Artificial intelligence-based medical data management method and system

Abstract

The application relates to the technical field of artificial intelligence computer systems, and particularly discloses a medical data management method and system based on artificial intelligence, which comprises the following steps: collecting a medical data set and obtaining attribute parameters thereof; analyzing a data complexity index; simultaneously considering performance parameters of a management device to which the medical data set belongs; comprehensively evaluating device performance; using artificial intelligence technology to perform encryption processing on the medical data set according to performance evaluation factors, and marking the medical data set as an encrypted medical data set, thereby enhancing the security of the medical data, ensuring privacy protection of the data in the transmission and storage process, simultaneously collecting encryption parameters of the encrypted medical data set, evaluating storage complexity, intelligently judging whether the encrypted data set needs to be compressed and stored according to the storage complexity, the method effectively balances data security, privacy protection and storage efficiency, reduces possible risk abnormalities in the storage process, improves the intelligent level and the security level of medical data management, and realizes efficient and safe storage of medical data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence computer systems, in particular to a medical data management method and system based on artificial intelligence. BACKGROUND

[0002] With the continuous deepening of medical informatization, medical data is growing explosively. These data have characteristics such as large data volume, data complexity, and high data value. Intelligent and effective management and utilization of these data to improve the quality and efficiency of medical services have become an important issue facing the current medical industry.

[0003] For example, the invention patent with publication number CN117993500B discloses a medical teaching data management method and system based on artificial intelligence, which specifically includes the following steps: obtaining multi-dimensional medical data of a patient; performing heterogeneous data fusion on the multi-dimensional medical data of the patient to generate heterogeneous fusion medical feature data; performing variational coding on the heterogeneous fusion medical feature data to generate medical feature coding data; performing latent variable coding on the medical feature coding data to generate medical feature vector data; performing nonlinear coding mapping on the medical feature vector data to generate latent variable space coding data; performing fine-grained segmentation on the multi-dimensional medical data of the patient to generate medical fine-grained node data; and performing vector similarity calculation on the latent variable space coding data to generate a latent variable vector similarity index.

[0004] For example, the invention patent with publication number CN118428493B discloses an artificial intelligence data analysis method and system based on machine learning, which relates to the technical field of data analysis and includes collecting data in a data source and preprocessing the data, extracting features from the preprocessed data, constructing a multi-modal machine learning model for data prediction to predict the probability of a category, implementing data encryption, constructing an interactive data visualization interface, and storing collected and analyzed artificial intelligence data.

[0005] However, in the process of implementing the embodiments of the present application, the above-mentioned technology at least has the following technical problems: the existing technology mainly focuses on the processing and analysis process of medical data when managing medical data. The additional data volume in the analysis process may face threats to data integrity and availability, such as data corruption or slow data response, which hinders the continuous utilization and value mining of medical data and reduces the accuracy and efficiency of medical data storage. SUMMARY

[0006] To overcome the deficiencies of the prior art, the present application provides a medical data management method and system based on artificial intelligence, which can effectively solve the problems involved in the background art.

[0007] To achieve the above object, the present application is implemented by the following technical solutions: the present application provides a medical data management method based on artificial intelligence, comprising: step one, collecting a medical data set, obtaining attribute parameters of the medical data set, and analyzing a data complexity index of the medical data set, while obtaining performance parameters of a management device to which the medical data set belongs, and comprehensively analyzing a performance evaluation factor of the management device to which the medical data set belongs based on the data complexity index of the medical data set; step two, according to the performance evaluation factor of the management device to which the medical data set belongs, using artificial intelligence technology to encrypt the medical data set, and marking the processed medical data set as an encrypted medical data set, collecting encryption parameters of the encrypted medical data set, and evaluating the storage complexity of the encrypted medical data set; step three, according to the storage complexity of the encrypted medical data set, determining whether to use artificial intelligence technology to compress and store the encrypted medical data set.

[0008] As a further method, according to the performance evaluation factor of the management device to which the medical data set belongs, the artificial intelligence technology is used to encrypt the medical data set, and the specific process is: the performance evaluation factor of the management device to which the medical data set belongs is compared with the first performance evaluation factor interval, the second performance evaluation factor interval and the third performance evaluation factor interval stored in the medical database; if the performance evaluation factor of the management device to which the medical data set belongs belongs to the first performance evaluation factor interval, the first performance evaluation factor interval corresponding to the preset artificial intelligence first encryption strategy in the database is matched; if the performance evaluation factor of the management device to which the medical data set belongs belongs to the second performance evaluation factor interval, the second performance evaluation factor interval corresponding to the preset artificial intelligence second encryption strategy in the database is matched; if the performance evaluation factor of the management device to which the medical data set belongs belongs to the third performance evaluation factor interval, the third performance evaluation factor interval corresponding to the preset artificial intelligence third encryption strategy in the database is matched.

[0009] As a further method, the artificial intelligence technology is used to encrypt the medical data set, and the specific encryption process is: if the artificial intelligence first encryption strategy is matched according to the performance evaluation factor of the management device to which the medical data set belongs, the medical data set is encrypted according to the artificial intelligence first encryption strategy, and the encrypted medical data set is obtained after the processing is completed; if the artificial intelligence second encryption strategy is matched according to the performance evaluation factor of the management device to which the medical data set belongs, the medical data set is encrypted according to the artificial intelligence second encryption strategy, and the encrypted medical data set is obtained after the processing is completed; if the artificial intelligence third encryption strategy is matched according to the performance evaluation factor of the management device to which the medical data set belongs, the medical data set is encrypted according to the artificial intelligence third encryption strategy, and the encrypted medical data set is obtained after the processing is completed.

[0010] As a further method, the determination whether to compress store the encrypted medical data set by using the artificial intelligence technology is specifically: according to the performance evaluation factor of the management equipment to which the encrypted medical data set belongs, a storage complexity threshold is matched from the detection database; the storage complexity of the encrypted medical data set is compared with the storage complexity threshold, if the storage complexity of the encrypted medical data set is less than or equal to the storage complexity threshold, it is determined that the encrypted medical data set is not compressed stored by using the artificial intelligence technology, and the encrypted medical data set is directly stored; if the storage complexity of the encrypted medical data set is greater than the storage complexity threshold, it is determined that the encrypted medical data set is compressed stored by using the artificial intelligence technology.

[0011] As a further method, the compressed storage of the encrypted medical data set by using the artificial intelligence technology is specifically: the storage complexity of the encrypted medical data set is processed by using the storage complexity threshold, and the processing result is marked as the storage compression reference ratio of the encrypted medical data set; the data capacity of the encrypted medical data set is compressed by using the artificial intelligence technology according to the storage compression reference ratio of the encrypted medical data set, and the encrypted medical data set after compression is stored.

