WO2017193581A1 - 乳腺筛查图像自动处理系统及方法 - Google Patents

乳腺筛查图像自动处理系统及方法 Download PDF

Info

Publication number
WO2017193581A1
WO2017193581A1 PCT/CN2016/109736 CN2016109736W WO2017193581A1 WO 2017193581 A1 WO2017193581 A1 WO 2017193581A1 CN 2016109736 W CN2016109736 W CN 2016109736W WO 2017193581 A1 WO2017193581 A1 WO 2017193581A1
Authority
WO
WIPO (PCT)
Prior art keywords
breast
image
screening
mammary gland
screening image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2016/109736
Other languages
English (en)
French (fr)
Inventor
张贯京
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Qianhai AnyCheck Information Technology Co Ltd
Original Assignee
Shenzhen Qianhai AnyCheck Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Qianhai AnyCheck Information Technology Co Ltd filed Critical Shenzhen Qianhai AnyCheck Information Technology Co Ltd
Publication of WO2017193581A1 publication Critical patent/WO2017193581A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
    • G16Z99/00—Subject matter not provided for in other main groups of this subclass
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/30—Noise filtering
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03—Recognition of patterns in medical or anatomical images

Definitions

  • the present invention relates to the field of image processing and recognition technologies, and in particular, to a breast screening image automatic processing system and method.
  • Breast disease is one of the most common diseases in women. Especially with the accelerated pace of modern urban life, women's life pressure and work pressure are increasing. The incidence of breast diseases is increasing year by year. Serious breast diseases may be serious. Lead to breast cancer. According to incomplete survey statistics, the incidence and mortality of breast diseases are also on the rise. In fact, the cause of death in most breast cancer patients is because they have not found the best treatment for the disease. Therefore, it is necessary for women to have a breast screening, early detection of breast disease, and early diagnosis for women's health.
  • breast screening images for assisted diagnosis has become an important method for screening and diagnosing early stage breast cancer.
  • computer-aided detection methods for breast screening and screening of breast screening images usually require doctors to manually animate the approximate range of breast cancer suspicious area (ROI) in breast images, breast screening and detection efficiency is not high, no A census for a large number of breast samples.
  • ROI breast cancer suspicious area
  • the main object of the present invention is to provide a mammography screening image automatic processing system and method, aiming at solving the problem that the existing computer-aided detection method requires a doctor's hand to animate the suspicious region of the breast lesion, resulting in low efficiency and accuracy. .
  • the present invention provides an automatic mammography screening image processing system for use in a medical terminal device connected to a medical information system platform via a communication network.
  • the milk The gland screening image automatic processing system comprises: a mammary image acquisition module, configured to acquire a user's mammary gland screening image from the breast image database of the medical information system platform according to the user's mammography screening number; a mammary image processing module, The breast screening image is subjected to distortion-free noise filtering to obtain a distortion-free breast screening image, and the undistorted breast screening image is subjected to gray layering processing to obtain a gray-layered breast screening image; a texture feature extraction module, configured to compare a texture distribution difference between a gray-scale layered breast screening image and a normal breast image to extract a breast texture feature region from the breast screening image; a breast image output module, configured to The mammographic texture feature region is marked in the mammography screening image after grayscale stratification, and the mammary gland screening image labeled with
  • the breast image acquisition module is further configured to receive a breast screening number input by a user through an input unit of the medical terminal device.
  • the breast image output module is further configured to display the breast screening image indicating the characteristic area of the mammary gland on the display screen of the output unit, and the printer printing station that controls the output unit.
  • the breast image database stores breast screening images generated by different users in a breast examination and a screening, the normal breast images being stored in a storage unit of the medical terminal device, the texture distribution Differences include differences in tissue structure, size differences, and contour differences in the breast.
  • the grayscale layering process comprises dividing the breast screening image into different regions according to gray scales and performing color assignment processing on each region.
  • the present invention also provides a method for automatically processing a breast screening image, which is applied to a medical terminal device, and the medical terminal device is connected to a medical information system platform through a communication network, and the breast screening image is
  • the automatic processing method includes the steps of: obtaining a breast screening image of the user from the breast image database of the medical information system platform according to the user's breast screening number; performing the distortion-free noise filtering processing on the breast screening image to obtain no distortion The mammary gland screening image; the undistorted mammary gland screening image is subjected to gray layering to obtain a mammographic screening image after grayscale layering; comparing the mammographic screening image after grayscale layering with the normal mammary gland image a texture distribution difference to extract a mammary gland texture feature region from the mammary gland screening image; the mammary gland texture feature region is marked in a grayscale layered mammary gland screening image; output unit through the output of the medical terminal device Characteristic area of mammary gland Mammary screening
  • the breast screening image automatic processing method further comprises the step of: receiving, by the input unit of the medical terminal device, a breast screening number input by a user.
