A medical service geographic information modeling method and a modeling platform

Abstract

The application relates to a medical service geographic information modeling method and a modeling platform. The modeling platform comprises a satellite remote sensing image acquisition unit, a DEM data acquisition unit, an image data processing unit, a thematic database input unit, a database and a front-end platform which are connected in communication with each other. The modeling method obtains a high-quality and continuously browsable frontier area map through image preprocessing, data splicing, color adjustment and data fusion of geographic information data, thereby providing guarantee for efficient and accurate data fusion. Thematic data is input into the database and is secondarily fused with the geographic information image. On the basis of the built database, multiple front-end platforms are developed, medical service data, disease data, drug support data, water quality data and system management data are visually displayed on the map, and the medical service efficiency and the medical service quality of remote areas are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent medical technology, and particularly relates to a medical service medical geographic information modeling method and a modeling platform. BACKGROUND

[0002] Medical service refers to the professional organization and work of the medical service personnel of the army using medical theory, technology, equipment and facilities to provide health services for its members, which is referred to as "medical service". The basic task of medical service is to maintain and promote the health of army members and consolidate and improve the combat effectiveness of the army. Medical service usually includes health prevention, medical care, health protection, medicine supply, medical training, medical research, medical service management and other professional work. It needs the cooperation of relevant departments, the implementation of professional support agencies and the participation of army members. However, in the existing medical service support system, due to the lack of effective large databases and medical service information analysis systems, the data resources are scattered, the information fusion degree is poor, and especially in Xinjiang Uygur Autonomous Region of China, the land area is wide, and the medical services of each army stationed area are basically independent. In the face of the prevention and treatment of infectious diseases of the army, the relationship between diseases and geographic information (for example, some army members stationed in some areas are prone to acute and chronic digestive tract diseases, and some areas are prone to urinary system or digestive system diseases due to water source problems), medical personnel often have to consult experts and scholars and consult medical books to solve the problem, which is difficult to solve the problem as a whole. It can be seen that due to the lack of comprehensive big data analysis, the medical service efficiency and service quality of remote areas is low. SUMMARY

[0003] In order to solve the technical problems mentioned in the background art, the present application provides a medical service medical geographic information modeling method and a modeling platform, and the specific technical solutions are as follows:

[0004] A medical service medical geographic information modeling method comprises the following steps:

[0005] Step S1: obtaining an original image by satellite remote sensing; the original image includes remote sensing image data obtained by satellite remote sensing in a border defense area and DEM data of the range of the army stationed area in the border defense area, and the above data is integrated and classified.

[0006] Step S2: image preprocessing: data calibration and coordinate projection conversion are performed on the original image;

[0007] Step S3: precision detection is performed, when the calibration precision does not meet the requirements, returning to step S2, and when the calibration precision meets the requirements, entering step S4;

[0008] Step S4: data splicing: seamlessly splicing adjacent image images, removing redundant images and artifacts in the overlapping area on the basis of maintaining the consistency of ground objects, thereby forming a panoramic image image;

[0009] The process of the above data splicing further includes steps S41-S43:

[0010] Step S41: splicing the panchromatic single scene and the multispectral single scene, on the basis of which, the connection points between each image image are searched, the error is controlled within 3%, all image images are imported into the project and then the connection points between each image are searched, GCP / TP is selected in the processing step, image DEM is imported, other options are selected by default, "connection point" is selected to start automatic running of the connection point;

[0011] Step S42: calculating the Res value, selecting and deleting the values greater than 3, performing neighborhood search on the connection points through the K-means algorithm, and judging whether the connection points are uniformly distributed by setting a first threshold value, if yes, the connection points are uniformly distributed at the junction of the two images, then no treatment is needed, the data is compressed according to the first data format and saved, if not, GCP / TP is selected again, the DEMs of the two images with unevenly distributed connection points are imported, pixelization processing is performed to obtain pixel units, key node queries are performed on the two images through the Kd-tree algorithm, and it is judged whether there are key identical nodes, if yes, "connection point" is selected for connection;

[0012] Step S43: on the basis of step S42, removing redundant images and artifacts in the overlapping area through a high-pass filtering algorithm, thereby obtaining a panoramic image image and completing data splicing;

[0013] Step S5: checking the data splicing quality, if it meets the requirements, entering step S6, if the data splicing accuracy does not meet the requirements, returning to step S4 and performing data splicing again;