[0012] The second aspect of the present application provides an artificial intelligence-based medical data management system, comprising: a data set analysis module for collecting medical data sets, obtaining attribute parameters of the medical data sets, and analyzing data complexity indexes of the medical data sets, and simultaneously obtaining performance parameters of the management equipment to which the medical data sets belong, and comprehensively analyzing performance evaluation factors of the management equipment to which the medical data sets belong according to the data complexity indexes of the medical data sets; a data set encryption module for encrypting the medical data sets by using the artificial intelligence technology according to the performance evaluation factors of the management equipment to which the medical data sets belong, and marking the processed medical data sets as encrypted medical data sets, collecting encryption parameters of the encrypted medical data sets, and evaluating storage complexity of the encrypted medical data sets; and a data set compression module for determining whether to compress store the encrypted medical data set by using the artificial intelligence technology according to the storage complexity of the encrypted medical data set.

[0013] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0014] (1) This invention provides a medical data management method and system based on artificial intelligence. By collecting medical datasets and obtaining their attribute parameters, analyzing the data complexity index, and considering the performance parameters of the management equipment to which the medical datasets belong, the system comprehensively evaluates the equipment performance. Using artificial intelligence technology, the system encrypts the medical datasets according to the performance evaluation factors and marks them as encrypted medical datasets. This step not only enhances the security of medical data but also ensures the privacy protection of data during transmission and storage. Furthermore, the system collects the encryption parameters of the encrypted medical datasets, evaluates the storage complexity, and intelligently determines whether the encrypted datasets need to be compressed for storage. This method effectively balances data security, privacy protection, and storage efficiency, reduces the risks and anomalies that may occur during the storage process, improves the intelligence and security level of medical data management, and achieves efficient and secure storage of medical data.

[0015] (2) Based on the performance evaluation factors of the management equipment to which the medical dataset belongs, the present invention uses artificial intelligence technology to provide encryption protection for the medical dataset in a targeted manner. This approach effectively protects the privacy and security of medical data. At the same time, the encryption algorithm is flexibly adjusted according to the performance characteristics of the management equipment, and the medical data encryption processing flow is optimized. While ensuring data security, it also promotes faster and more efficient processing and application of data on high-performance management equipment.

[0016] (3) Based on the storage complexity of the encrypted medical dataset, this invention flexibly decides whether to use artificial intelligence technology to implement a compression storage strategy, thereby optimizing the use of storage space, reducing storage overhead, and ensuring the security and integrity of the data. In the context of the surge in medical data and the continuous growth of storage demand, this method not only improves the security and efficiency of storage, but also lays a solid foundation for convenient access and efficient management of data. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the method steps of the present invention.

[0019] Figure 2 This is a schematic diagram of the system module connections of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] Reference Figure 1 As shown, the first aspect of the present invention provides a medical data management method based on artificial intelligence, including: Step 1: collecting a medical dataset, obtaining the attribute parameters of the medical dataset, and analyzing the data complexity index of the medical dataset, while obtaining the performance parameters of the management device to which the medical dataset belongs, and analyzing the performance evaluation factor of the management device to which the medical dataset belongs by comprehensively analyzing the data complexity index of the medical dataset.

[0022] The aforementioned medical dataset refers to the medical data to be updated or saved received by the management device after receiving an update or save instruction, and the data as a whole is labeled as a medical dataset. The medical dataset includes, but is not limited to, patient physiological parameters, diagnostic images, and laboratory test data. The aforementioned management device includes, but is not limited to, computer equipment.

[0023] In one specific embodiment, the present invention uses artificial intelligence technology to provide targeted encryption protection for medical datasets based on the performance evaluation factors of the management device to which the medical dataset belongs. This approach effectively protects the privacy and security of medical data. At the same time, the encryption algorithm is flexibly adjusted according to the performance characteristics of the management device, optimizing the medical data encryption processing flow. While ensuring data security, it also promotes faster and more efficient processing and application of data on high-performance management devices.

[0024] Specifically, the analysis of performance evaluation factors for the management device to which the medical dataset belongs involves the following steps: The performance parameters of the management device include the real-time cache hit rate, real-time throughput, and real-time CPU wait queue length during the monitoring period. The monitoring period refers to the time interval for monitoring the performance of the management device, the specific duration of which is determined by the device administrator. It should also be noted that the time point for collecting the medical dataset is the end time of the monitoring period. The real-time cache hit rate represents the ratio between the number of times data was successfully retrieved from the management device's cache and the total number of data requests at each time point within the monitoring period; this can be obtained through performance monitoring tools (such as Grafana). The real-time throughput represents the total amount of data successfully transmitted and processed by the management device at each time point within the monitoring period; this can be obtained through application-level monitoring tools (such as application performance management). The real-time CPU wait queue length represents the number of tasks (or threads) waiting for the management device's CPU to process at each time point within the monitoring period; this can be obtained through performance monitoring tools (such as Grafana).

[0025] The real-time cache hit rate of the management device to which the medical dataset belongs during the monitoring period is processed to obtain the real-time cache miss rate of the management device to which the medical dataset belongs during the monitoring period. The specific data processing method is as follows: subtract the cache hit rate of the management device to which the medical dataset belongs at a certain time point during the monitoring period from 1 to obtain the cache miss rate at that time point. Through the above data processing method, the real-time cache miss rate of the management device to which the medical dataset belongs during the monitoring period can be obtained based on the real-time cache hit rate of the management device to which the medical dataset belongs during the monitoring period. The real-time cache miss rate represents the ratio of the number of times the management device's caching system failed to retrieve data from the cache to the total number of real-time data requests at each time point during the monitoring period.

