Medical data transmission method and device

[0056] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the following further describes the present invention in detail in conjunction with specific embodiments and with reference to the accompanying drawings.

[0057] It should be noted that all the expressions "first" and "second" in the embodiments of the present invention are used to distinguish two entities with the same name but not the same or parameters that are not the same, as shown in "first" and "second" Only for the convenience of presentation, it should not be construed as a limitation to the embodiments of the present invention, and subsequent embodiments will not describe this one by one.

[0058] The embodiment of the present invention provides a medical data transmission method. It should be noted that in the context of big data mobile medical care, how to solve the data transmission between the hospital and the patient, and the patient and the society is a very important problem. Patients and social relations usually use mobile APP devices to receive and send data. Such devices have small storage space and limited data transmission and reception. Due to the large amount of data composed of the electronic medical records and test results of each patient, if all the data is sent and received every time, it will cause serious consumption of mobile APP equipment. For this reason, in a specific application scenario of this application, a routing algorithm based on Wireless Sensor Effective Data Transmission (WSEDT) is established. The algorithm judges and analyzes the routing request sent by its neighbors by the node, predicts the number of data packets required by the node, calculates the sending and receiving process of the data packet, improves the transmission success rate of the node data packet, and reduces the result. Point routing cost.

[0059] Such as figure 1 Shown is a schematic flow chart of a medical data transmission method proposed by an embodiment of the present invention. The method specifically includes the following steps:

[0060] Step 101: Obtain medical diagnosis data input by the user;

[0061] The purpose of this step is to obtain the data generated by the client during the medical diagnosis process entered by the doctor. Among them, there are many ways to obtain data, such as direct import after the doctor completes a complete diagnosis report, and the doctor gradually enters and completes the medical report during the diagnosis. As long as the different acquisition methods can achieve the corresponding purpose, the different methods will not affect the protection scope of the present invention.

[0062] In specific application scenarios, when doctors diagnose patients, they can reduce the misdiagnosis rate by combining machine diagnosis and manual diagnosis. In the big data self-organizing sensor network, doctors and hospitals send all diagnostic reports and results to patients through mobile devices, and patients and doctors choose further treatment options based on the results. However, patients may generate a lot of data and information during the diagnosis and treatment process. This information includes the patient's personal information, past medical history, examination items, medical images, hospitalization records and other information.

[0063] Step 102: Analyze the medical diagnosis data, obtain personal information and diagnosis results therein, classify the medical diagnosis data according to the diagnosis results, and obtain key data items in the medical diagnosis data according to the classification results, Associate the key data item with the personal information;

[0064] This step aims to classify the diagnosis results of medical diagnosis data, and extract the most relevant key data items to associate with personal information. There are many ways to classify data, such as: according to different patients, according to different times, according to different diseases, according to different lesions, etc.; there are also many ways to obtain key data items at the same time, such as: obtaining records according to patient history , According to the acquisition record of the same disease or focus, the data items acquired according to the needs of the doctor preset, etc. As long as the different classification methods and acquisition methods can achieve the corresponding purpose, the different methods will not affect the protection scope of the present invention.

[0065] Further, in order to find the key data items more accurately and scientifically. In a preferred embodiment of the present application, the classification of the medical diagnosis data according to the diagnosis result, and obtaining the key data items in the medical diagnosis data according to the classification result specifically includes:

[0066] Classify the diagnosis results according to different diagnosis parts;

[0067] Establishing a decision tree according to the diagnosis part and the diagnosis result of each of the diagnosis parts;

[0068] Obtaining the key data item corresponding to the diagnosis result according to the decision tree.

[0069] Further, in order to ensure the special data viewing needs of the client or doctor, and at the same time, quickly determine all required data when the client or doctor has no need. In a preferred embodiment of the present application, the obtaining the key data item corresponding to the diagnosis result according to the decision tree specifically includes:

[0070] Setting the corresponding data item in the decision tree as the key data item according to the user's selection;

[0071] and / or

[0072] Acquire historical acquisition records of the same diagnosis result, select the key data items whose acquisition rate exceeds a threshold value as comparison information according to the historical acquisition records, and compare the priority in the decision tree according to the comparison information Data, filtering out relevant data in the decision tree according to the priority data, and setting the relevant data as the key data item.

[0073] In a specific application scenario, the patient can send a request to obtain the data package he needs. The data package can be a diagnosis report, an electronic medical record, an image data, and so on. Doctors can also classify valid data packets through mobile devices and then choose to download them for patients. The purpose of this is to reduce the number of downloads of complete personal electronic medical records. In this way, data transmission and energy consumption are reduced. A patient’s data packet consists of {K i ,K i+1 …K n } Composition, where K is a certain item of data. The only data packets actually needed are Patients can be screened by doctors upon request to reduce the amount of data. In this way, a large amount of data can be turned into an effective amount of data.

