Molecular pathology automatic diagnosis method and system and storage medium

Abstract

The invention relates to the field of medical systems, in particular to a molecular pathology automatic diagnosis method and system and a storage medium, and the method comprises the steps: reading pathology data, and establishing a pathology database; reading case data; determining diagnosis data by using the case data and the pathological data; and outputting the diagnosis data and the case data corresponding to the diagnosis data to a user for confirmation. The method has the effect of improving the diagnosis efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical systems, and in particular to a molecular pathology automatic diagnosis method, system and storage medium. BACKGROUND

[0002] Molecular pathology is a branch of pathology and a relatively new discipline. Molecular pathology mainly studies and diagnoses diseases by testing molecules in organs, tissues and blood. The specific means can be to analyze pathology by testing the content of each molecule collected in organs, tissues and blood, or to analyze pathology of sliced tissues, organs and blood by staining, slicing and observing under a microscope. However, due to the large magnification of the microscope, the observation area is small, and in order to prevent accidental phenomena, a large number of slices often need to be observed, which is low in efficiency. SUMMARY

[0003] In order to improve the problem of low efficiency of molecular pathology analysis, the present application provides a molecular pathology automatic diagnosis method, system and storage medium.

[0004] The molecular pathology automatic diagnosis method provided by the present application adopts the following technical solution: A molecular pathology automatic diagnosis method comprises: reading pathological data and establishing a pathological database; reading case data; determining diagnosis data by comparing the case data with the pathological data; outputting the diagnosis data and the corresponding case data of the diagnosis data to a user for confirmation.

[0005] By adopting the above technical solution, the diagnosis data is automatically calculated, and then output to the user, and the basis for obtaining the diagnosis data, that is, the corresponding case data, is also output, so as to facilitate the user to confirm the accuracy of the diagnosis data through the case data, which is convenient and efficient, and greatly improves the efficiency.

[0006] Optionally, determining diagnosis data by comparing the case data with the pathological data comprises: The case data comprises case text data and case image data, and the pathological data comprises pathological text threshold and pathological image standard threshold. comparing the case text data with the pathological text threshold to determine whether the case text data falls within the range of the pathological text threshold, and obtaining corresponding diagnosis data; The case image data is compared with the pathological image standard threshold to determine the similarity between the case image data and the pathological image standard threshold, and corresponding diagnosis data is obtained.

[0007] By using the above technical solution, the test results and other case text data are calculated by comparing the case text data with the pathological text threshold, and the similarity data of the case image data is obtained by calculating the case image data and the pathological image standard threshold, thereby realizing multi-directional automatic judgment of text and images, and being more comprehensive.

[0008] Optionally, determining the diagnosis data by comparing the case data with the pathological data comprises: reading a wall threshold, and the pathological image standard threshold comprises a pathological range threshold and a lesion degree threshold; determining wall-in data and wall-out data by comparing the case image data with the wall threshold, determining wall-in proportion data by comparing the wall-in data with the pathological range threshold, and determining wall-in diagnosis data by comparing the wall-in proportion data with the lesion degree threshold, wherein the diagnosis data comprises the wall-in diagnosis data.

[0009] By using the above technical solution, the cell wall is recognized by calculating the wall threshold, so as to recognize and judge the intracellular fluid and the extracellular fluid, and the accuracy is improved.

[0010] Optionally, determining the diagnosis data by comparing the case data with the pathological data comprises: reading invalid threshold data; determining wall-out effective data by comparing the wall-out data with the invalid threshold data, determining wall-out proportion data by comparing the wall-out effective data with the pathological range threshold, and determining wall-out diagnosis data by comparing the wall-out proportion data with the lesion degree threshold, wherein the diagnosis data comprises the wall-out diagnosis data.

[0011] By using the above technical solution, the extracellular fluid that is spilled out is excluded from the calculation by calculating the invalid threshold data, the misleading of the calculation is reduced, and the accuracy is improved.

[0012] Optionally, determining the diagnosis data by comparing the case data with the pathological data comprises: the case data comprises family case data; determining potential cause data by comparing the family case data with the pathological data, determining potential case data by reading the potential cause data in the past and the potential cause data this time, and outputting the potential case data to a user.