[0026] The average throughput of the managed devices belonging to the medical dataset during the monitoring period is obtained by averaging the real-time throughput of the managed devices during the monitoring period. The maximum and minimum values ​​of the real-time throughput of the managed devices belonging to the medical dataset during the monitoring period are located and their differences are processed. The ratio of the difference processing result to the average throughput of the managed devices belonging to the medical dataset during the monitoring period is processed to obtain the throughput volatility of the managed devices belonging to the medical dataset during the monitoring period, and this value is marked as the real-time throughput volatility of the managed devices belonging to the medical dataset during the monitoring period. The aforementioned average throughput represents the average level of real-time throughput of the managed devices belonging to the medical dataset during the monitoring period. The throughput volatility of the managed device belonging to the medical dataset during the monitoring period represents the ratio of the maximum change in throughput of the managed device to the average throughput during the monitoring period. It reflects the stability and consistency of the managed device's data processing capability. The lower the throughput volatility, the more stable the processing capability of the managed device and the smaller the fluctuation. It should be explained that the throughput volatility of the managed device belonging to the medical dataset during the monitoring period is marked as the real-time throughput volatility of the managed device belonging to the medical dataset during the monitoring period. Therefore, the real-time throughput volatility of the managed device belonging to the medical dataset during the monitoring period is a constant value. The marking in this embodiment is to lay the foundation for a reasonable description of throughput volatility in the following performance evaluation factors.

[0027] By comprehensively analyzing the data complexity index of the medical dataset, the real-time cache miss rate of the management device of the medical dataset during the monitoring period, the real-time CPU wait queue length of the management device of the medical dataset during the monitoring period, and the real-time throughput volatility of the management device of the medical dataset during the monitoring period, a performance evaluation factor for the management device of the medical dataset is derived. The performance evaluation factor of the management device of the medical dataset represents a quantitative evaluation of the performance of the management device of the medical dataset. The larger the performance evaluation factor of the management device of the medical dataset, the better the performance of the management device of the medical dataset in data storage and processing.

[0028] The specific analysis method for the performance evaluation factors of the management equipment to which the medical dataset belongs is as follows:

[0029]

[0030] Specifically, PTD(ts) is:

[0031]

[0032] In the formula, PEF is the performance evaluation factor of the management device to which the medical dataset belongs, CWQ(YDC,YDM,YNC) is the data complexity index of the medical dataset, W is the weight factor corresponding to the preset data complexity index in the medical database, ts2 is the end time of the monitoring period, ts1 is the start time of the monitoring period, ts is any time point within the monitoring period, ts∈[ts1,ts2], PTD(ts) is the performance evaluation function of the management device to which the medical dataset belongs within the monitoring period, and RC(ts) is the performance evaluation function of the management device to which the medical dataset belongs within the monitoring period at time point ts. The cache miss rate is defined as follows: MD(ts) is the CPU wait queue length of the management device to which the medical dataset belongs at time point ts within the monitoring period; FR(ts) is the throughput volatility of the management device to which the medical dataset belongs at time point ts within the monitoring period; e is a natural constant; fs1 is the performance evaluation weighting factor corresponding to the preset unit value of cache miss rate in the medical database; fs2 is the performance evaluation weighting factor corresponding to the preset unit value of CPU wait queue length in the medical database; and fs3 is the performance evaluation weighting factor corresponding to the preset unit value of throughput volatility in the medical database.

[0033] The above PTD(ts) is the performance evaluation function of the managed equipment to which the medical dataset belongs during the monitoring period, representing the relationship between the performance of the managed equipment to which the medical dataset belongs and the time variable ts.

[0034] The weighting factors corresponding to the aforementioned data complexity indices represent the proportion of the data complexity index of the medical dataset to the performance evaluation factor of the management device to which the medical dataset belongs. The medical database stores the correspondence between the data complexity index and its corresponding weighting factor. For example, inputting the data complexity index of the medical dataset into the medical database will allow the database to match the corresponding weighting factor, with values ​​ranging from 0 to 1. The performance evaluation weighting factor corresponding to the aforementioned cache miss rate unit value represents the degree of influence of the cache miss rate unit value on the performance evaluation factor of the management device to which the medical dataset belongs. The performance evaluation weighting factor corresponding to the aforementioned CPU wait queue length unit value represents the degree of influence of the CPU wait queue length unit value on the performance evaluation factor of the management device to which the medical dataset belongs. The aforementioned throughput volatility unit value corresponds to... The performance evaluation weighting factor represents the degree of influence of the throughput volatility unit value on the performance evaluation factor of the management device to which the medical dataset belongs. The medical database stores the correspondence between cache miss rate, CPU wait queue length, and throughput volatility and their corresponding performance evaluation weighting factors. For example, if the real-time cache miss rate, real-time CPU wait queue length, and real-time throughput volatility of the management device to which the medical dataset belongs during the monitoring period are input into the medical database, the medical database can match the performance evaluation weighting factor corresponding to the cache miss rate unit value, the CPU wait queue length unit value, and the throughput volatility unit value, respectively. The values ​​are all between 0 and 1.

[0035] It's important to explain that an increase in cache miss rate directly reflects that many data accesses on the management device containing the medical dataset are not effectively utilizing the cache. This forces the management device to frequently access its underlying storage system. This frequent storage access not only increases the burden on the storage device but also significantly prolongs the CPU's waiting time for the required data. Because the CPU must wait for the underlying storage to return data before it can continue processing, the length of the CPU wait queue increases accordingly. This directly reflects increased latency in the management device when processing data requests. The increase in the CPU wait queue length not only means that the management device responds to requests more slowly but also reduces its overall data processing and storage efficiency. This decrease in efficiency further affects the throughput of the management device, leading to increased throughput volatility. Specifically, due to the increased latency in data access and the increased CPU waiting time, the stability of the management device in processing continuous data requests is affected, and its processing capacity becomes inconsistent, manifesting as unstable fluctuations in throughput. Therefore, an increase in cache miss rate, by exacerbating CPU waiting time, prolonging storage access time, and increasing the length of the CPU wait queue, ultimately leads to an increase in the throughput volatility of the management device. This series of chain reactions not only affects the efficiency of the management device in processing data but also has a significant adverse impact on its overall performance, especially the performance of stored data.

[0036] Meanwhile, this embodiment incorporates a data complexity index into the performance evaluation factor of the management device to which the medical dataset belongs, which is a comprehensive assessment factor. Medical datasets often contain a large amount of information of various types, such as images, videos, text, and physiological parameters, and these data often have complex relationships and hierarchical relationships. Therefore, the level of the data complexity index directly reflects the requirements of the medical dataset on the processing capacity and storage performance of the management device. Secondly, by incorporating the data complexity index into the performance evaluation factor of the management device, the overall performance of the management device in processing medical datasets can be evaluated more comprehensively. Traditional performance evaluation methods often only focus on single indicators such as the throughput and response time of the management device, while ignoring the impact of the complexity of the medical dataset itself on the performance of the management device. By introducing the data complexity index, the performance of the device in processing medical datasets of different complexities can be more accurately reflected, thereby providing more reliable device performance evaluation results.