[0014] Step S6: color adjustment processing of the image image, fusion of the original image color, restoration of the true color of the ground object or according to the requirements of different application objects, achieving the expected ground color;

[0015] Step S7: data fusion: through an improved Pansharp algorithm, multispectral remote sensing image data sets and panchromatic image data after color adjustment processing are fused, so as to achieve complementary advantages of panchromatic and multispectral data, enhance spatial details, reduce color distortion, and form a relatively complete information description map of the ground target;

[0016] In step S7, the following subdivided steps S71-S73 are further included:

[0017] Step S71: first, the MS image is resampled and interpolated to obtain the LRMS image, and then the I component is separated from the LRMS image by using the weighted average method.

[0018] Step S72: details are extracted from the PAN image, and a guided filter is used to ensure that the trend of detail change is consistent with the LRMS image, and the spatial scaling coefficient β k is solved.

[0019] Step S73: on the basis of step S72, the extracted details are input into the LRMS image, and the intensity modulation is performed by using the spectral modulation parameter α k , so that α k and β k reach the optimal combination, thereby greatly reducing the probability of spatial image distortion.

[0020] Step S74: the spectral modulation parameter α k is used as the correction coefficient of the LRMS image data, and β k is used as the correction coefficient of the PAN image data, and on this basis, the two are added to calculate the fused image data.

[0021] Step S8: on the basis of data fusion, the fused data is compressed and imported into the database.

[0022] On the basis of the medical service geographic information modeling method, a medical service geographic information modeling platform is further proposed.

[0023] A medical service geographic information modeling platform comprises a satellite remote sensing image acquisition unit, a DEM data acquisition unit, an image data processing unit, a special database input unit, a database and a front-end platform, wherein the satellite remote sensing image acquisition unit and the DEM data acquisition unit are in communication connection with the image data processing unit, the image data processing unit is in electrical connection with the database, and the database is in communication connection with the front-end platform.

[0024] The front-end platform comprises a medical service support platform, a disease management platform, a drug support platform, a water source water quality platform and a system management platform. The disease management platform and the drug support platform can realize data sharing.

[0025] The special database input unit sorts, labels and imports the data of the camp, common diseases (including training injuries, psychological diseases, natural epidemic diseases and infectious diseases), drug support and supply, medical evacuation, water source water quality data into the database, and performs secondary fusion with the information description map in the database, that is, combines the corresponding labeled labels with geographic information coordinates and ground information description.

[0026] In summary, the medical geographic information modeling method and modeling platform of the application have the advantages of the prior art.

[0027] 1) The medical geographic information modeling method of the application can obtain high-quality and continuously browsable border defense area maps after image preprocessing, data splicing, color adjustment, data fusion and other processing of geographic information data, thereby providing guarantee for efficient and accurate data fusion.

[0028] 2) Compared with the single parameter adjustment mode and image fusion mode based on GIS images of the prior art, the application combines the K-means algorithm and the Kd-tree algorithm to realize the search for connection points during data splicing, and performs query reconfiguration for uneven connection points, thereby improving the quality of data splicing.

[0029] 3) The application establishes a parameterized model from a multi-target perspective, and through the improved Pansharp algorithm, the edge and texture of the multispectral image are enhanced while the details are integrated, and the multispectral image is sharpened through spectral modulation, so that the quality of all pixels in the fused image can be maximized, and the adverse effects of detail injection are eliminated.

[0030] 4) The application imports the image data of fused geographic information into the database, and enters the thematic data into the database for secondary fusion with the geographic information image, develops multiple front-end platforms based on the built database, and visualizes the medical service, disease data, drug support data, water quality data, and system management data on the map, thereby improving the medical service efficiency and quality of medical service in remote areas. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 The application is a medical geographic information modeling method flowchart.

[0032] Figure 2 The application is an improved Pansharp algorithm block diagram.

[0033] Figure 3 The application is a medical geographic information resource position distribution diagram.

[0034] Figure 4 The application is a medical disease management platform statistical analysis interface. DETAILED DESCRIPTION

[0035] Example 1

[0036] Please check Figure 1 The application is a medical geographic information modeling method, which comprises the following steps:

[0037] Step S1: obtaining original image by satellite remote sensing;

[0038] The original image includes remote sensing image data (data accuracy is 0.5-2m) obtained by satellite remote sensing in the border defense area and DEM data (i.e. digital elevation model, 30m) in the internal team station range of the border defense area, and the above data is integrated and classified.