[0013] Through the calculation of the family case data, whether the patient has the tendency to develop the family genetic disease is predicted in advance, and the accuracy is further improved.

[0014] Optionally, the diagnostic data and the case data corresponding to the diagnostic data are output to the user for confirmation, including: reading region division data; The case division region data is determined through the region division data and the case image data, the lesion region data is determined through the case division region data, the intramural diagnostic data and the extramural diagnostic data, the diagnostic data is determined through the lesion region data, and is output to the user.

[0015] Through the above technical solution, the region where the lesion occurs is calculated, and the lesion region data is output to the user, so that the user can directly observe the image of the region and confirm the correctness of the diagnostic data, which is convenient and fast.

[0016] Optionally, the diagnostic data and the case data corresponding to the diagnostic data are output to the user for confirmation, including: reading re-division data; If the regions represented by the plurality of lesion region data are adjacent, the adjacent lesion region data are merged to obtain lesion combined region data, the re-division region data is determined through the lesion combined region data and the re-division data, the new lesion region data is determined through the re-division region data, the intramural diagnostic data and the extramural diagnostic data, and the lesion region data is replaced by the new lesion region data.

[0017] Through the above technical solution, the accuracy of the lesion region data is improved, and the workload of user confirmation is further reduced, which is convenient and fast.

[0018] Optionally, the diagnostic data and the case data corresponding to the diagnostic data are output to the user for confirmation, including: reading division reference data; The region difference value is determined through the original lesion region data and the current lesion region data, the region difference value is compared with the division reference data, if the region difference value falls within the range of the division reference data, the diagnostic data is determined through the current lesion region data and is output to the user; If the region difference value falls outside the range of the division reference data, it is judged again whether the regions represented by the plurality of lesion region data are adjacent, if not, the diagnostic data is determined through the current lesion region data and is output to the user, if yes, the step of merging the adjacent lesion region data to obtain lesion combined region data is returned.

[0019] By adopting the technical scheme, the lesion region data is reduced to be accurate enough and then output to the user through circulation calculation, so that the accuracy is greatly improved.

[0020] The application provides a molecular pathology automatic diagnosis system, which adopts the following technical scheme. A molecular pathology automatic diagnosis system comprises: A control panel is used for inputting data by a user to realize human-computer interaction. A processor is used for processing the data input by the user through the control panel and outputting diagnosis data. A database is used for storing data and calling the stored data by the processor. A display module is used for displaying the diagnosis data output by the processor.

[0021] By adopting the technical scheme, the human-computer interaction is realized through the control panel, the data storage and calculation are realized through the processor and the database, and the display output of the data is realized through the display module, so that the convenience and efficiency are improved.

[0022] The application provides a computer readable storage medium, which adopts the following technical scheme. A computer readable storage medium stores a computer program capable of being loaded and executed by a processor to perform a molecular pathology automatic diagnosis method.

[0023] By adopting the technical scheme, the computer program is stored through the computer readable storage medium.

[0024] In summary, the application has at least one of the following beneficial technical effects: 1. Convenient and efficient, greatly improving the efficiency.

[0025] 2. The accuracy of automatic diagnosis is improved, and the workload of the user is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flowchart of a molecular pathology automatic diagnosis method in the embodiment of the application.

[0027] Figure 2 is a flowchart of step S2.

[0028] Figure 3 is a flowchart of step S3.

[0029] Figure 4 is a flowchart of step S5.

[0030] Figure 5 is a flowchart of step S6.

[0031] Figure 6 Fig. 1 is a schematic diagram of a module of a molecular pathology automatic diagnosis system in an embodiment of the present application.

[0032] Reference signs: 1, control panel; 11, processor; 12, database; 13, display module. DETAILED DESCRIPTION

[0033] The following will be described in detail with reference to the accompanying drawings. Figure 1-6 The present application will be further described in detail.

[0034] An embodiment of the present application discloses a molecular pathology automatic diagnosis method. Referring to Fig. 1, the molecular pathology automatic diagnosis method comprises the following steps: Figure 1 S1, reading user-input pathology data, establishing a pathology database for storing the input pathology data; S1, reading user-input pathology data, establishing a pathology database for storing the input pathology data; S11, reading user-input case data; S12, determining diagnosis data by comparing the case data with the pathology data; S13, outputting the diagnosis data and the corresponding case data to the user for confirmation.