[0037] Furthermore, the analysis of the data complexity index of the medical dataset is carried out through the following specific analysis process: The attribute parameters of the medical dataset include the data capacity, data dimension, missing field ratio, and redundant field ratio. The data capacity of the medical dataset represents the size of the data contained in the dataset. The data dimension represents the number of data attributes in the dataset. For example, if the data attributes in the medical dataset include age, gender, disease type, and ID number, then the data dimension is marked as 4. The missing fields represent the ratio of missing fields to the total number of fields in the dataset. The redundant field ratio represents the ratio of duplicate fields to the total number of fields. This can be obtained by deduplicating the medical dataset using data analysis software (such as a statistical analysis system), obtaining the number of deduplicated fields, and then comparing this number with the total number of fields before deduplication to obtain the redundant field ratio. The data capacity, data dimension, and missing field ratio of the medical dataset can all be extracted from the data processing logs of the management device.

[0038] The data noise factor of the medical dataset is obtained by multiplying the missing field ratio by the missing field ratio factor and then adding the redundant field ratio by the redundant field ratio factor. The missing field ratio factor, extracted from the medical database, ranges from 0 to 1, representing the degree of influence of the missing field ratio on the data noise factor. Similarly, the redundant field ratio factor, also extracted from the medical database, ranges from 0 to 1, representing the degree of influence of the redundant field ratio on the data noise factor. The data noise factor indicates the level of data impurities introduced by missing and redundant fields in the medical dataset. A high data noise factor not only reveals defects in the dataset itself but also indicates that the management equipment needs to invest more performance resources to cope with these complexities during storage, thus significantly negatively impacting the overall data processing efficiency and performance of the management equipment.

[0039] By comprehensively analyzing the data volume, data dimension, and data noise factor of the medical dataset, a data complexity index is derived. Specifically, the data complexity index is obtained by substituting the data volume, data dimension, and data noise factor of the medical dataset into the data complexity index function, resulting in the data complexity index CWQ(YDC,YDM,YNC). It should be noted that the above data complexity index represents the degree of complexity of the medical dataset.

[0040] In the formula, CWQ(YDC,YDM,YNC) is the data complexity index of the medical dataset, YDC is the data volume of the medical dataset, YDM is the data dimension of the medical dataset, and YNC is the data noise factor of the medical dataset; the data complexity index function is CWQ(DC,DM,NC), specifically expressed as follows:

[0041]

[0042] CWQ(DC,DM,NC) is the data complexity exponential function, where DC is the data capacity, DM is the data dimension, NC is the data noise factor, gt1 is the data complexity exponential weight factor corresponding to a preset unit value of data capacity in the medical database, gt2 is the data complexity exponential weight factor corresponding to a preset unit value of data dimension in the medical database, gt3 is the data complexity exponential weight factor corresponding to a preset unit value of data noise factor in the medical database, and e is a natural constant.

[0043] The aforementioned data complexity index function is a mathematical expression that describes how the data complexity index dynamically changes with variations in data volume, data dimensionality, and data noise factor.

[0044] The data complexity index weighting factor corresponding to the aforementioned data capacity unit value represents the degree of influence of the data complexity index on the data capacity unit value. In this embodiment, the medical database stores the correspondence between data capacity and its corresponding data complexity index weighting factor. For example, by inputting the data capacity of a medical dataset into the medical database, the medical database can match the data complexity index weighting factor corresponding to the data capacity unit value, with a value range between 0 and 1. Similarly, the data complexity index weighting factor corresponding to the aforementioned data dimension unit value represents the degree of influence of the data dimension unit value on the data complexity index. In this embodiment, the medical database stores the data dimension and its corresponding data complexity index weighting factor. The correspondence between data noise factors and their corresponding data complexity index weight factors is as follows: For example, if the data dimensions of a medical dataset are input into a medical database, the medical database can match the data complexity index weight factor corresponding to the unit value of the data dimension, with a value range between 0 and 1. The data complexity index weight factor corresponding to the unit value of the data noise factor represents the degree of influence of the unit value of the data noise factor on the data complexity index. In this embodiment, the medical database stores the correspondence between the data noise factor and its corresponding data complexity index weight factor. For example, if the data noise factor of a medical dataset is input into the medical database, the medical database can match the data complexity index weight factor corresponding to the unit value of the data noise factor, with a value range between 0 and 1.

[0045] It's important to explain that increasing data dimensionality means each data point will contain more feature information. During data processing and storage, since different dimensions may occupy different amounts of space and content, this naturally leads to an increase in data volume. Increased data volume not only means more physical storage space is needed but also that more data points need to be processed during data processing and analysis, thus increasing computational complexity and time costs. As data volume increases, the probability of anomalies in the dataset also increases accordingly. These anomalies may manifest as missing fields, redundant fields, etc., all of which are forms of data noise. An increase in data noise factors means that there are more uncertainties and interference factors in the dataset, which can mask the true data. According to the signal, patterns and relationships in the data become more difficult to identify and extract. The expansion of data dimensions not only increases data volume but also significantly enhances data complexity. In high-dimensional space, the relationships between data points become more intricate. This increased complexity not only increases the difficulty of data processing but may also lead to problems such as the curse of dimensionality, severely affecting the accuracy and reliability of data analysis. In summary, there is a close and complex interaction between data dimensions, data volume, and data noise factors. The increase in data dimensions leads to an increase in data volume, which in turn may trigger more data anomalies and data noise, jointly exacerbating data complexity and making data analysis by management equipment more complicated.

[0046] Step 2: Based on the performance evaluation factors of the management device to which the medical dataset belongs, use artificial intelligence technology to encrypt the medical dataset, mark the processed medical dataset as encrypted medical dataset, collect the encryption parameters of the encrypted medical dataset, and evaluate the storage complexity of the encrypted medical dataset.

[0047] Specifically, the storage complexity of the encrypted medical dataset was evaluated, and the evaluation process is as follows:

[0048] The encryption parameters of the encrypted medical dataset include the data size of the encrypted medical dataset, the encryption complexity level of the encryption algorithm to which the encrypted medical dataset belongs, and the encryption speed of the encrypted medical dataset. The data size of the encrypted medical dataset represents the total amount of data contained in the encrypted dataset, which can be extracted from the data processing logs of the management device. The encryption complexity level of the encryption algorithm to which the encrypted medical dataset belongs is used to quantify the computational resources and time complexity required by the encryption algorithm when processing the encrypted medical dataset. The medical database stores the encryption complexity level corresponding to each encryption algorithm. The specific quantification rules for the encryption complexity level are formulated by the data analysis administrator based on the time complexity and space complexity of the encryption algorithm. The encryption complexity level of the encryption algorithm to which the medical dataset belongs can be obtained directly by querying the medical database. The encryption speed of the encrypted medical dataset refers to the amount of data that the management device can encrypt per unit time during the encryption process of the encrypted medical dataset. The encryption time of the encrypted medical dataset is extracted from the data processing logs of the management device, and the data size of the encrypted medical dataset is compared with the encryption time of the encrypted medical dataset to obtain the encryption speed of the encrypted medical dataset.