  • the automatic mammography screening image processing method further comprises the steps of: displaying the mammography screening image marked with a mammographic texture feature area on a display screen of the output unit; controlling the output unit The printer prints a breast screening image labeled with a characteristic area of the mammary gland.
  • the breast image database stores breast screening images generated by different users in a breast examination and a screening, the normal breast images being stored in a storage unit of the medical terminal device, the texture distribution Differences include differences in tissue structure, size differences, and contour differences in the breast.
  • the grayscale layering process comprises dividing the breast screening image into different regions according to gray scales and performing color assignment processing on each region.
  • the breast screening image automatic processing system and method of the present invention adopts the above technical solutions, and achieves the following technical effects: by performing noise removal and grayscale layering processing on the breast screening image, The breast texture feature area is extracted from the processed breast screening image and the breast texture characteristic area is marked in the breast screening image for reference for diagnosis and screening of the breast disease, thereby assisting the doctor to improve the breast disease.
  • the efficiency and accuracy of testing and screening have improved the social efficiency of breast screening.
  • FIG. 1 is a schematic diagram of an application environment of a preferred embodiment of the mammography screening image automatic processing system of the present invention
  • FIG. 2 is a flow chart of a preferred embodiment of the method for automatically processing breast screening images of the present invention.
  • FIG. 1 is a schematic diagram of an application environment of a preferred embodiment of the automatic mammography screening image processing system of the present invention.
  • the mammography screening image automatic processing system 10 is installed and operated in the medical terminal device 1.
  • the medical terminal device 1 establishes a communication connection with the medical information system platform 2 via the communication network 3.
  • the medical terminal device 1 may be disposed in a medical examination institution such as a large hospital or a community medical workstation, and the medical terminal device 1 may be a computing device having a data processing and communication function such as a computer or a cloud server.
  • the medical information system platform 2 may be a cloud platform or a server in a cloud platform, and may be disposed in a medical inspection institution such as a hospital or a community medical workstation.
  • the medical information system platform 2 includes a breast image database 20 that stores breast screening images of female breast examinations and census sputums.
  • the communication network 3 may be an internet network including a local area network, a wide area network, or a wireless transmission network including GSM, GPRS, and CDMA.
  • the medical terminal device 1 includes, but is not limited to, a mammography screening image automatic processing system 10, an input unit 11, a storage unit 12, a processing unit 13, and an output unit 14.
  • the input unit 11, the storage unit 12 and the output unit 14 are all connected to the processing unit 13 via a data bus and can be exchanged with the breast screening image automatic processing system 10 via the processing unit 13.
  • the input unit 11 may be an input device such as a keyboard or a handwriting touch screen for inputting user information by the user. For example, information such as a breast screening number or a user name of the user may be input on the input unit 11.
  • the storage unit 12 can be a read only memory unit ROM, an electrically erasable memory unit EEPROM or a flash memory unit FLASH.
  • the processing unit 13 may be a central processing unit (CPU), a microprocessor, a microcontroller (MCU), a data processing chip, or an information processing unit having data processing functions.
  • the output unit 14 can be a display screen for displaying a breast screening report or a printer for printing a breast screening report.
  • the breast screening image automatic processing system 10 includes, but is not limited to, a breast image acquisition module 101, a breast image processing module 102, a texture feature extraction module 103, and a breast image output module 104.
  • module refers to a series of computer program instruction segments that can be executed by the processing unit 13 of the medical terminal device 1 and capable of performing a fixed function, which is stored in the medical end In the storage unit 12 of the end device 1.
  • the breast image acquisition module 101 is configured to receive a breast screening number input by a user through the input unit 11, and acquire a breast screening image of the user from the breast image database 20 according to the breast screening number of the user.
  • the user can input the user's mammary screening number through an input unit 11 (e.g., a keyboard or a handwritten touch screen), and the breast image database 20 stores breast screening images generated by different users for breast examination and screening.
  • the breast image processing module 102 is configured to perform the distortion-free noise filtering process on the breast screening image to obtain a distortion-free breast screening image.
  • the breast image processing module 102 performs a distortion-free noise filtering process on the breast screening image by using a Gaussian filter function to remove impurities of the breast screening image to obtain a distortion-free breast screening image, thereby Improve the accuracy of testing and screening for breast diseases.
  • the Gaussian filter function is a conventional method for filtering an image by a person skilled in the art, and is not specifically described in this embodiment.
