Feature database building system for abnormal development of penis

Abstract

The invention discloses a feature database building system for abnormal development of a penis. The system comprises a visual operation interface, an image processing module and a data storage module, the visual operation interface is used for receiving import and labeling operations of a clinician on original penis data; the image processing module is used for generating a feature screenshot according to the operation of a clinician, automatically completing feature representation and classification labeling according to the feature screenshot, and generating a penis feature database; and the data storage module is used for storing the penis feature database as a relational database. By adopting the system provided by the invention, a clinician can directly carry out annotation and database establishment on the data, so that the whole process independence from data acquisition, annotation to database establishment is realized, the data safety and the annotation accuracy are guaranteed, and the dependence on computer professionals is reduced.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of data processing, in particular to a feature database construction system for abnormal penis development. BACKGROUND

[0002] In the clinical diagnosis of abnormal penis development, the AI-enabled diagnosis technology generally has the following problems: (1) High computer professional threshold: high-quality image annotation, structured processing and database saving need to rely on data annotation team or IT personnel, which is time-consuming and costly; (2) Data privacy risk: sensitive medical data needs to be transferred across departments, which inevitably leaks patient privacy data; (3) Slow response to clinical needs: computer team cannot update the annotation rules required by doctors or supplement rare case data in time, resulting in slow AI model iteration. SUMMARY

[0003] In order to solve the problems in the prior art, the present application provides a feature database construction system for abnormal penis development, through which the clinician can directly annotate and construct the database, realizing the full-process self-automation from data acquisition, annotation to database construction, ensuring data security and annotation accuracy, and reducing the dependence on computer professionals.

[0004] The embodiment of the present application provides a feature database construction system for abnormal penis development, comprising: a visual operation interface, an image processing module and a data storage module. The visual operation interface is used to receive the import and annotation operation of the clinician on the original penis data. The image processing module is used to generate feature screenshots according to the operation of the clinician, and automatically complete feature representation and classification annotation according to the feature screenshots, and generate a penis feature database. The data storage module is used to store the penis feature database as a relational database.

[0005] Optionally, the system further presets an abnormal penis feature label library for supporting the clinician to perform feature annotation and custom extension through the visual operation interface.

[0006] Optionally, the system further comprises a display module for displaying the penis feature data in the form of a data table.

[0007] Optionally, each data record in the data table includes: a unique identification ID, an original image containing an annotation box, a feature screenshot, a feature name, a feature annotation result of the clinician and a note.

[0008] Optionally, the system also has a data verification module built-in, for verifying and prompting the labeling integrity of the clinical doctors.

[0009] Optionally, the system also includes a security synchronization module, for synchronizing the database to the hospital data center or AI training platform through an encrypted channel.

[0010] The present application provides a penile abnormal development feature database building system specially designed for doctors, through which the clinical doctors can directly label and build the database, realizing the full-process self-automation from data collection, labeling to database building, without relying on IT technical personnel or complex technical support; the present application helps to improve the multi-center data consistency through a data structure and labeling specification; and the present application supports expanding feature labels and supplementing rare case data on demand, helping to continuously optimize the AI model. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 A visual operation interface schematic diagram is provided for the present embodiment; Figure 2 A data labeling process schematic diagram is provided for the present embodiment. DETAILED DESCRIPTION

[0012] The present application will be further described in detail below in combination with the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings, not the whole structure. EMBODIMENT

[0013] The present embodiment provides a penile abnormal development feature database building system, which includes a visual operation interface, an image processing module and a data storage module.

[0014] Referring to Figure 1 The visual operation interface in the present embodiment is used to receive the import and labeling operation of the clinical doctors on the original penile data.

[0015] Specifically, the clinical doctors can import the medical image original graph to be labeled through the visual operation interface, that is, the medical image original graph of the patient's penile part, which supports formats such as JPG / PNG. The clinical doctors can use the mouse to operate and select the abnormal feature area (ROI) in the original graph with a rectangular frame.

[0016] The visual operation interface also supports multi-format image preview and provides auxiliary tools such as magnifying glass and rotation. The present embodiment also uses color display to mark the progress, such as using yellow background in the image list to represent the completion of labeling of a certain row of images.