[0074] Under normal circumstances, there are three main methods for effective data selection.

[0075] First, according to their own needs, patients can propose local lesions or observed data in multiple detection and inspection reports, and download them in categories to reduce the amount of data received by the device.

[0076] Second, it is up to the doctor to judge and send some key diagnostic results to the patient, reducing the process of sending all data.

[0077] Third, scan and compare through the hospital's database, and set the priority to send reports through the intelligent screening mode to reduce the amount of data.

[0078] The diagnosis data of each patient contains information such as time, location, and diagnosis results. The diagnosis results include names, diagnosis categories, diagnosis conclusions, etc. The categories include information such as diastole, calcification, heartbeat, and liver function. By analyzing the diagnosis results and then using a decision tree to filter the patient's diagnosis report. The patient's condition diagnosis data contains numerous data items, most of which can only reflect the patient's physical condition, but cannot reflect the patient's condition, and the key data items that can intuitively reflect the patient's condition can be filtered through a decision tree.

[0079] When a patient requests to inquire about his condition, the patient can select the key data items related to his condition to download according to the classification results of the decision tree, or the doctor can make judgments, and the doctor can use the decision tree to assist in selecting the key diagnosis results to send to the patient, or According to the download records of key data items with the same historical diagnosis results in the hospital database, and all data items in the download records whose download rate exceeds a certain proportion (for example: 40%, 50%, 60%, etc.) are selected as priority data Compare with the current decision tree, select the corresponding data item as the key data item and send it to the patient, in this way reduce the amount of data sent.

[0080] Step 103: Obtain the extraction instruction input by the user;

[0081] This step aims to obtain the user's instruction to extract the key data item after selecting the key data item. Among them, the extraction instruction may be an extraction instruction directly set by a doctor, or an extraction instruction sent by a patient through a mobile terminal or the like. As long as the different methods for acquiring the fetch instruction can achieve the corresponding purpose, the different methods will not affect the protection scope of the present invention.

[0082] Step 104: According to the personal information and the target terminal in the extraction instruction, the key data items associated with the personal information are transmitted to the target terminal. The purpose of this step is to transmit key data items to the final target terminal. Among them, there are many ways to transmit data, such as: directly transmitting to the target terminal through the network; transmitting the data to the target terminal through the intermediate server and cloud server; transmitting to the target through the intermediate mobile terminal through the "carry-store-forward" method Terminal etc. As long as the different transmission modes can achieve the corresponding purpose, the different methods will not affect the protection scope of the present invention.

[0083] Furthermore, in order to speed up the efficiency of data transmission, at the same time it is more convenient and convenient for customers to view the data. In a preferred embodiment of the present application, the transmitting the key data items associated with the personal information to the target terminal specifically includes:

[0084] The target terminal is a target mobile terminal, and the key data is transmitted to the target mobile terminal through at least one intermediate mobile terminal.

[0085] Further, in order to solve the incomplete coverage of the mobile data network, in areas lacking medical resources and medical conditions, doctors can form a wireless self-organizing sensor network with patients in the communication domain by moving. In a preferred embodiment of the present application, the key data item is transmitted to the target mobile terminal through at least one intermediate mobile terminal, which specifically includes:

[0086] Detect the intermediate mobile terminals within the communication range, and send detection messages to all intermediate mobile terminals, and establish at least one communication domain according to the returned first response message, wherein the first response message includes at least the corresponding The number of other intermediate mobile terminals that the intermediate mobile terminal can communicate with and the second response messages of the other intermediate mobile terminals are respectively calculated according to the first response message. The node meeting degree, node similarity, node timeliness and node centrality of the terminal, and each node is calculated according to the node meeting degree, node similarity, node timeliness and node centrality. The central intermediate mobile terminal in the communication domain establishes the connection between the communication domain and the adjacent communication domain through the central intermediate mobile terminal, and uses this to screen out at least one preferred transmission path to transmit the key data item to The intermediate mobile terminal, and instruct the intermediate mobile terminal to detect that other intermediate mobile terminals in the communication domain perform the transmission of the key data item, and use the central intermediate mobile terminal to perform the key inter-communication domain Transmission of data items;

[0087] Instruct the intermediate mobile terminal to transmit the key data item to the target mobile terminal when the target mobile terminal is detected within the communication range.

[0088] In specific application scenarios, the hospital first sends data to field doctors or specific doctors, and then moves or transfers the data to other doctors to approach the target mobile terminal, and enter the wireless transmission of the doctor where the current data is located on the patient’s mobile terminal. When in the sense domain, the data is transmitted to the patient's mobile terminal. After receiving the data provided by the hospital, the patient can forward the data to his social relationship. In this way, patients no longer need to go to the hospital to obtain their own test results, and the patient’s social relationship can also be obtained directly with the patient’s equipment. Through this kind of data transmission, the pressure for patients and other personnel to enter the hospital and doctors is reduced.