[0049] The difference between the data capacity of the encrypted medical dataset and the data capacity of the medical dataset is processed, and the result is compared with the data capacity of the medical dataset to obtain the data expansion rate of the encrypted medical dataset, which represents the degree of expansion of the data capacity of the encrypted medical dataset compared with the data capacity before encryption.

[0050] Obtain the allowed storage capacity of the management device to which the encrypted medical dataset belongs, and perform a difference operation between this allowable storage capacity and the data capacity of the encrypted medical dataset to obtain the storage reserve capacity of the management device to which the encrypted medical dataset belongs. The allowed storage capacity of the management device to which the encrypted medical dataset belongs represents the remaining capacity of the management device before storing the encrypted medical dataset, and can be extracted from the storage management log of the management device. The storage reserve capacity of the management device to which the encrypted medical dataset belongs represents the capacity that the management device can still use to store other data after storing the encrypted medical dataset.

[0051] It should be explained that the encrypted medical dataset is the encrypted name of the medical dataset, and the management device to which the encrypted medical dataset belongs is the same management device as the management device to which the medical dataset belongs.

[0052] The performance evaluation factor of the management device to which the encrypted medical dataset belongs is obtained. The storage capacity of the management device to which the encrypted medical dataset belongs, the data expansion rate of the encrypted medical dataset, the encryption complexity level of the encryption algorithm to which the medical dataset belongs, and the encryption speed of the encrypted medical dataset are comprehensively evaluated to obtain the storage complexity of the encrypted medical dataset. The aforementioned performance evaluation factor of the management device to which the encrypted medical dataset belongs is the performance evaluation factor of the management device to which the medical dataset belongs. In this embodiment, in order to keep the performance evaluation factor and the storage complexity below in the same naming dimension, the performance evaluation factor of the management device to which the medical dataset belongs is marked as the performance evaluation factor of the management device to which the encrypted medical dataset belongs. The aforementioned storage complexity of the encrypted medical dataset represents the degree of complexity in the storage process of the encrypted medical dataset.

[0053] Specifically, the storage complexity of the encrypted medical dataset is evaluated using the following method:

[0054]

[0055] In the formula, SMS represents the storage complexity of the encrypted medical dataset, PEFD represents the performance evaluation factor of the management device to which the encrypted medical dataset belongs, LR represents the storage reserve capacity of the management device to which the encrypted medical dataset belongs, LP represents the data expansion rate of the encrypted medical dataset, OT represents the encryption complexity level of the encryption algorithm to which the medical dataset belongs, VM represents the encryption speed of the encrypted medical dataset, YU represents the storage complexity weight factor corresponding to the preset performance evaluation factor in the medical database, ow1 represents the storage complexity weight factor corresponding to the preset unit value of the storage reserve capacity in the medical database, ow2 represents the storage complexity weight factor corresponding to the preset unit value of the data expansion rate in the medical database, ow3 represents the storage complexity weight factor corresponding to the preset unit value of the encryption complexity level in the medical database, ow4 represents the storage complexity weight factor corresponding to the preset unit value of the encryption speed in the medical database, and e represents a natural constant.

[0056] The storage complexity weight factors corresponding to the aforementioned performance evaluation factors represent the proportion of the performance evaluation factors in the storage complexity of the encrypted medical dataset. This establishes the correspondence between the storage performance evaluation factors and their corresponding storage complexity weight factors in the medical database. For example, inputting the performance evaluation factor of the management device to which the encrypted medical dataset belongs into the medical database will allow the database to match the corresponding storage complexity weight factor, with values ​​ranging from 0 to 1. Similarly, the storage complexity weight factor corresponding to the aforementioned unit value of storage reserve capacity represents the degree of influence of the unit value of storage reserve capacity on the storage complexity of the encrypted medical dataset. This establishes the correspondence between the storage reserve capacity and its corresponding storage complexity weight factors in the medical database. For example, inputting the storage reserve capacity of the management device to which the encrypted medical dataset belongs into the medical database will allow the database to match the corresponding storage complexity weight factor, with values ​​ranging from 0 to 1. Finally, the storage complexity weight factor corresponding to the aforementioned unit value of data inflation rate represents the degree of influence of the unit value of data inflation rate on the storage complexity of the encrypted medical dataset. This establishes the correspondence between the storage inflation rate and its corresponding storage complexity weight factor in the medical database. The correspondence between complexity weight factors is as follows: For example, inputting the data inflation rate of an encrypted medical dataset into a medical database will allow the database to match the storage complexity weight factor corresponding to each unit value of the data inflation rate, with values ​​ranging from 0 to 1. Similarly, the storage complexity weight factor corresponding to each unit value of encryption complexity level represents the degree of influence of that unit value on the storage complexity of the encrypted medical dataset. The medical database stores the correspondence between encryption complexity levels and their corresponding storage complexity weight factors. For instance, inputting the encryption complexity level of the encryption algorithm to which the medical dataset belongs will allow the database to match the storage complexity weight factor corresponding to each unit value of the encryption complexity level, with values ​​ranging from 0 to 1. Likewise, the storage complexity weight factor corresponding to each unit value of encryption speed represents the degree of influence of that unit value on the storage complexity of the encrypted medical dataset. The medical database stores the correspondence between encryption speed and its corresponding storage complexity weight factor. For instance, inputting the encryption speed of an encrypted medical dataset will allow the database to match the storage complexity weight factor corresponding to each unit value of the encryption speed, with values ​​ranging from 0 to 1.

[0057] It's important to explain that the performance of the management device not only affects its ability to handle encryption and decryption operations but also the encryption speed. High-performance devices can execute encryption algorithms faster, thus shortening the encryption process time, which is crucial for improving overall storage efficiency. Storage capacity refers to the amount of space the management device can hold for encrypted medical datasets. After encryption, the medical dataset may occupy more storage space due to the characteristics of the encryption algorithm (such as data bloat rate), increasing the data bloat rate of the encrypted dataset. A high data bloat rate will exacerbate the consumption of storage space on the management device. Therefore, when selecting an encryption algorithm, a balance needs to be struck between its security and data bloat rate. Furthermore, the encryption complexity level of the encryption algorithm used for the encrypted medical dataset not only affects encryption speed but also indirectly affects storage complexity. While high-complexity encryption algorithms offer higher security, they typically require more computing resources and time to complete the encryption process. This can lead to latency and performance degradation during the storage of the encrypted dataset. Additionally, complex encryption algorithms may increase the data bloat rate of the encrypted medical dataset, further increasing the storage pressure under the limitations of the management device's performance, resulting in higher storage complexity and a greater susceptibility to packet loss.