[0035] For example, the pathology data is data for confirming illness, such as white blood cell content exceeding a certain value, which is the pathology data; the case data is data contained in the patient's examination report; by comparing the patient's case data with the pathology data, it is determined that, for example, when the data in the case data falls within the pathology data, it is normal, and when the case data exceeds the range of the pathology data, the patient's body has an abnormality, and the case data is automatically judged, and the corresponding diagnosis data is output to the user, i.e. the abnormal case data and the automatically judged diagnosis data are output to the doctor for further confirmation, such as excessive helicobacter pylori content as case data, and helicobacter pylori infection as diagnosis data.

[0036] Referring to Fig. 2, step S12 comprises the following steps: Figure 2 S2, the case data comprises case text data and case image data, and the pathology data comprises pathology text threshold and pathology image standard threshold; S21, comparing the case text data with the pathology text threshold to determine whether the case text data falls within the range of the pathology text threshold, and obtaining the corresponding diagnosis data; S22, comparing the case image data with the pathology image standard threshold to determine the similarity between the case image data and the pathology image standard threshold, and obtaining the corresponding diagnosis data. S22, comparing the case image data with the pathology image standard threshold to determine the similarity between the case image data and the pathology image standard threshold, and obtaining the corresponding diagnosis data.

[0037] For example, the case text data is text, numerical data, etc., including data for molecular testing of organs, tissues, blood, etc., which is simply the content of various substances in human tissues. The case image data is image data, including image data obtained by observing and recording cells, tissues, etc. after staining and observing them under an electron microscope. The case text data is directly compared with the pathology text threshold to obtain the corresponding diagnosis data. For example, the Helicobacter pylori content is directly compared with the normal range to determine whether it is within the normal range. The Helicobacter pylori content is the case text data, and the normal Helicobacter pylori content range is the range of the pathology text threshold. The case image data is compared with the pathology image standard threshold to obtain the similarity diagnosis data. The pathology image standard threshold is the minimum standard for abnormal staining tissues. For example, red is a treatment substance, and five red dots within the display range of an electron microscope are normal. If six red dots are identified in the case image data, it indicates that there may be an abnormality, and the corresponding diagnosis data is output.

[0038] Referring to Figure 3 , step S22 further includes the following steps: S3, reading the wall threshold, the pathology image standard threshold includes the pathology range threshold and the lesion degree threshold; S31, determining the wall-in data and the wall-out data from the case image data and the wall threshold, determining the wall-in proportion data from the wall-in data and the pathology range threshold, and determining the wall-in diagnosis data from the wall-in proportion data and the lesion degree threshold. The diagnosis data includes the wall-in diagnosis data.

[0039] For example, the preset wall threshold can be a color interval. If it is set according to the red-green-blue three primary color parameters, it can be a parameter range such as a (150±5, 20±5, 190±5). The tissue wall, cell wall, etc. are automatically determined by the points falling within the wall threshold range and being continuous, so as to distinguish the inside and outside of the cell. The wall-in data is the data of the inside of the identified cell, and the wall-out data is the data of the outside of the identified cell. The pathology range threshold is the color interval of the pathogenic substance after staining, for example, b (15±5, 15±5, 225±5). The wall-in proportion data is obtained by identifying all the stained substances in the cell, identifying the stained substances falling within the pathology range threshold, and calculating the proportion, i.e. the wall-in proportion data. For example, there are 1000 stained points in the wall-in data, and 100 stained points fall within the pathology range threshold b (15±5, 15±5, 225±5). The wall-in proportion data is 100 / 1000=10%, and the lesion degree threshold is 5%. Since 10%>5%, the cell has been lesioned, and the wall-in diagnosis data of the lesion is output.

[0040] Referring to Figure 3 , step S31 also includes the following steps: S4, reading invalid threshold data; S41, determining wall-outside valid data from wall-outside data and invalid threshold data, determining wall-outside proportion data from wall-outside valid data and pathological range threshold, determining wall-outside diagnosis data from wall-outside proportion data and lesion degree threshold, and diagnosis data includes wall-outside diagnosis data.