[0089] In a multi-aware communication network, due to the mobility of nodes (ie, mobile terminals), the degree of encounter between nodes is different, the data transmission time interval is different, the content of the data information is different, and so on. In order to choose a more reasonable association node, the following concepts need to be defined.

[0090] Node meeting degree (Meeting Degree). The node meeting degree indicates the degree of meeting between two nodes in the T time interval. Suppose nodes m and n, their degree of encounter is MD (m, n), md (m, N) represents the number of encounters between node m and the summary point N in the self-organizing communication domain, md (m, n) represents The number of encounters between node m and node n. Then the formula for the degree of encounter between nodes m and n is:

[0091]

[0092] Node similarity (Similarity Degree). Node similarity indicates the degree of neighbors shared between two nodes. The greater the similarity of the nodes, the greater the number of neighbors shared between the information of the nodes, and the closer the relationship between the nodes.

[0093] Suppose node m and n, their similarity is SD (m, n), sd (m, N) represents the similarity between node m and the summary point N in the self-organizing communication domain, sd (m, n) represents the node The similarity between point m and node n. Then the similarity formula between nodes m and n is:

[0094]

[0095] Node Timely Effective Degree. The node timeliness indicates the degree of timeliness of the routing time between two nodes. That is, the shorter the route delay time between two nodes is, the higher their timeliness is. Since the delay of data transmission in opportunistic networks is very common, the research on node timeliness is very important.

[0096] Suppose nodes m and n, their time validity is TD(m,n), td(m,N) represents the time validity of node m and the summary point N in the self-organizing communication domain, td(m,n) represents Timeliness of node m and node n. Then the time validity formula between nodes m and n is:

[0097]

[0098] among them, Indicates the timeliness of the k-th route established by nodes m and n in the T time period.

[0099] The central degree of the node (Central Degree). The node centrality indicates the degree of connectivity of the node in the self-organizing communication domain. The higher the node centrality, the more neighbors it can connect to, and the more valid data packets it can transmit. Node centrality is the most important parameter index for research on opportunistic network data transmission.

[0100] Suppose a node m, its node centrality is determined by the degree of data transmission between it and its neighbors cd(m, n) out And into the degree cd (m, n) in Decided. Therefore, the center degree CD(m,n) of node m is:

[0101]

[0102] Among them, l represents the number of neighbors n of node m. CD(N) represents the total degree of self-organizing communication domain.

[0103] Through the above definition, we can get the condition that the associated node meets in the opportunistic network communication domain: the associated node has the highest weight among all the nodes in the self-organizing network, that is, the overall utility weight of the node U weight (m, d). U weight (m, d) represents the utility weight from node m to destination node d. According to the aforementioned formula, the calculation formula of the associated node can be obtained:

[0104] U weight (m, d)=αMD(m,n)+βSD(m,n)+γTD(m,n)+δCD(m,n)

[0105] Among them, α, β, γ, and δ represent weighting factors, and α+β+γ+δ=1. Since the node m belongs to the full node N, obviously 0≤U weight (m, d)≤1.

[0106] By calculating the utility value U in the communication domain weight (m, d), the most suitable node in the communication domain can be obtained, the correlation node characteristic (Relationship node Characteristic) CR(m),

[0107] CR(m)=U weight (m, d) max

[0108] That is, the associated node has the largest associated node characteristic value in the communication domain.

[0109] Through analysis and calculation, it is possible to determine the associated nodes in each communication domain, and establish data transmission between different communication domains by setting the associated nodes.

[0110] The specific execution process is as follows:

[0111] Step 1: At time t, divide the communication domain of the self-organizing opportunistic network.

[0112] Step 2: Calculate the central node in each communication domain according to all the aforementioned formulas.

[0113] The third step: The central node responds to its neighbors and establishes a connection channel.

[0114] Step 4: The central node transfers the data packet to the next communication domain node. If the next communication domain node is also the central node, the data packet is iteratively processed and passed to the next communication domain; if it is not the central node The point finds the central node according to all the aforementioned formulas and transmits the data packet to the central node.

[0115] Step 5: Repeat the third and fourth steps. When the target node appears in the communication domain, the data packet is passed to the target node through the central node to complete the communication process.

