports, characterized in that: The components of the management system include: inspection report acquisition module, inspection type identification module, urgency assessment module, importance rating module, and cloud storage module; The examination report acquisition module connects to the hospital's internal medical examination equipment and information system to collect raw medical data from various medical examination reports, including structured data and unstructured data. It also has a built-in data verification mechanism to perform preliminary verification and cleaning on the collected data. The examination type recognition module identifies and classifies the examination type from the report text, and classifies the report into different types of blood tests, imaging tests, pathological tests, and electrocardiogram tests by learning and pattern recognition from a large number of historical reports; The urgency assessment module comprehensively considers multiple factors, including the patient's symptoms, the critical values ​​of the examination indicators, and the clinician's annotations, and divides the report into three categories: urgent, general, and non-urgent. The importance rating module analyzes the impact of the examination results on disease diagnosis, treatment plan formulation, and prognosis assessment, thereby distinguishing between important and ordinary reports; The cloud storage module adopts a distributed storage architecture and adopts differentiated storage strategies for reports of different categories according to their importance and access frequency, including short-term storage area, long-term storage area, and archive storage area.

2. A medical examination report cloud storage management system according to claim 1, characterized in that: The examination report acquisition module communicates with the medical examination equipment within the hospital to obtain examination data, integrates with the hospital information system, obtains electronic medical records and report information, and comprehensively collects the original medical data of various medical examination reports, including structured data, i.e., the numerical values ​​and date and time information of various examination indicators, and unstructured data, i.e., the text descriptions, images, and videos of the pathology report. The collected data will immediately enter the built-in data verification mechanism for preliminary verification and cleaning. The verification mechanism includes: Duplicate data processing: For duplicate data, keep the latest valid records and delete and mark the old records; Outlier processing: For obviously abnormal values, they will be checked again with the data source, and judgment and correction will be made based on the distribution range of historical data.

3. A medical examination report cloud storage management system according to claim 1, characterized in that: The examination type identification module extracts text data from the original medical examination report received by the examination report acquisition module, preprocesses the text data, including removing irrelevant symbols, special characters, word segmentation, and part-of-speech tagging. For the input examination report text, it is converted into a series of feature vector representations, namely {v1, v2, ..., v T }, where T is the length of the inspection report text, and the query vector q, key vector k and value vector v are defined. The query vector q is regarded as an abstract representation of the currently concerned feature, the key vector k is used to measure the relevance of each input feature vector with the query vector, and the value vector v is the vector that actually carries the information. The attention score e is calculated. ij , that is, e ij =f(q i ,k j ), function f(q i ,k j ) is the dot product operation, that is Measure the reported data similarity between the query vector q and the key vector k, normalize the attention score, and get the attention weight α ij ,Right now The feature vectors are weighted and summed according to the attention weights to obtain the final attention output y, that is, y = ∑ j α ij v j ,According to the calculation results, we focus on the key parts in the input text in a targeted manner, so as to more accurately identify and classify the inspection type.

4. A medical examination report cloud storage management system according to claim 3, characterized in that: The examination report acquisition module identifies the medical examination report by focusing on the final results output, emphasizing the keywords and phrases related to specific examination types in the report, namely blood test, imaging examination, pathological examination, and electrocardiogram examination. The classification results include: Blood tests: Automatically identify and weight blood indicators mentioned in the report, i.e. red blood cell count, platelets; Imaging examinations: Assign higher weights to imaging terms that appear in reports, such as CT, MRI, and X-ray, to help distinguish different types of imaging reports. For descriptions of imaging findings, such as whether lung nodules or fracture lines are clearly visible, capture descriptive sentences. Pathology examination: Highlight the description of pathological changes in the report, such as the presence of cancer cells and obvious inflammatory reactions in tissue samples. Extract the characteristics of pathological changes in the report, such as cancer cell infiltration and inflammatory reactions. The diagnostic criteria contained in the pathology report, such as histological grade and tumor stage, will be given higher weight to ensure accurate classification. Electrocardiogram examination: Pay more attention to the key information in the electrocardiogram report, namely ST segment elevation, abnormal QRS wave morphology, and arrhythmia.

5. A medical examination report cloud storage management system according to claim 1, characterized in that: The specific steps of the urgency assessment module are: Data preparation: Receive raw data from the examination report acquisition module and evaluate the urgency of symptoms in the examination report based on the classification results of the examination type identification module; Feature extraction: Extract the patient's symptom description from the report text, such as difficulty breathing and persistent high fever; extract key examination indicator values, such as white blood cell count and blood sugar level from blood tests; and extract the emergency instructions marked by the doctor in the report; Urgency Assessment: Symptom severity score, which is based on the symptom characteristics, scores the severity of symptoms; indicator criticality score, which is based on the characteristics of the examination indicators, assesses whether the indicators are outside the normal range and their criticality; doctor's weight, that is, the emergency description marked by the doctor, is given additional weight; Comprehensive score: Calculates a comprehensive score for the urgency assessment and classifies reports as urgent, general, or non-urgent based on the comprehensive score.