[0041] For example, multiple cells in a tissue can lock extracellular fluid in the gap between cells, and when the cells are located at the edge of the slice, the extracellular fluid will overflow, resulting in a much lower concentration of substances in the extracellular fluid at this location than the normal level. The invalid threshold data is used to remove this overflowed extracellular fluid from the calculation. The invalid threshold data can be a concentration threshold of extracellular fluid, or a concentration change threshold of extracellular fluid. If it is a concentration threshold, the wall-outside data is compared with the invalid threshold data. Wall-outside data higher than the invalid threshold data is considered normal and used for calculation, while wall-outside data lower than the invalid threshold data is excluded from the calculation as abnormal. For example, if the invalid threshold data is 100, i.e. there are 100 substances in a unit area, then the range of wall-outside data with a density greater than 100 is defined as wall-outside valid data. If the invalid threshold data is a concentration change threshold, the density change in the wall-outside data is identified and calculated. For example, if the invalid threshold data is 10, i.e. the concentration of one region in the wall-outside data decreases by more than 10 compared to another region, it indicates that the other region has started to overflow, and the other region is excluded from the calculation. Wall-outside proportion data is obtained by identifying all the dyed substances in the extracellular fluid outside the cell in the wall-outside valid data, then identifying the dyed substances within the pathological range threshold, and calculating the proportion, i.e. wall-outside proportion data. For example, if the wall-outside valid data has 1000 dyed points, and 100 dyed points have colors within the pathological range threshold b (15±5, 15±5, 225±5), the wall-outside proportion data is 100 / 1000=10%, and if the lesion degree threshold is 5%, since 10%>5%, it indicates that the extracellular fluid has been diseased, and the wall-outside diagnosis data is output.

[0042] Referring to Figure 4 , step S12 also includes the following steps: S5, case data includes family case data; S51, determining potential cause data from family case data and pathological data, and determining potential case data from reading past potential cause data and this potential cause data, and outputting to the user.

[0043] For example, the family case data is the case data of the relatives of the current patient retrieved from the database, and whether there is a genetic disease history is inquired. If there is a genetic disease history, the data related to the genetic disease in the case data of the patient himself / herself is retrieved, and is compared with the pathological data to obtain potential cause data. For example, if there is a history of diabetes in the family, the urine sugar data, urine protein data, urine leukocyte data and other related data in the case data of the patient himself / herself are directly retrieved, compared with the pathological data to obtain potential cause data, and then the potential cause data of this time is compared with the past potential cause data to obtain potential case data. The potential case data indicates whether the patient himself / herself has a tendency to develop in the direction of the genetic disease, and is output to the doctor for judgment.

[0044] With reference to Figure 5 After step S41, the following steps are further included: S6, reading region division data; S61, determining case division region data through the region division data and the case image data, determining lesion region data through the case division region data, the intramural diagnosis data and the extramural diagnosis data, and determining diagnosis data through the lesion region data; S62, reading re-division data; S63, if the regions represented by the multiple lesion region data are adjacent, merging the adjacent lesion region data to obtain lesion combined region data, determining re-division region data through the lesion combined region data and the re-division data, determining new lesion region data through the re-division region data, the intramural diagnosis data and the extramural diagnosis data, and replacing the original lesion region data with the lesion region data.

[0045] For example: first, distinguish between the inside and outside of the cell, then divide by area division data, for example, set the area division data to 27, then divide the display range of the electron microscope into 27 parts, and then calculate the intramural diagnostic data and the extramural diagnostic data of the 27 divided areas respectively, and then obtain the abnormal area, that is, the lesion area data; If there are multiple adjacent regions represented by lesion area data, it means that the division of this part of the region may not be accurate, for example, in the three consecutive regions of QWE, if the main lesion is the W region, and the lesion material spills from W to both sides, it also causes the position of the Q region close to the W region and the position of the E region close to the W region to also exceed the data, but the position of the Q region away from the W region and the position of the E region away from the W region do not exist lesion material, so further accurate division of the region is needed to reduce the workload of subsequent doctor confirmation, and the re-division data is to combine the connected lesion area data together as a whole region, and then divide the region again through the re-division data, for example, set the re-division data to 9, then combine the three regions of QWE together, and then divide them into 9 parts to obtain 9 re-division region data, and then calculate the intramural diagnostic data and the extramural diagnostic data of the 9 divided re-division region data respectively, and then obtain the abnormal region, that is, the new lesion area data, and then replace the original lesion area data with the new lesion area data as the diagnostic data output.