  • the breast image processing module 102 is further configured to perform grayscale layering processing on the distortion-free breast screening image to obtain a gray-layered breast screening image to enhance layering of the breast screening image. display effect.
  • the grayscale layering process is also referred to as density layering processing, which comprises dividing the breast screening image into different regions according to gray scale and performing color assignment processing on each region, so that the breast screening image reaches the score. The effect of the layer display.
  • the mammary gland screening image after grayscale layering can more clearly display the texture distribution on the mammary gland screening image, such as the tissue structure, size and contour of the mammary gland.
  • the texture feature extraction module 103 is configured to compare the texture distribution difference between the gray-layered breast screening image and the normal breast image to extract the breast texture feature region from the breast screening image.
  • the normal breast image is stored in the storage unit 12 as a reference for comparison with the user's breast screening image.
  • the difference in texture distribution includes differences in tissue structure, size difference, and contour profile of the mammary gland. This extraction of breast texture features from breast screening images can help doctors improve the efficiency and accuracy of detection and screening for breast diseases.
  • the breast image output module 104 is configured to mark a breast texture feature area in the gray-scale layered breast screening image, and output a breast screening image indicating the breast texture feature area through the output unit 14 for the doctor For the diagnosis and screening of breast diseases.
  • the breast image output module 104 is also used to The breast screening image indicating the characteristic area of the mammary gland is displayed on the display screen of the output unit 14, or the printer controlling the output unit 14 prints the breast screening image indicating the characteristic area of the mammary gland for the doctor Provide a basis for diagnosis and treatment of breast diseases.
  • the present invention also provides a method for automatically processing a breast screening image, which is applied to a medical terminal device.
  • Fig. 2 is a flow chart showing a preferred embodiment of the automatic processing method for mammography screening images of the present invention.
  • the automatic processing method for mammography screening image includes the following steps:
  • Step S21 receiving a breast screening number input by the user through the input unit; in the embodiment, the breast image acquiring module 101 receives the breast screening number input by the user through the input unit 11 (for example, a keyboard or a handwriting touch screen).
  • the input unit 11 for example, a keyboard or a handwriting touch screen.
  • Step S22 obtaining the breast screening image of the user from the breast image database according to the user's breast screening number; specifically, the breast image acquiring module 101 acquires the breast image database 20 according to the user's breast screening number.
  • the user's breast screening image In this embodiment, the user can input the user's mammary gland screening number through the input unit 1 1 , and the breast image database 20 stores breast screening images generated by different users for breast examination and screening.
  • Step S23 performing a distortion-free noise filtering process on the breast screening image to obtain a distortion-free breast screening image; specifically, the breast image processing module 102 performs the distortion-free noise filtering process on the breast screening image. Undistorted breast screening images.
  • the breast image processing module 102 performs a distortion-free noise filtering process on the breast screening image by using a Gaussian filter function to remove impurities of the breast screening image, thereby improving detection and screening of breast diseases. Check the accuracy.
  • Step S24 performing a grayscale layering process on the undistorted breast screening image to obtain a grayscale layered breast screening image; specifically, the breast image processing module 102 images the distortionless breast screening image.
  • the gray layering process is performed to obtain a gray layered breast screening image to enhance the layered display effect of the breast screening image.
  • the grayscale layering process is also referred to as density layering processing, which comprises dividing the breast screening image into different regions according to gray scale and performing color assignment processing on each region, thereby achieving the gray scale image of the breast.
  • the effect of the layer display the mammography screening image after the grayscale layering process can more clearly show the texture distribution on the mammary gland screening image, such as the tissue structure, size and contour of the mammary gland.
  • Step S25 comparing the texture distribution between the mammographic screening image and the normal breast image after the gray layering
  • the breast texture feature region is extracted from the breast screening image; specifically, the texture feature extraction module 103 compares the texture distribution difference between the gray-scaled breast screening image and the normal breast image to screen from the breast.
  • the breast texture feature area is extracted from the image.
  • the texture distribution difference includes a difference in tissue structure, a size difference, and a contour difference of the breast.
  • Step S26 marking the mammary gland texture feature region in the grayscale layered breast screening image; specifically, the mammary gland image output module 104 indicates the mammary gland texture feature in the grayscale layered mammary gland screening image Area to help doctors improve the efficiency and accuracy of testing and screening for breast diseases.
  • Step S27 outputting a breast screening image indicating a characteristic area of the mammary gland by the output unit for the doctor to diagnose and screen the mammary gland; specifically, the mammographic image output module 104 outputs the mammary gland texture characteristic through the output unit 14.
  • Regional breast screening images are available for physicians to diagnose and screen for breast disease.
