Batch desensitization and format conversion method for ultrasonic cardiogram dicom files

Abstract

Provided in the present invention is a batch desensitization and format conversion method for ultrasonic cardiogram DICOM files, the method comprising: acquiring a DICOM file to be processed; performing a desensitization operation on the DICOM file to be processed to obtain a desensitized DICOM file; extracting a region of interest of the desensitized DICOM file based on a self-adaptive mask image and a priori mask image to obtain a DICOM file of interest; and on the basis of a DICOM image output format, performing encoding and image pixel information processing on the DICOM file of interest to obtain a DICOM file of interest in a specified storage format. The present invention solves the problems of low DICOM data analysis efficiency and low data usability in the prior art.

Description

A method for batch desensitization and format conversion of echocardiogram DICOM files Technical Field

[0001] The present invention relates to the technical field of DICOM file processing, and in particular to a method for batch desensitization and format conversion of echocardiogram DICOM files. Background Art

[0002] Echocardiograms are primarily stored in the DIOCM format in PACS (Picture Archiving and Communication System), including single-frame images and multi-frame videos. Echocardiography, with its advantages such as rich temporal information, lack of radiation, and high cost-effectiveness, has led to an increasing demand for ultrasound examinations in clinical practice. In recent years, artificial intelligence (AI) has rapidly advanced in the field of intelligent echocardiographic analysis. AI-based echocardiographic analysis can automatically obtain a variety of cardiac structural and functional parameters, possessing significant clinical value for the diagnosis and treatment of cardiovascular disease. Large-sample, standardized ultrasound data are the cornerstones of AI algorithm development. Therefore, DICOM image desensitization and format conversion are key steps in intelligent analysis using large-sample ultrasound data. However, DICOM images contain extensive pixel information and patient-related metadata. Manual processing and management are cumbersome, time-consuming, and prone to errors, resulting in low accuracy and efficiency in data analysis. Furthermore, image parameters (such as size and sector angle) vary, and different medical devices and manufacturers may generate DICOM files with varying standards and settings, increasing data heterogeneity and limiting data usability.

[0003] Summary of the Invention

[0004] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide a method for batch desensitization and format conversion of echocardiogram DICOM files, which solves the problems of low efficiency of DICOM data analysis and low data usability in the prior art.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for batch desensitization and format conversion of echocardiogram DICOM files, comprising:

[0007] Get the DICOM file to be processed;

[0008] Performing a desensitization operation on the DICOM file to be processed to obtain a desensitized DICOM file;

[0009] Extracting a region of interest from the desensitized DICOM file based on the adaptive mask image and the priori mask image to obtain a DICOM file of interest;

[0010] According to the DICOM image output format, the DICOM file of interest is encoded and the image pixel information is processed to obtain the DICOM file of interest in a specified storage format.

[0011] Preferably, performing a desensitization operation on the DICOM file to be processed to obtain a desensitized DICOM file includes:

[0012] Using Python's pydicom library to parse the DICOM file to be processed and identify the DICOM file to be processed, replacing the patient's sensitive personal information with a preset string to obtain an initial desensitized DICOM file;

[0013] Using Python's opencv library to read the initial desensitized DICOM file, and identifying sensitive information to obtain sensitive areas;

[0014] The pixel values ​​of the sensitive areas are set to 0 to obtain a desensitized DICOM file.

[0015] Preferably, extracting the region of interest of the desensitized DICOM file based on the adaptive mask image and the priori mask image to obtain the DICOM file of interest includes:

[0016] Obtaining a single-frame DICOM file and a multi-frame DICOM file of the desensitized DICOM file;

[0017] Extracting a region of interest from the single-frame DICOM file based on the prior mask image to obtain a first DICOM file;

[0018] Extracting a region of interest from multiple DICOM files based on the adaptive mask image to obtain a second DICOM file;

[0019] A DICOM file of interest is obtained according to the first DICOM file and the second DICOM file.