[0046] Referring to Figure 5 After step S63, the following steps are further included: S7, read division reference data; S71, determine the region difference value through the original lesion area data and the current lesion area data, and compare the region difference value with the division reference data to determine whether the region difference value falls within the range of the division reference data, if yes, then determine the diagnostic data through the current lesion area data and output to the user; S72, if the region difference value falls outside the range of the division reference data, then determine whether there are multiple adjacent regions represented by lesion area data, if no, then determine the diagnostic data through the current lesion area data and output to the user, if yes, then return to the step of combining the adjacent lesion area data to obtain lesion area data.

[0047] For example, if the original lesion region data is complete and the lesion material is in the selected region, and there is no lesion material in other regions, the lesion region data obtained after division will be similar to the original lesion region data, and there is no need to calculate again. The judgment of the division reference data is used to identify and judge this situation, that is, the coordinates of the original lesion region after connection are compared with the coordinates of the new lesion region data after connection to obtain a region difference value, and then the region difference value is compared with the division reference data. For example, the two lesion region data is similar, the region difference value obtained is 1%, the division reference data is 5%, 1%<5%, the region difference value falls within the range of the division reference data, indicating that the lesion region data has sufficient accuracy, and the lesion region data can be output as diagnostic data to the doctor at this time. If the region difference value is 10%, 10%>5%, it indicates that the lesion region data is not accurate enough, and the step S6 is returned to calculate the division of the lesion region data again to obtain new lesion region data.

[0048] The implementation principle of the molecular pathology automatic diagnosis method of the embodiment of the application is that when the pathology database needs to be expanded, the pathology data can be output to realize it. After the patient's case data is input into the system, the diagnostic data and the case data related to the diagnostic data can be automatically judged to obtain the diagnostic data, so that the doctor can confirm whether the diagnostic data is correct according to the output case data.

[0049] The embodiment of the application discloses a molecular pathology automatic diagnosis system. Referring to Figure 6 , the molecular pathology automatic diagnosis system comprises a control panel 1, a processor 11, a database 12 and a display module 13. The control panel 1 is used for inputting data to the processor 11, so as to realize human-computer interaction. In the embodiment, the control panel 1 can adopt a touch screen, or a keyboard combined with a display screen and the like. The processor 11 receives the data input by the user, and can process the database 12. For example, the user inputs the pathology data through the control panel 1, and the processor 11 stores the pathology data into the database 12 to form a pathology database after receiving the pathology data. Or the user inputs the patient's case data through the control panel 1 (by scanning or data transmission), the processor 11 retrieves the pathology data from the database 12, so as to calculate the diagnostic data from the case data, and then displays the diagnostic data through the display module 13. In the embodiment, the display module 13 can be a touch screen, a display screen or a display device of a terminal device of a doctor (such as a mobile phone screen, a computer screen and the like).

[0050] The processor 11 can include a central processing unit such as a CPU or an MPU, or a host system built around a CPU or an MPU, and includes hardware or software. After the meter has the processor 11, people can freely control the meter by programming, and make it run according to people's wishes. The processor 11 can control the local meter, the remote meter, the remote communication component, etc. through the internal protocol. The internal protocol generally refers to all protocols for realizing mutual communication or linkage within the same meter or within the same system, including part or all of the following protocols: human-computer interaction protocol, software / hardware (interface) protocol, C-Bus protocol, I-Bus protocol, etc. With the development of integrated circuit technology, some protocols belonging to the E-Bus protocol are also included in the internal protocol after the E-Bus is integrated into the chip.

[0051] The embodiment of the present application discloses a computer readable storage medium. Referring to Figure 1 , the computer readable storage medium is used for storing a computer program running a molecular pathology automatic diagnosis method, so as to be loaded and executed by the processor 11.

[0052] The computer readable storage medium includes various media capable of storing program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0053] The above are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, so: all equivalent changes made according to the structure, shape, principle of the present application should be covered in the protection scope of the present application.