[0039] Step S2: image preprocessing: data calibration and coordinate projection conversion are performed on the original image; the data coordinate system is unified through image preprocessing, which is convenient for subsequent use;

[0040] Due to the geometric deformation caused by topography and other factors during image shooting, the real spatial coordinates and topographic and geomorphic points will be misaligned, in order to give the data real spatial coordinates for user use, therefore, data calibration process is needed. For example, reasonably distributed correction control points, these points can be selected from existing control data (such as topographic map) through the method of indoor point transfer, or obtained by field measurement. For geometric correction of high-resolution remote sensing image, indoor point transfer generally uses large-scale topographic map, such as 1:500 or 1:1000 topographic map; field measurement generally uses GPS positioning measurement method.

[0041] Step S3: precision detection, when the calibration accuracy does not meet the requirements, return to step S2, when the calibration accuracy meets the requirements, enter step S4;

[0042] Data quality includes: whether there is data missing, whether there is cloud cover, etc. After data calibration, the calibrated image should be evaluated for accuracy (RMS error is less than 1 pixel), only in the case of meeting the requirements, the next step can be carried out, otherwise the data calibration needs to be re-calibrated.

[0043] Step S4: data splicing: adjacent image images are seamlessly spliced, on the basis of maintaining the consistency of ground objects, removing redundant images and artifacts in the overlapping area, thereby forming a panoramic image;

[0044] The above data splicing process further includes steps S41-S43:

[0045] Step S41: splicing full color and multi-spectral single scene, on this basis, find the connection points between each image, the error is controlled within 3%, all image images are imported into the project and start searching the connection points between each image, select GCP / TP in the processing step, import image DEM, other options are selected by default, select "connection point" to start automatic running of connection point;

[0046] Step S42: calculate the Res value, select and delete the value of the Res value (the Res value is the residue in the complex function) greater than 3, perform neighborhood search on the connection points through the K-means algorithm, and judge whether the connection points are uniformly distributed by setting the first threshold value, if yes, the connection points are uniformly distributed at the junction of the two scene images, then there is no need to dispose, the data is compressed according to the first data format and saved, if not, the GCP / TP is selected again, the two scene image DEMs with unevenly distributed connection points are imported, pixelization processing is performed to obtain pixel units, the geodetic coordinate system number and the RGB true color value are automatically added to each pixel unit, key node query is performed on the two scene images through the Kd-tree algorithm, and it is judged whether there is a key same node, if yes, the connection point is selected for connection;

[0047] Step S43: on the basis of step S42, the overlapping area redundant image and the artifact are removed through the high-pass filtering algorithm, so that the panoramic image is obtained, and the data stitching is completed;

[0048] Step S5: the data stitching quality is checked, if the requirement is met, step S6 is entered, if the data stitching accuracy does not meet the requirement, step S4 is returned, and the data stitching is performed again;

[0049] Step S6: the image is processed, the original image color after fusion is processed, the true color of the ground object is restored, or the ground object color meeting the requirements of different application objects is achieved;

[0050] Step S7: data fusion: through the improved Pansharp algorithm, the multispectral remote sensing image data set and the panchromatic image data after color processing are fused, so that the advantages of panchromatic and multispectral data are complementary, the spatial details are enhanced, the color distortion is reduced, and a relatively complete information description map of the ground target is formed;

[0051] Step S8: on the basis of data fusion, the fused data is compressed and imported into the database.

[0052] It can be seen that, compared with the existing technology based on the single parameter adjustment mode and image fusion mode of the GIS image, the medical geographic information modeling method provided by the application realizes the search of the connection points in combination with the K-means algorithm and the Kd-tree algorithm, finds out the uneven connection points, performs query and reconfiguration, and improves the quality of data stitching; in addition, from the perspective of multiple targets, a parameterized model is established, through the improved Pansharp algorithm, the details are integrated and the edges and textures of the multispectral image are enhanced, and the multispectral image is sharpened through spectral modulation, so that the quality of all pixels in the fused image can be maximized, and the adverse effects of detail injection are eliminated.

[0053] Please checkFigure 2 As shown in the above step S7, the process of data fusion based on the improved Pansharp algorithm further comprises the following sub-steps S71-S74:

[0054] Step S71: First, the MS image is resampled and interpolated to obtain the LRMS image, and then the I component is separated from the LRMS image by using the weighted average method.