[0020] Preferably, extracting a region of interest from a multi-frame DICOM file based on the adaptive mask image to obtain a second DICOM file comprises:

[0021] Acquiring motion information of multiple frames of DICOM files, and obtaining an initial mask image according to the motion information;

[0022] Extract the bounding rectangle of the mask image based on Python's opencv library and construct an adaptive mask image;

[0023] The region of interest of the multi-frame DICOM file is extracted based on the adaptive mask image to obtain a second DICOM file.

[0024] Preferably, encoding the DICOM file of interest and processing image pixel information according to the DICOM image output format to obtain the DICOM file of interest in a specified storage format includes:

[0025] Parsing the DICOM file of interest using the pydicom library to obtain the color space information of the DICOM file of interest;

[0026] Determine whether the color space information is an RGB color space, and if so, obtain pixel information and bit depth information of the DICOM file of interest; if not, convert the color space information into a specified color space using a conversion function and obtain the pixel information and bit depth information;

[0027] Obtaining the number of frames and the dynamic range of pixel values ​​of the DICOM file of interest according to the pixel information and the bit depth information;

[0028] Determine whether the frame number of the DICOM file of interest is 1, and if so, obtain the image storage format of the first DICOM file; if not, obtain the frame rate, image length, and image width of the second DICOM file, and obtain the image storage format of the second DICOM file according to the pixel bit depth information, the frame rate, image length, and image width of the second DICOM file;

[0029] A DICOM file of interest in a specified storage format is obtained according to the image storage format of the second DICOM file and the image storage format of the first DICOM file.

[0030] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0031] The present invention provides a method for batch desensitization and format conversion of echocardiogram DICOM files, comprising: acquiring a DICOM file to be processed; performing a desensitization operation on the DICOM file to be processed to obtain a desensitized DICOM file; extracting a region of interest from the desensitized DICOM file based on an adaptive mask image and a priori mask image to obtain a DICOM file of interest; encoding the DICOM file of interest and processing image pixel information according to a DICOM image output format to obtain a DICOM file of interest in a specified storage format. By performing a desensitization operation on the file to be processed, the present invention improves data security, extracts the region of interest of the data, improves data analysis efficiency and accuracy, and performs batch format conversion on the data to improve data usability. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] FIG1 is a flow chart of a method for batch desensitization and format conversion of echocardiogram DICOM files provided by an embodiment of the present invention;

[0034] FIG2 is a flow chart of DICOM data batch desensitization processing according to an embodiment of the present invention;

[0035] FIG3 is a flow chart of batch extraction of regions of interest of echocardiograms according to an embodiment of the present invention;

[0036] FIG4 is a flow chart of batch format conversion of echocardiogram DICOM data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] The purpose of the present invention is to provide a method for batch desensitization and format conversion of echocardiogram DICOM files, which solves the problems of low DICOM data analysis efficiency and low data usability in the prior art.

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] As shown in FIG1 , the present invention provides a method for batch desensitization and format conversion of echocardiogram DICOM files, comprising:

[0041] Step 100: Obtain the DICOM file to be processed;

[0042] Step 200: performing a desensitization operation on the DICOM file to be processed to obtain a desensitized DICOM file;

[0043] Step 300: extracting the region of interest of the desensitized DICOM file based on the adaptive mask image and the priori mask image to obtain a DICOM file of interest;

[0044] Step 400: encoding and processing the image pixel information of the DICOM file of interest according to the DICOM image output format to obtain the DICOM file of interest in a specified storage format.

[0045] Further, performing a desensitization operation on the DICOM file to be processed to obtain a desensitized DICOM file includes:

[0046] Using Python's pydicom library to parse the DICOM file to be processed and identify the DICOM file to be processed, replacing the patient's sensitive personal information with a preset string to obtain an initial desensitized DICOM file;

[0047] Using Python's opencv library to read the initial desensitized DICOM file, and identifying sensitive information to obtain sensitive areas;

[0048] The pixel values ​​of the sensitive areas are set to 0 to obtain a desensitized DICOM file.