Claims

1. An automatic molecular pathology diagnosis method, characterized by, The method comprises: reading pathological data to establish a pathological database; reading case data; determining diagnosis data by comparing the case data with the pathological data; outputting the diagnosis data and the case data corresponding to the diagnosis data to a user for confirmation.

2. The method of claim 1, wherein the method comprises: Determining diagnosis data by comparing the case data with the pathological data comprises: the case data comprises case text data and case image data, and the pathological data comprises pathological text threshold and pathological image standard threshold; comparing the case text data with the pathological text threshold to determine whether the case text data falls within the range of the pathological text threshold, thereby obtaining corresponding diagnosis data; comparing the case image data with the pathological image standard threshold to determine the similarity between the case image data and the pathological image standard threshold, thereby obtaining corresponding diagnosis data.

3. The method of claim 2, wherein the method comprises: Determining diagnosis data by comparing the case data with the pathological data comprises: reading wall threshold, and the pathological image standard threshold comprises pathological range threshold and lesion degree threshold; determining wall-in data and wall-out data by comparing the case image data with the wall threshold, determining wall-in proportion data by comparing the wall-in data with the pathological range threshold, and determining wall-in diagnosis data by comparing the wall-in proportion data with the lesion degree threshold, wherein the diagnosis data comprises the wall-in diagnosis data.

4. The method of claim 3, wherein the method comprises: Determining diagnosis data by comparing the case data with the pathological data comprises: reading invalid threshold data; determining wall-out effective data by comparing the wall-out data with the invalid threshold data, determining wall-out proportion data by comparing the wall-out effective data with the pathological range threshold, and determining wall-out diagnosis data by comparing the wall-out proportion data with the lesion degree threshold, wherein the diagnosis data comprises the wall-out diagnosis data.

5. The method of claim 1, wherein the method comprises: Determining diagnosis data by comparing the case data with the pathological data comprises: the case data comprises family case data; determining potential cause data by comparing the family case data with the pathological data, determining potential case data by comparing the potential cause data in the past with the potential cause data this time, and outputting the potential case data to a user.

6. The automatic molecular pathology diagnosis method according to claim 4, characterized in that, Outputting the diagnosis data and the case data corresponding to the diagnosis data to a user for confirmation comprises: reading region division data; determining case division region data by comparing the region division data with the case image data, determining lesion region data by comparing the case division region data, the wall-in diagnosis data and the wall-out diagnosis data, determining the diagnosis data by comparing the lesion region data, and outputting the diagnosis data to a user.

7. The method of claim 6, wherein the method comprises: Outputting the diagnosis data and the case data corresponding to the diagnosis data to a user for confirmation comprises: reading re-division data; If multiple said lesion region data represent adjacent regions, then said adjacent lesion region data are combined to obtain lesion combined region data, said lesion combined region data and said re-division data are used to determine re-division region data, said re-division region data, said intramural diagnostic data and said extramural diagnostic data are used to determine new said lesion region data, and said new lesion region data are used to replace the original said lesion region data.

8. The method of claim 7, wherein the method comprises: Said diagnostic data and said case data corresponding to said diagnostic data are output to a user for confirmation, including: reading division reference data; determining region difference value from the original said lesion region data and the current said lesion region data, comparing said region difference value with said division reference data, if said region difference value falls within the range of said division reference data, then determining said diagnostic data from the current said lesion region data and outputting said diagnostic data to the user; if said region difference value falls outside the range of said division reference data, then determining whether multiple said lesion region data represent adjacent regions, if not, then determining said diagnostic data from the current said lesion region data and outputting said diagnostic data to the user, if yes, then returning to the step of combining said adjacent lesion region data to obtain lesion combined region data.

9. A system for using the automatic molecular pathology diagnosis method according to claims 1 to 8, characterized in that, including: a control panel (1) for a user to input data to realize human-computer interaction; a processor (11) for processing data input by the user through the control panel (1) and outputting diagnostic data; a database (12) for storing data and for calling stored data by the processor (11); a display module (13) for displaying diagnostic data output by the processor (11).

10. A computer-readable storage medium, characterized in that: A computer program capable of being loaded and executed by a processor to implement any one of the molecular pathology automatic diagnosis methods of claims 1 to 8 is stored.

Images