[0116] By applying the technical solution of the present application, the solution obtains medical diagnosis data input by the user; analyzes the medical diagnosis data, obtains personal information and diagnosis results therein, and classifies the medical diagnosis data according to the diagnosis results, Obtain the key data items in the medical diagnosis data according to the classification result, associate the key data items with the personal information; obtain the extraction instruction input by the user; according to the personal information and the target in the extraction instruction The terminal transmits the key data items associated with the personal information to the target terminal. By applying the technical scheme of this application, the patient data is classified and screened according to the doctor’s diagnosis, so that the hospital only needs to transmit a small part of the key information after completing a medical report, which greatly reduces the load on the communication network and the The waste of communication resources reduces the amount of data sent from the source. When receiving medical report data, patients will not place excessively high requirements on the receiving environment, receiving network, and receiving terminal due to excessive data volume when receiving medical report data, making customers more convenient and quicker View the information you want, which greatly improves the user experience.

[0117] In order to further illustrate the technical idea of ​​the present invention, the technical scheme of the present invention will now be described in combination with specific application scenarios.

[0118] 1. Transmission network design

[0119] In the context of big data mobile regional medical care, the transmission network design is like figure 2 Shown. When the relationship between the patient and the hospital and the patient and the doctor is established, as long as the patient is within the communication range of the self-organizing network, the hospital and the doctor can transmit the data to the patient's mobile device by sending data. After receiving the data provided by the hospital, the patient can forward the data to his social relationship. At the same time, patients can also send their real-time data and needs to doctors or hospitals. After receiving the data, the doctors and hospitals can process the patient data as soon as possible to ensure the safety of the patient. In this way, patients no longer need to go to the hospital to obtain their own test results, and the patient’s social relationship can also be obtained directly with the patient’s equipment. Through this data transfer model, the pressure on patients and other personnel to enter the hospital and doctors is reduced.

[0120] In particular, in areas lacking medical resources and medical conditions, doctors can form a wireless self-organizing sensor network with patients within the communication domain by moving. In this network environment, patients send their own data to doctors through mobile devices. According to the request of the patient, the diagnosis result and inspection report can also be sent to the patient, which solves the problem of medical resource allocation.

[0121] 2. Effective data analysis

[0122] When a doctor diagnoses a patient, he can reduce the misdiagnosis rate through a combination of machine diagnosis and manual diagnosis. In the big data self-organizing sensor network, doctors and hospitals send all diagnostic reports and results to patients through mobile devices, and patients and doctors choose further treatment options based on the results. However, patients may generate a lot of data and information during the diagnosis and treatment process. This information includes the patient's personal information, past medical history, examination items, medical images, hospitalization records and other information. According to statistics from the PET-CT center of a certain hospital, patients who have undergone the PET-CT project examination will produce about 300 images per person on average, with a total size of about 1G. The hospital has more than 3 million patients a year. If all the generated data participates in network communication, it will cause huge network resource consumption.

[0123] In addition, based on these image data, doctors need to fill in the relevant CRF form, that is, the case report form, and record the information reflected in the image. Taking the prostate cancer CRF table as an example, the first-level fields reach 17 items, and the second-level fields reach more than 100 items. It takes about 30 minutes to complete the entry of information for a patient. It is conceivable that, in the face of such complex and cumbersome big data, if the doctor sends all of it to the patient, it will inevitably cause great pressure on network resources and cause unnecessary resource consumption; at the same time, the patient does not have relevant professional knowledge , It is impossible to judge which parts of these data are more important, and which ones need long-term attention to its changing trends, which will make the effective information submerged in other data, resulting in patients not being able to obtain key information.

[0124] In the PET-CT center, on average, each patient will produce hundreds of image data, including image information of the whole body of the subject. These hundreds of images can reflect the distribution of cell metabolism throughout the patient’s body. If there is an abnormal increase in cell metabolism, a bright yellow concentrated area will be produced in the picture, and the area of ​​the concentrated area can be directly measured using related software. Concentration value (SUV) size. For most patients, the lesion is confined to a certain part, such as the liver, prostate, and lymph nodes. Therefore, after the doctor has examined the whole body image of the patient, he only needs to select the image of the concentrated part, pay attention to it, and send it to the patient. In this way, the amount of information sent can be greatly compressed and the amount of data sent can be reduced. For example, in a PET-CT scan with dozens of images, only 3 effective images are actually used for diagnosis. Extracting these 3 images not only reduces the transmission of all data, but also increases the probability of diagnosis. Therefore, in the process of researching big data mobile medical care, data analysis and separation need to be considered.

[0125] Such as image 3 Shown is the data transfer selection process. The patient can send a request to obtain the data package that he needs. The data package can be the diagnosis report, electronic medical record, image and image data and so on that the patient needs. Doctors can also classify valid data packets through mobile devices and then choose to download them for patients. The purpose of this is to reduce the number of downloads of complete personal electronic medical records. In this way, data transmission and energy consumption are reduced. A patient’s data packet consists of {K i , K i+1 …K n }constitute. The only data packets actually needed are Patients can be screened by doctors upon request to reduce the amount of data. In this way, a large amount of data can be turned into an effective amount of data.