6. A medical examination report cloud storage management system according to claim 5, characterized in that: In the specific steps of the urgency assessment module, the urgency assessment is combined with the frequency, severity and correlation factors of the symptoms to create a scoring model, and each symptom is assigned a severity level according to medical knowledge, i.e. mild, moderate, severe, and a severity level value is assigned to each severity level. For a single symptom S i , that is, S i =Freq(s i )×Severity(s i ), where Freq(s i ) is symptom S i Frequency of occurrence, Severity(s i ) is symptom S i The severity level value for both symptoms s i and s j , PMI P(s i ,s j ) is a symptom i and s j The probability of simultaneous occurrence, P(s i ) and P(s j ) are symptoms i and s j The probability of a single symptom appearing is used to construct a symptom co-occurrence matrix, record the probability of different symptoms appearing at the same time, calculate the score of each symptom, and calculate the comprehensive symptom severity score based on the correlation between the symptoms. That is, the comprehensive symptom severity score S is: n is the number of symptoms in the symptom list, S i Is a single symptom i Rating, PMI(s i ,s j ) is a symptom i and s j Point mutual information between, Severity(s i ) and Severity(s j ) are symptoms i and s j The severity level value.

7. A medical examination report cloud storage management system according to claim 6, characterized in that: The urgency assessment module combines the symptom severity score, the indicator criticality score and the doctor's weighted score to calculate a comprehensive score, namely C = w S ·S+w I I+w D D, C are comprehensive scores, w S 、w I and w D The weight coefficients are the symptom severity score, the indicator criticality score and the doctor's annotation weight, respectively. S, I and D are the symptom severity score, the indicator criticality score and the doctor's annotation score, respectively. The corresponding severity scoring standard is formulated for each type of examination result, including blood test, imaging test, pathological examination and electrocardiogram examination. For each type of examination result c i , we define its severity score C i ,Right now m is the number of abnormal indicators, a j is the abnormality index j, Freq(a j ) is the abnormality indicator a j The frequency of occurrence in this type of inspection, Severity (a j ) is the abnormality indicator a j The severity level value of the comprehensive inspection results is expressed as: n is the number of check types, C i is the severity score of the i-th inspection result. According to the comprehensive score T, a threshold is set to divide the urgency category, that is, for T>T 紧急 , it is considered urgent. For T 非紧急 <T≤T 紧急 , then it is judged as general, for T≤T 非紧急 , it is judged as non-urgent.

8. A medical examination report cloud storage management system according to claim 1, characterized in that: The importance rating module collects the annotated and verified historical medical examination report data and the data identified by the examination type identification module and the urgency assessment module. These data contain detailed information on examination results, disease diagnosis, treatment plan and prognosis assessment. Features are extracted from the training set, including the numerical value and description of the examination results, the correlation with the disease diagnosis, the influence weight on the treatment plan selection, and the predictive value for the prognosis assessment. For each feature x i , calculate its impact on disease diagnosis, treatment selection and prognosis assessment, that is, N is the number of features, w j is the weight of feature j, Impact ij is feature x i The degree of influence of feature j is used to define the objective function of the comprehensive influence on disease diagnosis, treatment plan selection and prognosis evaluation, that is, Z = α·f1+β·f2+γ·f3, where f1 represents the degree of influence on disease diagnosis, f2 represents the degree of influence on treatment plan selection, and f3 represents the degree of influence on prognosis evaluation. α, β and γ are the weights of disease diagnosis, treatment plan selection and prognosis evaluation, respectively, and α+β+γ=1. The objective function Z is calculated using the feature importance score and weight to distinguish between important and ordinary examination reports.

9. A medical examination report cloud storage management system according to claim 1, characterized in that: The specific steps of the cloud storage module are: Architecture design: Determine the distributed storage architecture, including the layout of server nodes and data redundancy schemes. Predetermine the classification standards for short-term storage, long-term storage, and archive storage areas based on urgency, importance, and access frequency. Data reception: receiving medical examination report data that has been processed and classified by the examination type identification module, urgency assessment module, and importance rating module; Initial storage: According to the classification rules, newly received data is stored in the corresponding initial storage area, that is, important and urgent data is stored in the short-term storage area; Access monitoring: Continuously monitor the access frequency of each report and regularly assess whether the report meets the migration criteria based on time thresholds, access frequency thresholds, and changes in importance. Data migration: When a report meets the migration conditions, it will be automatically migrated to the corresponding storage area. That is, reports with reduced access frequency will be migrated from the short-term storage area to the long-term storage area. The index and access path of the data will be updated after migration to make the data accessible.

10. A medical examination report cloud storage management system according to claim 9, characterized in that: The differentiated storage strategies of the cloud storage module specifically include: Short-term storage area: used to store urgent and important reports that are frequently accessed recently, using high-speed, high-capacity storage media for rapid reading and response; Long-term storage area: Uses storage media with larger capacity to store various reports that are retained for a period of time, optimizing the balance between storage cost and access speed; Archive storage area: used for cloud cold storage, storing reports that are rarely accessed but still need to be retained, saving storage costs but with slower access speeds.

Images