[0055] Step S72: The details of the PAN image are extracted, and the trend of detail changes is ensured to be consistent with the LRMS image by using a guide filter. The spatial scaling coefficient β k is solved;

[0056] Step S73: On the basis of step S72, the extracted details are input into the LRMS image, and the intensity modulation is performed by using the spectral modulation parameter α k , so that α k and β k achieve an optimal combination, thereby greatly reducing the probability of spatial image distortion;

[0057] Step S74: The spectral modulation parameter α k is taken as the correction coefficient of the LRMS image data (multispectral remote sensing image data set), and β k is taken as the correction coefficient of the PAN image data (panchromatic image data after color matching), and on this basis, the two are added to calculate the fused image data.

[0058] In the above step S72, the guide filter (guide filter) is obtained for the first time to obtain the details of the PAN image. The guide filter is a spatial filter belonging to the MRA-based method. After filtering the input image to produce its low-pass version, the lost high-frequency information is calculated as the input details by image subtraction. Two input images are needed in the guide filter, and the filtering operation is based on a local linear model. One input image is taken as the filtered image, and the other is taken as the guide image, so as to ensure that the trend of change of the output image is consistent with the guide image. As a result, the obtained details will have a high correlation with the guide image. In this embodiment, the I component is taken as the guide image, and the PAN image is taken as the filtered image. Specifically, the I component is first obtained by using the simplest function, as shown in formula (1):

[0059] (1)

[0060] The above K is the MS order.

[0061] Let GF(PAN, I) denote the operation of the guided filter, and OP = GF(PAN, I) is the output image, then the filtering operation can be obtained by a local linear model given by equation (2) as follows:

[0062] (2)

[0063] where OP i and I i are the i-th pixel in the output image OP and the guided image I, respectively, W s is a local window of size (2r + 1) x (2r + 1) centered at pixel s. a s and b s can be estimated by minimizing the squared error between the input and output images as follows:

[0064] (3)

[0065] where PAN i is the i-th pixel of the filtered PAN image, and η is a regularization parameter. a s and b s are obtained by solving equation (3) as follows:

[0066] (4)

[0067] (5)

[0068] Based on the output image OP, PAN detail is calculated as follows:

[0069] (6)

[0070] where l is the number of filters, and OP l is the PAN image of the L-th filter input.

[0071] Spectral modulation of the fused image is one of the key issues of HFDI-based panchromatic sharpening. Therefore, an effective method is needed to maintain high spectral resolution in the fused image. To solve this problem, the spectral modulation parameter a k is calculated by a spectral modulation algorithm based on edge recovery:

[0072] (7)

[0073] On the basis of the calculation of the above formulas (1)-(6), the LRMS image data (multispectral remote sensing image data set) and the PAN image data (panchromatic image data after color matching) are fused, and the specific image fusion calculation formula FMS is as follows:

[0074] (8)

[0075] Unlike the prior art, in the embodiment, based on the improved Pansharp algorithm, a spectral modulation structure model parameter based on multi-objective optimization is constructed. The process of spectral modulation is first to construct a multi-objective decision algorithm to generate a decision mapping ω k . Then the mapping is applied to the LRMS image to perform spectral modulation. Compared with the image fusion algorithm in the prior art, the advantages are: 1) effectively enhancing the edge and texture space of the fused image; 2) effectively modulating the spectrum of the fused image of color information; 3) effectively modulating the intensity of the fused image, eliminating the adverse effects of detail injection.

[0076] Embodiment 2

[0077] A medical service geographic information modeling platform, comprising a satellite remote sensing image acquisition unit, a DEM data acquisition unit, an image data processing unit, a thematic database input unit, a database and a front-end platform, wherein the satellite remote sensing image acquisition unit and the DEM data acquisition unit are in communication connection with the image data processing unit, the image data processing unit is in electrical connection with the database, and the database is in communication connection with the front-end platform.

[0078] A thematic database is established, and according to the input data, the camp data, common diseases (including training injuries, psychological diseases, natural epidemic diseases and infectious diseases), drug security and supply, medical evacuation, water quality data are sorted, labeled, and imported into the database, and secondary fusion is performed with the information description map in the database, that is, the corresponding labeled label is combined with geographic information coordinates and ground information description.

[0079] On the basis of the database, a plurality of front-end platforms are formed, including a medical service support platform, a disease management platform, a drug security platform, a water quality platform and a system management platform.