[0126] Such as Figure 4 As shown, it shows the effective data selection and decision-making process. Specifically, by categorizing the patient's diagnosis, data screening is achieved, thereby assisting the patient in selecting effective data. Assist doctors to provide a knowledge base when performing data analysis on patients. Such as Figure 4 As shown, the diagnosis data of each patient contains information such as time, location, diagnosis result, and the diagnosis result includes name, diagnosis category, diagnosis conclusion, etc. The category contains information such as diastole, calcification, heartbeat, and liver function. By analyzing the diagnosis results and then using a decision tree to filter the patient's diagnosis report. The patient's condition diagnosis data contains a large number of data items, most of which can only reflect the patient's physical condition, but cannot reflect the patient's condition, and the key data items that can directly reflect the patient's condition are filtered through the decision tree.

[0127] When a patient requests to inquire about his condition, the patient can select the key data items related to his condition to download according to the classification results of the decision tree, or the doctor can make the judgment. The doctor uses the decision tree to assist in selecting the key diagnosis results and send them to the patient. This way reduces the amount of data sent.

[0128] Effective data selection and decision-making process:

[0129] 1. Collect the personal information of the diagnosed patient and the information of the diagnosis result;

[0130] 2. Classify the diagnosis results of patients according to different diagnosis parts;

[0131] 3. Establish a decision tree based on factors such as the patient's diagnosis site and the diagnosis result of each site;

[0132] 4. The diagnosis results of each patient can find the corresponding key data items according to the decision tree;

[0133] 5. When the user requests to query the medical condition, he can directly filter to other data items, and return the effective number of items that the user cares about.

[0134] In order to make the data reception and forwarding more reasonable, this application analyzes and designs WSEDT. The process can follow image 3 Proceed as shown. First, the patient gets an electronic medical record by getting the diagnosis from the hospital and the doctor in the hospital. This case contains a large number of data packets. The hospital stores the electronic case data package on the mobile device. The patient makes a request that contains the data that the patient wants to know. And doctors can provide patients with real-time data analysis through APP, and provide real-time data that is effective for patients to mobile device terminals to provide services for patients. This way of data transmission can not only be transmitted between doctors and patients, but also between patients and patients, patients and society. Such a process constitutes a broad spread of information transmission.

[0135] The following proves that the above method is better than the two current data transmission methods in the mobile medical environment in terms of node energy consumption, node routing overhead, and node transmission success rate: Spray and Wait (S&W) algorithm and Binary Spray and Wait (BS&W) algorithm.

[0136] 1) Node energy consumption in mobile medical environment

[0137] This application takes the social relations of doctors, patients, and patients as nodes in the network. The node needs to consume energy when sending information. When the amount of information sent by a node exceeds the energy of the node, the node will die, that is, the APP device can no longer send information. For any node, only satisfy E send ≤E, where E send Is the energy consumed by sending data, and E is the energy carried by the node. That is, the energy consumed by sending data must not be greater than the energy carried by the node to ensure that the node sends information.

[0138] In the process of sending data packets by the node, every time a data packet is sent to the neighbor, it needs to consume energy. A functional relationship is formed by the number of data packets sent and the energy consumption of the node, namely:

[0139] E(p) {peer to peer }=C 1 ×p (1)

[0140] Where E(p) {peer to peer} Indicates the energy consumed by the node and a neighbor to send data packets, C 1 Represents the node energy coefficient, and p represents the number of data packets sent by the node.

[0141] In the process of sending data packets, the energy coefficient is related to the number of neighbors of the node. Therefore, the node coefficient C 1 for:

[0142] C 1 = N i ×e s (2)

[0143] Where n i Indicates the number of neighbors of the node, e s Indicates the unit of energy consumed by each data packet.

[0144] According to formula (1) and formula (2), the energy consumed by the node to send data can be calculated as:

[0145] E=n i ×e s ×p (3)

[0146] According to formula (3), combining the characteristics of the three algorithms to send data packets, calculate the energy required for each algorithm node to send data packets to all neighbors. which is:

[0147] WSEDT node energy consumption E WSEDT for:

[0148]

[0149] Spray and Wait node energy consumption E S&W for:

[0150]

[0151] The energy consumption of Binary Spray and Wait node is:

[0152]

[0153] Because WSEDT selects data packets for neighbor nodes, that is, the amount of data packets p m ≤p M. That is, the energy consumption of WSEDT is less than Spray and Wait. When WSEDT selects that the number of useful data packets is not more than half of the total number of data packets WSEDT energy consumption is less than Binary Spray and Wait; on the contrary, it is greater than Binary Spray and Wait. However, in the big data environment, the number of nodes is large. As the number of nodes increases, the amount of data transfer of the Binary Spray and Wait algorithm will increase, and the cumulative trend will tend to the Spray and Wait algorithm.