[0058] Specifically, the process of encrypting the medical dataset using artificial intelligence technology based on the performance evaluation factors of the management device to which the medical dataset belongs involves the following steps: comparing the performance evaluation factors of the management device to which the medical dataset belongs with the first, second, and third performance evaluation factor intervals stored in the medical database; if the performance evaluation factors of the management device to which the medical dataset belongs belong to the first performance evaluation factor interval, then matching the first artificial intelligence encryption strategy corresponding to the preset first performance evaluation factor interval in the database; if the performance evaluation factors of the management device to which the medical dataset belongs belong to the second performance evaluation factor interval, then matching the second artificial intelligence encryption strategy corresponding to the preset second performance evaluation factor interval in the database; if the performance evaluation factors of the management device to which the medical dataset belongs belong to the third performance evaluation factor interval, then matching the third artificial intelligence encryption strategy corresponding to the preset third performance evaluation factor interval in the database.

[0059] Furthermore, the encryption process using artificial intelligence technology for the medical dataset is as follows: if a first encryption strategy is matched based on the performance evaluation factor of the management device to which the medical dataset belongs, the medical dataset is encrypted according to the first encryption strategy, resulting in an encrypted medical dataset; if a second encryption strategy is matched based on the performance evaluation factor of the management device to which the medical dataset belongs, the medical dataset is encrypted according to the second encryption strategy, resulting in an encrypted medical dataset; if a third encryption strategy is matched based on the performance evaluation factor of the management device to which the medical dataset belongs, the medical dataset is encrypted according to the third encryption strategy, resulting in an encrypted medical dataset.

[0060] In one example embodiment, assuming the performance evaluation factor of the management device to which the medical dataset belongs is G, the first performance evaluation factor interval is [G-220%, G-120%], the second performance evaluation factor interval is (G-120%, G-20%), and the third performance evaluation factor interval is (G-20%, G+80%), and the performance evaluation factor of the management device to which the medical dataset belongs falls within the third performance evaluation factor interval, then the AI ​​third encryption strategy is matched to encrypt the medical dataset. In this example embodiment, the AI ​​third encryption strategy includes the Advanced Encryption Standard - 256-bit encryption algorithm. It should be noted that in this example embodiment, the management device corresponding to the first performance evaluation factor interval is a low-performance device, and the AI ​​first encryption strategy has a simplified... The first performance evaluation factor range corresponds to management devices with limited performance. The second AI-based encryption strategy is balanced, ensuring basic data security by employing relatively simple encryption algorithms and techniques. The third performance evaluation factor range corresponds to management devices with medium performance. The AI-based encryption strategy is efficient, fully utilizing the computing power of the superior management devices for rapid encryption and decryption. It also provides a high level of security by employing advanced encryption algorithms and techniques to provide the highest level of protection for medical datasets, preventing data leakage and unauthorized access.

[0061] Step 3: Based on the storage complexity of the encrypted medical dataset, determine whether to use artificial intelligence technology to compress and store the encrypted medical dataset.

[0062] In one specific embodiment, the present invention flexibly decides whether to implement a compression storage strategy using artificial intelligence technology based on the storage complexity of the encrypted medical dataset. This optimizes the use of storage space, reduces storage overhead, and ensures the security and integrity of the data. In the context of the surge in medical data and the continuous growth in storage demand, this method not only improves the security and efficiency of storage, but also lays a solid foundation for convenient access and efficient management of data.

[0063] Specifically, the method for determining whether to use artificial intelligence technology to compress and store encrypted medical datasets is as follows: based on the performance evaluation factors of the management device to which the encrypted medical dataset belongs, a storage complexity threshold is matched from the detection database; the aforementioned storage complexity threshold represents the maximum value of the reasonable range of storage complexity for the encrypted medical dataset; the specific matching process for the storage complexity threshold is as follows: the medical database stores the storage complexity thresholds corresponding to each performance evaluation factor interval; the performance evaluation factor interval to which the storage complexity of the encrypted medical dataset belongs is queried; the storage complexity threshold corresponding to this performance evaluation factor interval is the storage complexity threshold matched by the performance evaluation factors of the management device to which the encrypted medical dataset belongs.

[0064] The storage complexity of the encrypted medical dataset is compared with a storage complexity threshold. If the storage complexity is less than or equal to the threshold, it is determined that the encrypted medical dataset will not be compressed using artificial intelligence technology and will be stored directly. If the storage complexity is greater than the threshold, it is determined that the encrypted medical dataset will be compressed using artificial intelligence technology. If the storage complexity is less than or equal to the threshold, it indicates that the storage requirements of the encrypted medical dataset are within the performance range of the management device, and the management device has sufficient resources and capabilities to directly store the encrypted medical dataset without additional compression processing. If the storage complexity is greater than the threshold, it indicates that the storage requirements of the encrypted medical dataset have exceeded the direct storage capacity of the management device. Direct storage may lead to problems such as low storage efficiency, resource shortages, and data loss. In order to make full use of the storage resources of the management device and ensure the integrity and security of the encrypted medical dataset, it is necessary to use artificial intelligence technology for compression storage. This decision-making process considers both storage efficiency and data security, as well as the performance limitations of the management device.

[0065] Furthermore, the use of artificial intelligence technology to compress and store the encrypted medical dataset involves the following compression process: The storage complexity of the encrypted medical dataset is compared to a storage complexity threshold; the result is marked as the storage compression reference ratio for the encrypted medical dataset. The artificial intelligence technology then compresses the data capacity of the encrypted medical dataset based on this storage compression reference ratio and stores the compressed dataset. The aforementioned storage compression reference ratio represents the ratio between the storage complexity of the encrypted medical dataset and the storage complexity threshold. In an example embodiment, assuming the storage compression reference ratio for the encrypted medical dataset is 80%, the artificial intelligence... The intelligent technology compresses the data capacity of the encrypted medical dataset according to the storage compression reference ratio, that is, compresses the data capacity of the encrypted medical dataset to 80% of the current data capacity. The artificial intelligence compression process is as follows: the JPEG2000 image compression algorithm is used to compress the images in the encrypted medical dataset, setting the compression ratio to 80%; the Gzip data compression algorithm is used to compress the text in the encrypted medical dataset, setting a high compression level to achieve a compression ratio of 80%; and the H.265 compression algorithm is used to compress the video in the encrypted medical dataset, setting appropriate encoding parameters to achieve a compression ratio of 80% while maintaining the smoothness and clarity of the video.