  • the method further includes the steps of: displaying the mammography screening image marked with the mammographic texture feature area on the display screen of the output unit 14, or controlling the output unit 14
  • the printer prints a breast screening image labeled with a characteristic area of the mammary gland.
  • the breast image output module 104 displays the breast screening image indicating the characteristic area of the mammary gland on the display screen of the output unit 14, or controls the printer of the output unit 14 to print the texture marked with mammary gland.
  • Mammography screening images of the characteristic areas provide a basis for doctors to diagnose and screen for breast diseases.
  • the breast screening image automatic processing system and method provided by the present invention removes noise and gray layer layering processing on the breast screening image, and extracts the breast texture characteristic region from the processed breast screening image and in the breast screen.
  • the image of the mammary gland is marked in the image for reference by doctors for diagnosis and screening of breast diseases, thus assisting doctors to improve the efficiency and accuracy of detection and screening of breast diseases, and improve the social efficiency of breast screening. .
  • the breast screening image automatic processing system and method of the present invention adopts the above technical solutions, and achieves the following technical effects: removing noise and grayscale layering by screening the breast image
  • the breast texture feature area is extracted from the processed breast screening image and the breast texture feature area is marked in the breast screening image for reference by the doctor for diagnosis and screening of the breast disease, thereby assisting the doctor to improve the mammary gland
  • the efficiency and accuracy of disease detection and screening have improved the social efficiency of breast screening.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Image Processing (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

一种乳腺筛查图像自动处理系统及方法，应用于医疗终端设备中，该方法包括步骤：通过输入单元接收用户输入的乳腺筛查编号（S21）；根据用户的乳腺筛查编号从乳腺图像数据库中获取该用户的乳腺筛查图像（S22）；将乳腺筛查图像进行无失真去除噪声滤波处理得到无失真的乳腺筛查图像（S23）；将无失真的乳腺筛查图像进行灰度分层处理得到灰度分层后的乳腺筛查图像（S24）；比较灰度分层后的乳腺筛查图像与正常乳腺图像两者的纹理分布差异以从灰度分层后的乳腺筛查图像中提取乳腺纹理特征区域（S25）；在灰度分层后的乳腺筛查图像中标示出乳腺纹理特征区域（S26）；通过输出单元输出标示有乳腺纹理特征区域的乳腺筛查图像以供医生对乳腺进行诊断与筛查参考（S27）。本方法和系统能够自动在乳腺筛查图像中标示出乳腺纹理特征区域以供医生参考，从而辅助医生提高对乳腺疾病进行检测与筛查的效率及准确性。