[0154] 2) Node routing overhead in the mobile medical environment

[0155] Since the routing cost of a node is related to the total number of data forwarded by the node, the larger the amount of forwarded data, the greater the routing cost and the higher the copy redundancy. So controlling the routing overhead of the node helps to improve the quality of the network. Especially in a big data environment, the smaller the routing overhead between nodes, the lower the redundancy of the copy, and the less node consumption. Therefore, the routing overhead between nodes in the control algorithm is the key.

[0156] In the mobile medical environment, the routing overhead of sending data packets between doctors and patients, and between patients is O {peer to peer} Yes:

[0157]

[0158] among them, Indicates that the average routing cost of each node in the algorithm is Where o i Represents the routing cost of the i-th node, n represents the number of node pairs; p represents the number of data packets sent by the node.

[0159] According to formula (4), the computer can calculate the routing cost of each node in the network algorithm. which is:

[0160] WSEDT node routing cost O WSEDT for:

[0161]

[0162] Spray and Wait node routing cost O S&W for:

[0163]

[0164] Binary Spray and Wait node routing cost O BS&W for:

[0165]

[0166] Because WSEDT selects data packets for neighbor nodes, that is, the amount of data packets p m ≤p M. WSEDT routing overhead is less than Spray and Wait. When WSEDT selects that the number of useful data packets is not more than half of the total number of data packets, that is When the WSEDT routing overhead is less than Binary Spray and Wait.

[0167] 3) Node transmission success rate in mobile medical environment

[0168] In the network, the success rate of data transmission between nodes is an important indicator to measure network performance. In the big data mobile regional medical environment, improving the success rate of data transmission not only ensures that network data packets can be accurately transmitted between patients and doctors, and between patients and social relations, but also relates to the integrity of patient information, ensuring that patients can get Accurate medical diagnosis. Therefore, improving the success rate of information transmission is the most important indicator of medical care in the big data mobile area.

[0169] For mobile devices with limited energy and storage space, the average transmission success rate is related to the number of data packets sent and received, the energy consumption of the device, and the overhead. When the mobile device is unable to send or receive data packets due to excessive energy consumption or excessive routing overhead resulting in insufficient memory space or excessive consumption of the device. As a result, the relationship between patients and doctors, patients and society cannot communicate information normally.

[0170] So in the big data environment, the transmission success rate between nodes is composed of a triple function.

[0171] which is:

[0172] D {peer to peer }=f(E, O, p{ peer to peer }) (5)

[0173] Where D{ peer to peer} Represents the transmission success rate between nodes, E represents energy consumption, O represents routing overhead, p{ peer to peer } Indicates the number of packets sent and received, namely Where p receive Is the number of packets received, p send Is the number of packets sent.

[0174] In the big data mobile regional medical environment, the average transmission success rate between node pairs is:

[0175]

[0176] Among them, Dk {peer to peer } Represents the transmission success rate between any two nodes, and n represents the number of node pairs.

[0177] According to the functional relationship of formulas (5) and (6), the average transmission success rate between node pairs in the network is:

[0178]

[0179] Where E k , O k , P k { peer to peer } Are the energy consumption, routing overhead, and the number of packets sent and received between any two nodes.

[0180] According to formula (3) and formula (4), E k And O k Is about the data packet p between any two nodes k The function. So according to formula (7), it can be expressed as:

[0181]

[0182] According to formulas (3), (4), (8), the average transmission success rate is a function of the number p of data packets sent by the node.

[0183] When the node has enough energy and storage space. The transmission success rate of a node is determined by the ratio between the data packet received by the neighbor node and the data packet sent by the node. That is, the transmission success rate D is: Among them, D receive Is the reception success rate, D send Is the sending success rate. For the entire network, measure the average transmission success rate of the network Yes: The larger the value, the higher the success rate of all nodes in the network to transmit data packets, and the lower the data loss rate. Therefore, the average transmission success rate can be verified through simulation experiments.

[0184] Algorithm design and simulation experiment

[0185] The WSEDT algorithm design is shown in Table 1.

[0186] Table 1 WSEDT algorithm design

[0187]

[0188] In this algorithm, the routing overhead and node energy consumption change with the sending of data packets, and the time complexity is O(n). The transmission success rate varies according to iterations and changes with data packets. The time is complicated. The degree is O(n).

[0189] The simulation experiment will pass ONE 1.4 for simulation test, and the selected simulation test parameter settings are shown in Table 2.