[0066] In one specific embodiment, the present invention provides an artificial intelligence-based medical data management method. This method involves collecting medical datasets and obtaining their attribute parameters, analyzing the data complexity index, and simultaneously considering the performance parameters of the management equipment to which the medical datasets belong. The method comprehensively evaluates the equipment performance and, using artificial intelligence technology, encrypts the medical datasets based on performance evaluation factors, labeling them as encrypted medical datasets. This step not only enhances the security of medical data but also ensures privacy protection during data transmission and storage. Furthermore, the encryption parameters of the encrypted medical datasets are collected, and the storage complexity is assessed. Based on this, an intelligent judgment is made as to whether the encrypted datasets need to be compressed for storage. This method effectively balances data security, privacy protection, and storage efficiency, reduces potential risks and anomalies during storage, improves the intelligence and security level of medical data management, and achieves efficient and secure storage of medical data.

[0067] Reference Figure 2 As shown, the second aspect of the present invention provides an artificial intelligence-based medical data management system, including: a dataset analysis module, a dataset encryption module, and a dataset compression module.

[0068] The second aspect of this invention provides an artificial intelligence-based medical data management system, which further includes a medical database. The medical database stores the following factors: a data complexity index weighting factor corresponding to a unit value of data capacity; a data complexity index weighting factor corresponding to a unit value of data dimension; a data complexity index weighting factor corresponding to a unit value of data noise; a weighting factor corresponding to the data complexity index; a performance evaluation weighting factor corresponding to a unit value of cache miss rate; a performance evaluation weighting factor corresponding to a unit value of CPU wait queue length; a performance evaluation weighting factor corresponding to a unit value of throughput volatility; a first performance evaluation factor interval; a second performance evaluation factor interval; a third performance evaluation factor interval; a first artificial intelligence encryption strategy; a second artificial intelligence encryption strategy; a third artificial intelligence encryption strategy; a storage complexity weighting factor corresponding to the performance evaluation factors; a storage complexity weighting factor corresponding to a unit value of storage reserve capacity; a storage complexity weighting factor corresponding to a unit value of data expansion rate; a storage complexity weighting factor corresponding to a unit value of encryption complexity level; a storage complexity weighting factor corresponding to a unit value of encryption speed; an encryption complexity level; a storage complexity threshold; a redundant field ratio factor; and a missing field ratio factor.

[0069] The dataset analysis module is connected to the dataset encryption module, the dataset encryption module is connected to the dataset compression module, and the dataset analysis module, dataset encryption module, and dataset compression module are all connected to the information feedback center.

[0070] The dataset analysis module is used to collect medical datasets, obtain attribute parameters of the medical datasets, analyze the data complexity index of the medical datasets, and obtain the performance parameters of the management equipment to which the medical datasets belong. Based on the data complexity index of the medical datasets, the module analyzes the performance evaluation factor of the management equipment to which the medical datasets belong.

[0071] The dataset encryption module is used to encrypt the medical dataset using artificial intelligence technology based on the performance evaluation factors of the management device to which the medical dataset belongs, and to mark the processed medical dataset as an encrypted medical dataset, collect the encryption parameters of the encrypted medical dataset, and evaluate the storage complexity of the encrypted medical dataset.

[0072] The dataset compression module is used to determine whether to use artificial intelligence technology to compress and store the encrypted medical dataset based on the storage complexity of the encrypted medical dataset.

[0073] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A medical data management method based on artificial intelligence, characterized in that, include: Step 1: Collect medical datasets, obtain the attribute parameters of the medical datasets, and analyze the data complexity index of the medical datasets. At the same time, obtain the performance parameters of the management equipment to which the medical datasets belong, and analyze the performance evaluation factors of the management equipment to which the medical datasets belong by combining the data complexity index of the medical datasets. Step 2: Based on the performance evaluation factors of the management device to which the medical dataset belongs, use artificial intelligence technology to encrypt the medical dataset, mark the processed medical dataset as encrypted medical dataset, collect the encryption parameters of the encrypted medical dataset, and evaluate the storage complexity of the encrypted medical dataset. The storage complexity of the encrypted medical dataset was evaluated, and the specific evaluation process is as follows: The encryption parameters of the encrypted medical dataset include the data capacity of the encrypted medical dataset, the encryption complexity level of the encryption algorithm to which the encrypted medical dataset belongs, and the encryption speed of the encrypted medical dataset. The difference between the data capacity of the encrypted medical dataset and the data capacity of the medical dataset is processed, and the processing result is compared with the data capacity of the medical dataset to finally obtain the data expansion rate of the encrypted medical dataset. Obtain the allowed storage capacity of the management device to which the encrypted medical dataset belongs, and perform a difference operation between the data capacity of the encrypted medical dataset to obtain the storage reserve capacity of the management device to which the encrypted medical dataset belongs. Obtain the performance evaluation factor of the management device to which the encrypted medical dataset belongs, and comprehensively evaluate the performance evaluation factor of the management device to which the encrypted medical dataset belongs, the storage capacity of the management device to which the encrypted medical dataset belongs, the data expansion rate of the encrypted medical dataset, the encryption complexity level of the encryption algorithm to which the medical dataset belongs, and the encryption speed of the encrypted medical dataset to obtain the storage complexity of the encrypted medical dataset. Step 3: Based on the storage complexity of the encrypted medical dataset, determine whether to use artificial intelligence technology to compress and store the encrypted medical dataset.

2. The artificial intelligence-based medical data management method according to claim 1, characterized in that: The analysis of the data complexity index of the medical dataset is as follows: The attribute parameters of the medical dataset include the data capacity of the medical dataset, the data dimension of the medical dataset, the missing field ratio of the medical dataset, and the redundant field ratio of the medical dataset. The data noise factor of the medical dataset is obtained by multiplying the missing field ratio by the missing field ratio factor and adding the redundant field ratio by the redundant field ratio factor. By comprehensively analyzing the data volume, data dimension, and data noise factor of the medical dataset, a data complexity index is derived. Specifically, the data complexity index is obtained by substituting the data volume, data dimension, and data noise factor of the medical dataset into a data complexity index function.