[0080] Among them, the medical service support platform has the functions of two-dimensional map browsing, resource query and resource location distribution display.

[0081] In the resource query process, the attribute query is carried out through point selection, circle selection, frame selection, polygon selection and the like, and the attribute query can also be carried out. Taking the attribute query as an example, a user can select a certain area range, and the related information in the area range is queried out, the query result is displayed in a list form, and is highlighted on the map. On this basis, the user can select a certain object of interest on the list or the map, and further view detailed information of the object, and the detailed information is displayed in a form similar to a bubble chart around the object. The attribute query refers to the query of point of interest information with the attribute through a keyword information, and the query result is displayed in a form similar to a list, and is highlighted on the map.

[0082] The disease management platform includes statistical analysis, question bank management, questionnaire entry, questionnaire management, training injury, psychological disease, natural resource disease and infectious disease. Through superposition of position information of common diseases on a basic map (i.e. information description map), different thematic symbols are used to directly display position information of various diseases on the map, and the influence of geographical factors on disease occurrence can be indirectly analyzed.

[0083] The following table shows a "military common disease questionnaire" frequently used for questionnaire entry. Military users enter the questionnaire information through the "questionnaire entry" function of the platform. The questionnaire participated by the military users can be entered, and after the questionnaire entry is completed, the entered questions can be deleted and modified.

[0084]

[0085] In the questionnaire management process, the questionnaire content is adjusted for different border defense companies, and different types of questionnaires are added. Information entry is carried out by selecting different questionnaires, and statistical personnel can conveniently carry out statistical analysis. By establishing a border defense training injury questionnaire, the distribution data and rules of common training injuries such as acute high altitude reaction, contusion, joint sprain, abrasion, frostbite, acute waist sprain and lumbar muscle strain in various border defense forces can be imported into the disease management platform.

[0086] The statistical analysis platform mainly classifies and displays the data of the entered questionnaire and carries out statistical analysis, and generates corresponding thematic maps such as column chart, pie chart and line chart, so as to carry out data display and analysis (as shown in the following table). Figure 4

[0087] The drug support platform is used for investigating and entering the types and quantities of drugs consumed by the border defense company in a certain period, and collecting the drug demand information of the border defense company under different natural geographical conditions. The drug support platform and the disease management platform can realize data sharing.

[0088] ​The final goal of medical rescue is to cure the wounded. How to effectively evacuate the wounded to ensure their safety is an important content of medical rescue research. The selection of reasonable evacuation route is an important link to ensure effective evacuation. Taking the border company as the geographical coordinate, the rescue ability, rescue route and rescue transportation mode of the surrounding military and local medical units as data parameters, the GIS statistical module was used to calculate and output the evacuation route in the form of map. The evacuation route can be adjusted in real time according to the road, location of rescue institutions, transportation capacity and other conditions, so as to maintain a dynamic change process.

[0089] Water quality platform is to cooperate with the CDC of a military region to detect the water quality of border company. The detection is carried out according to the Standard Test Methods for Drinking Water GB / T5750-2006, and the evaluation method is carried out according to the Standards for Drinking Water Health GB5749-2006. The detection result of one item is unqualified, which is defined as unqualified sample. The detection items include turbidity, color, odor, visible matter, PH value, aluminum, iron, manganese, copper, zinc, chloride, sulfate, total dissolved solids, etc. The bacterial indicators include total bacterial count, total coliform bacteria, heat-resistant coliform bacteria and Escherichia coli. The instruments include UV-2550 visible spectrophotometer, PF6-3 atomic fluorescence spectrophotometer, AA6800 graphite furnace atomic absorption spectrophotometer and 882 ion chromatograph.

[0090] The system management platform includes user management, role management, menu management and log management, which is mainly used for the management and maintenance of basic data of medical support platform, disease management platform, drug support platform, water quality platform and database, to ensure the normal operation of the whole medical geographic information system.