[0049] Specifically, batch desensitization of DICOM data: In intelligent ultrasound imaging research, data desensitization is a key step to protect patient privacy. The specific steps are as follows: ① Use the Python pydicom library to parse DICOM files acquired by different ultrasound devices, identify patient personal information (such as name, age, date of birth, etc.) within the DICOM tag file, and replace sensitive information with "XXXX"; ② Use the Python opencv library to read the DICOM file, identify sensitive patient-related information above the image, and set the pixel values ​​in this area to 0 (i.e., completely black). This achieves data desensitization and ensures secure sharing of medical data. (See Figure 2.)

[0050] Further, extracting the region of interest of the desensitized DICOM file based on the adaptive mask image and the priori mask image to obtain the DICOM file of interest includes:

[0051] Obtaining a single-frame DICOM file and a multi-frame DICOM file of the desensitized DICOM file;

[0052] Extracting a region of interest from the single-frame DICOM file based on the prior mask image to obtain a first DICOM file;

[0053] Extracting a region of interest from multiple DICOM files based on the adaptive mask image to obtain a second DICOM file;

[0054] A DICOM file of interest is obtained according to the first DICOM file and the second DICOM file.

[0055] Specifically, the batch extraction of echocardiographic regions of interest (ROIs) is performed: Clinically acquired echocardiograms contain a large amount of useless pixel information around the sector-shaped ROI. To improve the performance of the intelligent ultrasound image analysis model, this patent uses the Python pydicom library to parse large samples of DICOM ultrasound data.

[0056] For single-frame images, regions of interest (ROIs) were extracted from single-sector, double-sector, or triple-sector DICOM files acquired by various ultrasound devices (including EPIQ 5, EPIQ 7, EPIQ 7C, IE Elite, iE33, etc.) based on the prior mask image.

[0057] Furthermore, the region of interest of the multi-frame DICOM file is extracted based on the adaptive mask image to obtain a second DICOM file, including:

[0058] Acquiring motion information of multiple frames of DICOM files, and obtaining an initial mask image according to the motion information;

[0059] Extract the bounding rectangle of the mask image based on Python's opencv library and construct an adaptive mask image;

[0060] The region of interest of the multi-frame DICOM file is extracted based on the adaptive mask image to obtain a second DICOM file.

[0061] Specifically, for multiple frames of images, an initial mask image is obtained based on the motion information of two consecutive frames. Then, the Python OpenCV library is used to extract the bounding rectangle of the mask image and construct an adaptive mask image (the pixel value within the rectangular area is 255, and the remaining pixel values ​​are 0). Finally, the region of interest of the multi-frame DICOM file is extracted based on the adaptive mask image.

[0062] Based on the position information of the minimum bounding rectangle of the fan-shaped area of ​​the mask image, the original image is cropped to increase the ratio of the region of interest to the background, which helps the artificial intelligence algorithm to quantitatively analyze the echocardiogram.

[0063] Furthermore, encoding and image pixel information processing are performed on the DICOM file of interest according to the DICOM image output format to obtain the DICOM file of interest in a specified storage format, including:

[0064] Parsing the DICOM file of interest using the pydicom library to obtain the color space information of the DICOM file of interest;

[0065] Determine whether the color space information is an RGB color space, and if so, obtain pixel information and bit depth information of the DICOM file of interest; if not, convert the color space information into a specified color space using a conversion function and obtain the pixel information and bit depth information;

[0066] Obtaining the number of frames and the dynamic range of pixel values ​​of the DICOM file of interest according to the pixel information and the bit depth information;

[0067] Determine whether the frame number of the DICOM file of interest is 1, and if so, obtain the image storage format of the first DICOM file; if not, obtain the frame rate, image length, and image width of the second DICOM file, and obtain the image storage format of the second DICOM file based on the pixel information and bit depth information, the frame rate, image length, and image width of the second DICOM file;

[0068] The DICOM file of interest in a specified storage format is obtained according to the image storage format of the second DICOM file and the image storage format of the first DICOM file, as shown in FIG3 .