3. The medical data management method based on artificial intelligence according to claim 1, characterized in that: The analysis identified performance evaluation factors for the management devices associated with the medical dataset. The specific analysis process is as follows: The performance parameters of the management device to which the medical dataset belongs include the real-time cache hit rate, the real-time throughput, and the real-time CPU wait queue length during the monitoring period. The real-time cache hit rate of the management device to which the medical dataset belongs during the monitoring period is processed to obtain the real-time cache miss rate of the management device to which the medical dataset belongs during the monitoring period. The average throughput of the management device to which the medical dataset belongs during the monitoring period is averaged to obtain the average throughput of the management device to which the medical dataset belongs during the monitoring period. The maximum and minimum values ​​are located from the real-time throughput of the management device to which the medical dataset belongs during the monitoring period, and the difference is processed. The difference processing result is compared with the average throughput of the management device to which the medical dataset belongs during the monitoring period to obtain the throughput volatility of the management device to which the medical dataset belongs during the monitoring period. The throughput volatility of the management device to which the medical dataset belongs during the monitoring period is marked as the real-time throughput volatility of the management device to which the medical dataset belongs during the monitoring period. By comprehensively analyzing the data complexity index of the medical dataset, the real-time cache failure rate of the management device to which the medical dataset belongs during the monitoring period, the real-time CPU wait queue length of the management device to which the medical dataset belongs during the monitoring period, and the real-time throughput volatility of the management device to which the medical dataset belongs during the monitoring period, the performance evaluation factor of the management device to which the medical dataset belongs is derived.

4. The medical data management method based on artificial intelligence according to claim 1, characterized in that: The process of encrypting the medical dataset using artificial intelligence technology, based on the performance evaluation factors of the management device to which the medical dataset belongs, is as follows: The performance evaluation factors of the management device to which the medical dataset belongs are compared with the first performance evaluation factor interval, the second performance evaluation factor interval, and the third performance evaluation factor interval stored in the medical database; If the performance evaluation factor of the management device to which the medical dataset belongs belongs to the first performance evaluation factor interval, then the first artificial intelligence encryption strategy corresponding to the preset first performance evaluation factor interval in the database is matched. If the performance evaluation factor of the management device to which the medical dataset belongs belongs to the second performance evaluation factor interval, then the artificial intelligence second encryption strategy corresponding to the preset second performance evaluation factor interval in the database is matched. If the performance evaluation factor of the management device to which the medical dataset belongs falls within the third performance evaluation factor range, then the artificial intelligence third encryption strategy corresponding to the preset third performance evaluation factor range in the database will be matched.

5. The artificial intelligence-based medical data management method according to claim 4, characterized in that: The encryption process for medical datasets using artificial intelligence technology is as follows: If the first encryption strategy of artificial intelligence is matched according to the performance evaluation factor of the management device to which the medical dataset belongs, the medical dataset is encrypted according to the first encryption strategy of artificial intelligence, and the encrypted medical dataset is obtained after the processing is completed. If a second AI encryption strategy is matched based on the performance evaluation factor of the management device to which the medical dataset belongs, then the medical dataset is encrypted according to the second AI encryption strategy, and the encrypted medical dataset is obtained after the processing is completed. If an artificial intelligence third encryption strategy is matched based on the performance evaluation factors of the management device to which the medical dataset belongs, then the medical dataset is encrypted according to the artificial intelligence third encryption strategy, and the encrypted medical dataset is obtained after the processing is completed.

6. The medical data management method based on artificial intelligence according to claim 1, characterized in that: The storage complexity of the encrypted medical dataset is evaluated using the following method: ； In the formula, To address the storage complexity of encrypted medical datasets, For the performance evaluation factors of the management device to which the encrypted medical dataset belongs, Storage capacity of the management device containing the encrypted medical dataset. The data inflation rate of encrypted medical datasets, The encryption complexity level of the encryption algorithm to which the medical dataset belongs. To improve the encryption speed of encrypted medical datasets, Storage complexity weighting factors corresponding to pre-defined performance evaluation factors in medical databases. The storage complexity weight factor corresponding to the preset storage capacity unit value in the medical database. The storage complexity weight factor corresponding to the preset data expansion rate unit value in the medical database. The storage complexity weight factor corresponding to the preset encryption complexity level unit value in the medical database. The storage complexity weight factor corresponding to the preset encryption speed unit value in the medical database, where e is a natural constant.

7. The artificial intelligence-based medical data management method according to claim 1, characterized in that: The specific method for determining whether artificial intelligence technology is used to compress and store encrypted medical datasets is as follows: Based on the performance evaluation factors of the management devices to which the encrypted medical dataset belongs, the storage complexity threshold is matched from the detection database. The storage complexity of the encrypted medical dataset is compared with a storage complexity threshold. If the storage complexity of the encrypted medical dataset is less than or equal to the storage complexity threshold, it is determined that artificial intelligence technology will not be used to compress and store the encrypted medical dataset, and the encrypted medical dataset will be stored directly. If the storage complexity of the encrypted medical dataset exceeds the storage complexity threshold, it is determined that artificial intelligence technology should be used to compress and store the encrypted medical dataset.

8. The medical data management method based on artificial intelligence according to claim 7, characterized in that: The specific compression process for compressing and storing encrypted medical datasets using artificial intelligence technology is as follows: The storage complexity of the encrypted medical dataset is compared with a storage complexity threshold. The result is marked as the storage compression reference ratio of the encrypted medical dataset. Artificial intelligence technology compresses the data capacity of the encrypted medical dataset according to the storage compression reference ratio and stores the compressed encrypted medical dataset.

9. A system applying the artificial intelligence-based medical data management method as described in any one of claims 1-8, characterized in that: include: The dataset analysis module is used to collect medical datasets, obtain the attribute parameters of the medical datasets, analyze the data complexity index of the medical datasets, obtain the performance parameters of the managed equipment to which the medical datasets belong, and analyze the performance evaluation factors of the managed equipment to which the medical datasets belong by combining the data complexity index of the medical datasets. The dataset encryption module is used to encrypt medical datasets using artificial intelligence technology based on the performance evaluation factors of the management device to which the medical dataset belongs, and to mark the processed medical datasets as encrypted medical datasets. It also collects the encryption parameters of the encrypted medical datasets and evaluates the storage complexity of the encrypted medical datasets. The dataset compression module is used to determine whether to use artificial intelligence technology to compress and store encrypted medical datasets based on their storage complexity.

Images