Claims

1. A method for modeling geographic information in medical support, characterized in that, The modeling method includes the following steps: Step S1: Acquire raw image data through satellite remote sensing; the raw image data includes remote sensing image data of the border defense area and DEM data of the troop stationing area within the border defense area, and integrate and classify the above data; Step S2: Image preprocessing: Perform data calibration and coordinate projection transformation on the original image; Step S3: Perform accuracy detection. If the calibration accuracy does not meet the requirements, return to step S2. If the calibration accuracy meets the requirements, proceed to step S4. Step S4: Data stitching: Seamlessly stitch adjacent image images together. While maintaining the continuity and consistency of ground features, remove redundant images and artifacts in overlapping areas to form a panoramic image. Step S5: Check the data splicing quality. If it meets the requirements, proceed to step S6. If the data splicing accuracy does not meet the requirements, return to step S4 and splice the data again. Step S6: Perform color correction on the image to restore the true colors of the ground features or achieve the desired colors of the ground features according to the needs of different applications. Step S7: Data Fusion: Using the improved Pansharp algorithm, the multispectral remote sensing image dataset is fused with the color-corrected panchromatic image data to form a more complete information description map of the ground targets. Step S7 further includes steps S71-S74: Step S71: First, resample and interpolate the MS image to obtain the LRMS image, and then use a weighted average method to separate the I component from the LRMS image; Step S72: Extract details from the PAN image and use a guided filter to ensure that the trend of detail changes is consistent with that of the LRMS image, adjusting the spatial scaling factor β. k Perform the solution; Step S73: Based on step S72, the extracted details are input into the LRMS image and modulated by the spectral modulation parameter α. k Intensity modulation is performed so that α k With β k Achieve optimal group combination; Step S74: Set the spectral modulation parameter α k As a correction factor for LRMS image data, β k As a correction coefficient for PAN image data, the two are summed to calculate the fused image data. The image fusion calculation formula FMS in step S74 above is as follows: ； In the formula, PAN detail This represents the image attribute information of the panchromatic image data, with the subscript "k" indicating the "k" level image of the MS image; Step S8: Based on the data fusion, compress the fused data and import it into the database.

2. The method for modeling geographic information in medical support as described in claim 1, characterized in that, Step S4 includes steps S41 to S43: Step S41: Stitch together the panchromatic single scene and the multispectral single scene. Based on this, find the connection points between each scene image, with the error controlled within 3%. After all the image images are imported into the project, start searching for the connection points between each scene image. When processing step S41, select GCP / TP, import the image DEM, and select other options by default. Select "Connection Points" to start automatically running the connection points. Step S42: Calculate the Res value, select and delete values ​​with a Res value greater than 3, perform neighborhood search on the connection points using the K-means algorithm, and determine whether the connection points are evenly distributed by setting a first threshold. If so, the connection points are evenly distributed at the boundary of the two images, and no processing is required. Compress the data according to the first data format and save it. If not, select GCP / TP again, import the DEM images of the two images with unevenly distributed connection points, perform pixelation processing to obtain pixel units, and perform key node query on the two images respectively using the Kd-tree algorithm to determine whether there are key identical nodes. If so, select "connection points" to connect them. Step S43: Based on step S42, redundant images and artifacts in the overlapping area are removed by a high-pass filtering algorithm to obtain a panoramic image and complete the data stitching.

3. The method for modeling geographic information in medical support as described in claim 1, characterized in that, A specialized database was established. Based on the entered data, data on camp areas, common diseases, drug support and supply, medical evacuation, and water source and quality were organized, labeled, and imported into the database. The database was then further integrated with the information description map, that is, the corresponding labels were combined with geographic coordinates and ground information descriptions.

4. A medical geographic information modeling platform for healthcare workers, based on the medical geographic information modeling method described in claim 1, characterized in that, The modeling platform includes a satellite remote sensing image acquisition unit, a DEM data acquisition unit, an image data processing unit, a thematic database entry unit, a database, and a front-end platform. The satellite remote sensing image acquisition unit and the DEM data acquisition unit are both communicatively connected to the image data processing unit. The image data processing unit is electrically connected to the database, and the database is communicatively connected to the front-end platform.

5. The medical geographic information modeling platform for healthcare workers according to claim 4, characterized in that: The thematic database entry unit organizes and labels data on camp areas, common diseases, drug support and supply, medical evacuation, and water source and quality based on the entered data, and imports it into the database. It then performs a secondary fusion with the information description map in the database, that is, it combines the corresponding labels with geographic information coordinates and ground information descriptions.

6. The medical geographic information modeling platform for healthcare workers according to claim 4, characterized in that: The platform includes a medical support platform, a disease management platform, a drug supply platform, a water source and water quality platform, and a system management platform. Among them, the disease management platform and the drug supply platform can share data.

Images