[0017] Further, the system further comprises a preset penile abnormality feature label library (such as phimosis, hypospadias, etc.), which is used to support the clinical doctors to perform feature labeling and custom extension through a visual operation interface.

[0018] The image processing module is used to extract the ROI region bounding box according to the operation of the clinical doctors, complete necessary coordinate pixel automatic correction and even filtering functions, then generate a feature screenshot, and then automatically complete feature representation and classification labeling according to the feature screenshot, and generate a penile feature database. In the embodiment, multiple images of the same patient are processed according to multiple independent data of array enhancement logic. Since the number of features is small, labeling merging and conflict detection are not supported. In addition, the system in the application supports batch import / export of data.

[0019] The data storage module is used to store the penile feature database as a relational database (such as SQLite). The image content in the database is data for updating a large model. In order to avoid distortion, the image data is not encrypted and desensitized, but directly put into the database; the data can be accessed by setting special permissions to ensure the safety of the whole data.

[0020] In addition, in the embodiment, the database is preferentially stored locally, and can be synchronized to the cloud after the doctor is authorized, and the doctor's identity information is recorded to realize operation traceability.

[0021] Further, the system further comprises a preset penile abnormality feature label library (such as phimosis, hypospadias, etc.), which is used to support the clinical doctors to perform feature labeling and custom extension through a visual operation interface.

[0022] The system further comprises a display module, which is used to display the penile feature data in the form of a data table.

[0023] Each piece of data in the data table contains a unique identification ID, an original image containing a labeling box, a feature screenshot, a feature name, a feature labeling result of the clinical doctor, and a note description. The note description includes medical history, clinical observation points, doctor's name / ID, and other information.

[0024] Optionally, the system further comprises a data verification module, which is used to real-time prompt the labeling integrity of the clinical doctors.

[0025] Optionally, the system further comprises a security synchronization module, which is used to synchronize the database to a hospital data center or an AI training platform through an encrypted channel.

[0026] Further referring to Figure 2 , Figure 2 A data labeling process schematic diagram is provided for the embodiment.

[0027] As an example, the clinician operation process is as follows: After logging in, the original image of the penis collected in the clinic is uploaded in bulk through the "File Import" function, and the icon of each image is displayed in the image list on the left side of the visual operation interface; Click on a row of the image list, and its image will be opened in the central graphic display area of the program. Use the mouse to draw a rectangular frame to select the abnormal area. The system automatically generates an ROI screenshot and saves it to the same row in the data table on the right. The original image marked with a red selection box is also saved to the same row in the data table; In the fourth column of this row, select the feature type (such as "concealed penis") from the drop-down menu; In the fifth column of this row, enter the patient's age, medical history, and other information in the note description; Click "Save Record" to automatically generate a structured data row containing a unique ID; After completing all the annotations, generate an encrypted SQLite file through the "Export Database" function, which can be directly used for AI model training.

[0028] Note that the above is only the preferred embodiment of the present application and the technical principle applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and those skilled in the art can make various obvious changes, readjustments and substitutions without departing from the scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A database construction system for a feature database of abnormal penile development, characterized in that, The system comprises a visual operation interface, an image processing module, and a data storage module. The visual operation interface is used to receive the import and annotation operations of the original penis data by the clinician. The image processing module is used to generate feature screenshots according to the operations of the clinician, automatically complete feature representation and classification annotation according to the feature screenshots, and generate a penis feature database. The data storage module is used to store the penis feature database as a relational database. The system is also preconfigured with a penis abnormal feature label library to support the clinician to perform feature annotation and custom extension through the visual operation interface.

2. The system of claim 1, wherein, The system further comprises a display module for displaying the penis feature data in the form of a data table.

3. The system of claim 1, wherein, Each data record in the data table includes a unique identification ID, an original image containing an annotation box, a feature screenshot, a feature name, a feature annotation result of the clinician, and a note description.

4. The system of claim 3, wherein, The system further comprises a data verification module for verifying and prompting the annotation integrity of the clinician.

5. The system of claim 1, wherein, The system further comprises a security synchronization module for synchronizing the database to a hospital data center or an AI training platform through an encrypted channel.

6. The system of claim 1, wherein, ​

Images