[0190] Table 2 Simulation parameter settings

[0191]

[0192] Such as Figure 5 As shown, it shows the relationship between energy consumption and nodes in a big data environment. The energy consumption of WSEDT is greater than BS&W when the nodes participating in the transmission of information are before 300. When the number of nodes is greater than 400, the energy consumption of WSEDT is less than BS&W, which proves that WSEDT can save energy when more nodes are involved. When the number of nodes participating in the transmission of information reaches 1000, the energy consumption of the S&W algorithm is 2.5 times that of BS&W and 3 times that of WSEDT. Simulations can prove that WSEDT and BS&W can achieve better results with a large amount of investment in nodes, especially WSEDT is more suitable for use in a big data environment.

[0193] Such as Image 6 As shown, it represents the relationship between routing overhead and nodes. In the experiment, the routing overhead of the S&W algorithm is very high, and as the data that nodes participate in communication increases, the routing overhead is obvious. When the number of nodes is 500, the S&W is about 1000, and when the number of nodes is 1000, the routing cost exceeds 3200. Obviously, the S&W algorithm is not suitable for application in a big data environment. The increase in WSEDT and BS&W is not obvious. Especially when the number of nodes is less than 280, the routing overhead of BS&W is smaller than that of WSEDT. When the number of nodes exceeds 300, the routing overhead of WSEDT is less than BS&W. When the participating nodes reach 1000, the WSEDT routing overhead is 75% of BS&W consumption. It shows that WSEDT has a better optimization effect. It also proves that WSEDT can save routing overhead in a big data environment.

[0194] Such as Figure 7 As shown, the relationship between the transmission success rate and the node is shown. In the initial stage, when the number of nodes participating in the transmission of information is 100, the transmission success rate of WSEDT and BS&W exceeds 40%, and S&W also exceeds 30%. As the number of nodes increases, the transmission success rates of the three algorithms are increasing. This situation shows that the transmission success rate of the opportunistic network algorithm increases as the number of nodes increases. When the node reaches 600, the WSEDT transmission success rate exceeds 70%, when the node reaches 800, the WSEDT transmission success rate exceeds 80%, and when the node reaches 1000, the WSEDT algorithm transmission success rate exceeds 90%. It shows that the more data of WSEDT, the higher the transmission success rate. Moreover, when the node reaches 1000, WSEDT is 1.5 times the S&W transmission success rate. This shows that WSEDT has a very good effect on improving the success rate of network transmission.

[0195] Since the transmission success rate is the most important indicator under the big data mobile medical platform, it is related to the accurate arrival of information between nodes, so it is necessary to calculate the simulation of WSEDT sending data packets at different times. The simulation setting data packet sends out data every 5 seconds, 10 seconds and 15 seconds respectively. The simulation results are as

[0196] Figure 8 Shown.

[0197] From the beginning of the simulation, the transmission success rate of sending data packets in 15s is higher than that of 10s and 5s, but the difference in transmission success rate is less than 10%. This trend has continued. When the number of nodes reaches 500, the transmission success rate difference between 15s and 5s exceeds 10%, and the 10s is relatively stable. When the node reaches 1000, the transmission success rate of 10s and 15s exceeds 90%, and that of 5s is close to 90%. This means that WSEDT can get a higher transmission success rate by issuing data packets at different times. Moreover, when the data packet time interval is longer, the transmission success rate is higher.

[0198] Simulation experiments show that WSEDT is better than traditional algorithms in transmission success rate, routing overhead and energy consumption, and is more suitable for applications in a big data environment.

[0199] Picture 9 , Picture 10 versus Picture 11 It respectively represents the number of different types of medical electronic medical case data texts transmitted in various algorithms.

[0200] Picture 9 It can be clearly seen that in the Spray and wait algorithm, a large amount of data transmission storage space is occupied by image text, which means that when communicating between nodes, image transmission occupies a large amount of resources among the limited network resources. Others, such as diagnosis reports, follow-up information, conclusion reports, etc., do not have enough space for effective transmission, causing the phenomenon that the patient's needs and the actual received data are inconsistent, reducing the efficiency of the transmission process.

[0201] Picture 10 It shows the storage status of various texts in the Binary spray and wait algorithm. Obviously, because the Binary spray and wait algorithm limits the wireless transmission of data packets and reduces the flooding of large-scale data, the data transmission size of the image type is significantly reduced, and the data transmission of other types of data can increase. In particular, the increase in diagnostic reports and test reports is obvious. It shows that the Binary spray and wait algorithm is more optimized than the Spray and wait algorithm in the process of medical data transmission.

[0202] Picture 11 It shows the transmission of various texts in the process of transmitting medical data in the WSEDT algorithm. The WSEDT algorithm uses an effective data transmission mechanism to make the transmission of various texts more balanced. Each node sends demand information before getting the data packet, so that the type of data transmitted is randomly transmitted from a large sample to a more balanced transmission of various indicators. This algorithm strategy has obvious effects on improving the high random transmission of data packets in the opportunistic network; at the same time, for the effective transmission of medical data information, it has played a relatively good role in suppressing data such as image information that occupy a large amount of network resources.