[0069] Specifically, echocardiography DICOM data batch format conversion: Python's pydicom library is used to parse DICOM files, and the color space information of each data is extracted from (0028,0004) PhotometricInterpretation, where (0028,0004) is the Tag of DICOM data, which indicates the address where the corresponding data is stored. PhotometricInterpretation is the data description, indicating that the color space information of the data is stored at (0028,0004); it is mainly divided into "MONOCHROME", "YBR_FULL" or "YBR_FULL_422", and "RGB", where "MONOCHROME" indicates a single-channel grayscale image; "YBR_FULL" or "YBR_FULL_422" indicates a YCbCr (luminance, chroma blue, chroma red) color space image, and the opencv function cv2.cvtColor function is used to uniformly convert the image color space into "RGB", then extract pixel information from (7FE0,0010)PixelData, and extract pixel information from (0028,0100)Bits. The dynamic range of pixel values ​​extracted in the allocated data is generally divided into 8 bits (each pixel is represented by 8 bits) or 16 bits (each pixel is represented by 16 bits). Because the bit depth affects the image's ability to represent subtle changes in intensity, this patent can save a single-frame image as an image in a specified format (JPG, PNG, BMP, and TIFF) of the corresponding bit depth based on the bit depth information of the original DICOM file. For multi-frame images, the pixel information of each frame is extracted and stored in an array. At the same time, the frame rate is obtained from (0008,2144) Recommended Display Frame Rate, and the width and length of the image are obtained from (0028,0010) Rows and (0028,0011) Columns, respectively. Based on the pixel information, frame rate, length, and width data extracted from the DICOM file, the multi-frame image is encoded into a video in a specified format (AVI, MP4, MOV, and WMV) using Python's opencv library. Among them, (7FE0,0010), (0028,0100), (0008,2144), (0028,0010), and (0028,0011) are the addresses for storing pixel data in the DICOM file. PixelData, Bits Allocated, Recommended Display Frame Rate, Rows, and Columns are the tag descriptions for storing pixel data in the DICOM file. In addition, the present invention supports resaving desensitized data in DICOM format to increase the versatility and usability of the data, as shown in Figure 4.

[0070] The beneficial effects of the present invention are as follows:

[0071] (1) Batch Desensitization of DICOM Data: Currently, many single-frame or multi-frame DICOM images acquired by ultrasound devices face certain challenges in desensitization. Traditional technologies are unable to meet the needs of batch desensitization of DICOM files acquired by different devices, especially in terms of removing sensitive patient-related information from DICOM tag files or images. To address this issue, this patent proposes a batch desensitization method that can efficiently and reliably remove sensitive information and ensure the secure sharing of medical data.

[0072] (2) Batch extraction of regions of interest in echocardiograms: In addition to the fan-shaped regions of interest in echocardiograms, there is often a large amount of useless pixel information around them. This information not only takes up storage space, but may also interfere with the performance of the intelligent analysis model. However, traditional software and technology are usually unable to design adaptive mask images for different ultrasound devices and effectively extract regions of interest, resulting in a certain degree of impact on the accuracy and efficiency of intelligent analysis. This patent innovatively constructs adaptive mask images for single-frame and multi-frame DICOM images, respectively. It can achieve adaptive extraction of pixel information of regions of interest for DICOM images collected by different ultrasound devices, which helps to improve the accuracy and efficiency of intelligent analysis.