[0203] By establishing the WSEDT opportunistic network algorithm in the big data mobile medical environment, the routing overhead and energy consumption of the node are reduced, and the average transmission success rate is improved. This shows that applying WSEDT to the APP information transmission process can better realize the transmission of information in the big data mobile medical environment. Research on big data mobile medical information transmission algorithms can alleviate the contradiction between large population and small medical resources in developing countries.

[0204] Based on the same inventive concept, embodiments of the present invention also provide a medical data transmission device, such as Picture 12 Shown, including:

[0205] The first obtaining module 1201 obtains medical diagnosis data input by the user;

[0206] The screening module 1202 parses the medical diagnosis data, obtains personal information and diagnosis results therein, classifies the medical diagnosis data according to the diagnosis results, and obtains key data items in the medical diagnosis data according to the classification results To associate the key data item with the personal information;

[0207] The second obtaining module 1203 obtains the extraction instruction input by the user;

[0208] The transmission module 1204 transmits the key data items associated with the personal information to the target terminal according to the personal information and the target terminal in the extraction instruction.

[0209] In a specific application scenario, the screening module 1202 classifies the medical diagnosis data according to the diagnosis result, and obtains key data items in the medical diagnosis data according to the classification result, which specifically includes:

[0210] Classify the diagnosis results according to different diagnosis parts;

[0211] Establishing a decision tree according to the diagnosis part and the diagnosis result of each of the diagnosis parts;

[0212] Obtaining the key data item corresponding to the diagnosis result according to the decision tree.

[0213] In a specific application scenario, the filtering module 1202 obtains the key data item corresponding to the diagnosis result according to the decision tree, which specifically includes:

[0214] Setting the corresponding data item in the decision tree as the key data item according to the user's selection;

[0215] and / or

[0216] Acquire historical acquisition records of the same diagnosis result, select the key data items whose acquisition rate exceeds a threshold value as comparison information according to the historical acquisition records, and compare the priority in the decision tree according to the comparison information Data, filtering out relevant data in the decision tree according to the priority data, and setting the relevant data as the key data item.

[0217] In a specific application scenario, the transmission module 1204 block transmits the key data items associated with the personal information to the target terminal, which specifically includes:

[0218] The target terminal is a target mobile terminal, and the key data is transmitted to the target mobile terminal through at least one intermediate mobile terminal.

[0219] In a specific application scenario, the transmission module 1204 transmits the key data item to the target mobile terminal through at least one intermediate mobile terminal, which specifically includes:

[0220] Detect the intermediate mobile terminals within the communication range, and send detection messages to all intermediate mobile terminals, and establish at least one communication domain according to the returned first response message, wherein the first response message includes at least the corresponding The number of other intermediate mobile terminals that the intermediate mobile terminal can communicate with and the second response messages of the other intermediate mobile terminals are respectively calculated according to the first response message. The node meeting degree, node similarity, node timeliness and node centrality of the terminal, and each node is calculated according to the node meeting degree, node similarity, node timeliness and node centrality. The central intermediate mobile terminal in the communication domain establishes the connection between the communication domain and the adjacent communication domain through the central intermediate mobile terminal, and uses this to screen out at least one preferred transmission path to transmit the key data item to The intermediate mobile terminal, and instruct the intermediate mobile terminal to detect that other intermediate mobile terminals in the communication domain perform the transmission of the key data item, and use the central intermediate mobile terminal to perform the key inter-communication domain Transmission of data items;

[0221] Instruct the intermediate mobile terminal to transmit the key data item to the target mobile terminal when the target mobile terminal is detected within the communication range.

[0222] The device in the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0223] Those of ordinary skill in the art should understand that the discussion of any of the above embodiments is only exemplary, and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples; under the idea of ​​the present invention, the above embodiments or The technical features in different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in the details for the sake of brevity.

[0224] In addition, in order to simplify the description and discussion, and in order not to make the present invention difficult to understand, the well-known power/ground connections to integrated circuit (IC) chips and other components may or may not be shown in the drawings provided. . In addition, the devices may be shown in the form of block diagrams in order to avoid making the invention difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the invention will be implemented (ie These details should be completely within the understanding of those skilled in the art). In the case where specific details (for example, a circuit) are described to describe exemplary embodiments of the present invention, it is obvious to those skilled in the art that it may be possible without these specific details or when these specific details are changed. To implement the present invention. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0225] Although the present invention has been described in conjunction with specific embodiments of the present invention, many substitutions, modifications and variations of these embodiments will be apparent to those of ordinary skill in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the discussed embodiments.

[0226] The embodiments of the present invention are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.