[0073] (3) Batch format conversion of echocardiogram DICOM data: Generally, reading DICOM files usually requires the use of a dedicated parser. This process is not only complicated, but also reduces the usability of DICOM data. The existing technology mainly converts DICOM files into single-frame images in JPG format. This patent proposes a method for batch format conversion of DICOM data. By parsing DICOM files, multiple frames of ultrasound images are extracted and encoded into a universal playable file format, where single-frame images can support JPG, PNG, BMP and TIFF, while multiple-frame videos can support AVI, MP4, MOV and WMV. At the same time, this patent supports re-saving desensitized data in DICOM format. This technology not only greatly simplifies the processing flow of DICOM data, but also provides more flexible data format options for the widespread application of medical imaging.

[0074] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0075] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for batch desensitization and format conversion of echocardiogram DICOM files, characterized in that, it includes: Obtain the DICOM files to be processed; Perform desensitization operations on the DICOM files to be processed to obtain desensitized DICOM files; Extract the region of interest of the desensitized DICOM files based on the adaptive mask image and the prior mask image to obtain the DICOM files of the region of interest; According to the DICOM image output format, perform encoding and image pixel information processing on the DICOM files of the region of interest to obtain the DICOM files of the region of interest in the specified storage format.

2. The method for batch desensitization and format conversion of echocardiogram DICOM files according to claim 1, characterized in that, Performing desensitization operations on the DICOM files to be processed to obtain desensitized DICOM files includes: Using the pydicom library of Python to parse the DICOM files to be processed and identify the DICOM files to be processed, and using a preset string to replace the sensitive personal information of the patient to obtain the initial desensitized DICOM files; Using the opencv library of Python to read the initial desensitized DICOM files and identify the sensitive information to obtain the sensitive regions; Set the pixel values of the sensitive regions to 0 to obtain the desensitized DICOM files.

3. The method for batch desensitization and format conversion of echocardiogram DICOM files according to claim 1, characterized in that, Extracting the region of interest of the desensitized DICOM files based on the adaptive mask image and the prior mask image to obtain the DICOM files of the region of interest includes: Obtain the single-frame DICOM files and multi-frame DICOM files of the desensitized DICOM files; Based on the prior mask image, extract the region of interest of the single-frame DICOM files to obtain the first DICOM files; Extract the region of interest of the multi-frame DICOM files based on the adaptive mask image to obtain the second DICOM files; Obtain the DICOM files of the region of interest according to the first DICOM files and the second DICOM files.

4. The method for batch desensitization and format conversion of echocardiogram DICOM files according to claim 3, characterized in that, The extracting the region of interest of the multi-frame DICOM files based on the adaptive mask image to obtain the second DICOM files includes: Obtain the motion information of the multi-frame DICOM files, and obtain the initial mask image according to the motion information; Based on the opencv library of Python, extract the circumscribed rectangle of the mask image to construct the adaptive mask image; Extract the region of interest of the multi-frame DICOM files based on the adaptive mask image to obtain the second DICOM files.

5. The method for batch desensitization and format conversion of echocardiogram DICOM files according to claim 1, characterized in that, Encoding the DICOM file of interest and processing the image pixel information according to the DICOM image output format to obtain the DICOM file of interest in the specified image storage format, including: Parsing the DICOM file of interest using the pydicom library to obtain the color space information of the DICOM file of interest; Judging whether the color space information is the RGB color space. If so, obtaining the pixel information and bit depth information of the DICOM file of interest; if not, converting the color space to the specified color space using a conversion function and obtaining the pixel information and bit depth information; Obtaining the number of frames and the dynamic range of pixel values of the DICOM file of interest according to the pixel information and bit depth information; Judging whether the number of frames of the DICOM file of interest is 1. If so, obtaining the image storage format of the first DICOM file; if not, obtaining the frame rate, image length, and image width of the second DICOM file, and obtaining the image storage format of the second DICOM file according to the pixel bit depth information, the frame rate, image length, and image width of the second DICOM file; Obtaining the DICOM file of interest in the specified storage format according to the image storage format of the second DICOM file and the image storage format of the first DICOM file.

Images