Description

？ L腺筛査图像自动处理系统及方法 技术领域
[0001] 本发明涉及图像处理与识别技术领域， 尤其涉及一种乳腺筛査图像自动处理系 统及方法。
背景技术
[0002] 乳腺疾病是女性的多发病症之一， 特别是随着现代都市生活节奏的加快， 女性 面对的生活压力、 工作压力加重， 乳腺疾病的发病率在逐年升高， 严重的乳腺 疾病可能导致乳腺癌。 据不完全调査统计显示， 乳腺疾病的发病率和死亡率也 都成上升趋势。 其实， 大部分乳腺癌患者的死亡原因就是因为没有及吋发现， 错过了最佳治疗吋机， 因此， 对广大女性进行乳腺普査、 乳腺疾病早发现、 早 期诊断对于女性健康是十分必要的。
[0003] 目前， 利用乳腺筛査图像进行辅助诊断已经成为被广泛采用的筛査和诊断早期 乳腺癌的重要方法。 目前， 利用计算机辅助检测方法对乳腺筛査图像进行乳腺 检测与筛査吋， 通常需要医生在乳腺图像上手动画出乳腺癌可疑区域 （ROI) 的 大致范围， 乳腺筛査与检测效率不高， 不适合大量乳腺样本的普査情况。 在女 性乳腺癌的体检和普査活动中， 由于乳腺图像数量很多， 医生直接对每幅乳腺 图像手动画出 ROI进行检测难以保证效率及准确性， 从而容易造成漏诊和误诊的 情况发生。
技术问题
[0004] 本发明的主要目的在于提供一种乳腺筛査图像自动处理系统及方法， 旨在解决 现有计算机辅助检测方式需要医生手动画出乳腺病变可疑区域而造成效率及准 确性不高的问题。
问题的解决方案
技术解决方案
[0005] 为实现上述目的， 本发明提供了一种乳腺筛査图像自动处理系统， 应用于医疗 终端设备中， 该医疗终端设备通过通信网络连接至医疗信息系统平台。 所述乳 腺筛査图像自动处理系统包括： 乳腺图像获取模块， 用于根据用户的乳腺筛査 编号从所述医疗信息系统平台的乳腺图像数据库中获取用户的乳腺筛査图像； 乳腺图像处理模块， 用于将所述乳腺筛査图像进行无失真去除噪声滤波处理得 到无失真的乳腺筛査图像， 以及将无失真的乳腺筛査图像进行灰度分层处理得 到灰度分层后的乳腺筛査图像； 纹理特征提取模块， 用于比较灰度分层后的乳 腺筛査图像与正常乳腺图像两者的纹理分布差异以从所述乳腺筛査图像中提取 乳腺纹理特征区域； 乳腺图像输出模块， 用于在灰度分层后的乳腺筛査图像中 标示出所述乳腺纹理特征区域， 以及通过所述医疗终端设备的输出单元输出所 述标示有乳腺纹理特征区域的乳腺筛査图像。
[0006] 优选的， 所述乳腺图像获取模块还用于通过所述医疗终端设备的输入单元接收 用户输入的乳腺筛査编号。
[0007] 优选的， 所述乳腺图像输出模块还用于将所述标示有乳腺纹理特征区域的乳腺 筛査图像显示在所述输出单元的显示屏上， 以及控制所述输出单元的打印机打 印所述标示有乳腺纹理特征区域的乳腺筛査图像。
[0008] 优选的， 所述乳腺图像数据库存储有不同用户在乳腺体检和普査吋产生的乳腺 筛査图像， 所述正常乳腺图像存储在所述医疗终端设备的存储单元中， 所述纹 理分布差异包括乳腺的组织结构差异、 尺寸大小差异及外形轮廓差异。
[0009] 优选的， 所述灰度分层处理包括将所述乳腺筛査图像按灰度分割成不同的区域 并对每个区域进行色彩赋值处理。
[0010] 为实现上述目的， 本发明还提供了一种乳腺筛査图像自动处理方法， 应用于医 疗终端设备中， 该医疗终端设备通过通信网络连接至医疗信息系统平台， 所述 乳腺筛査图像自动处理方法包括步骤： 根据用户的乳腺筛査编号从所述医疗信 息系统平台的乳腺图像数据库中获取用户的乳腺筛査图像； 将所述乳腺筛査图 像进行无失真去除噪声滤波处理得到无失真的乳腺筛査图像； 将无失真的乳腺 筛査图像进行灰度分层处理得到灰度分层后的乳腺筛査图像； 比较灰度分层后 的乳腺筛査图像与正常乳腺图像两者的纹理分布差异以从所述乳腺筛査图像中 提取乳腺纹理特征区域； 在灰度分层后的乳腺筛査图像中标示出所述乳腺纹理 特征区域； 通过所述医疗终端设备的输出单元输出所述标示有乳腺纹理特征区 域的乳腺筛査图像。
[0011] 优选的， 所述乳腺筛査图像自动处理方法还包括步骤： 通过所述医疗终端设备 的输入单元接收用户输入的乳腺筛査编号。
[0012] 优选的， 所述乳腺筛査图像自动处理方法还包括步骤： 将所述标示有乳腺纹理 特征区域的乳腺筛査图像显示在所述输出单元的显示屏上； 控制所述输出单元 的打印机打印所述标示有乳腺纹理特征区域的乳腺筛査图像。
[0013] 优选的， 所述乳腺图像数据库存储有不同用户在乳腺体检和普査吋产生的乳腺 筛査图像， 所述正常乳腺图像存储在所述医疗终端设备的存储单元中， 所述纹 理分布差异包括乳腺的组织结构差异、 尺寸大小差异及外形轮廓差异。
[0014] 优选的， 所述灰度分层处理包括将所述乳腺筛査图像按灰度分割成不同的区域 并对每个区域进行色彩赋值处理。
发明的有益效果
有益效果
[0015] 相较于现有技术， 本发明所述乳腺筛査图像自动处理系统及方法采用上述技术 方案， 达到了如下技术效果： 通过对乳腺筛査图像进行去除噪音及灰度分层处 理， 从处理后的乳腺筛査图像中提取乳腺纹理特征区域并在乳腺筛査图像中标 示出乳腺纹理特征区域， 以供医生对乳腺疾病进行诊断与筛査提供参考， 从而 辅助医生提高对乳腺疾病进行检测与筛査的效率及准确性， 提高了乳腺筛査的 社会效率。
对附图的简要说明
附图说明
[0016] 图 1是本发明乳腺筛査图像自动处理系统优选实施例的应用环境示意图；
[0017] 图 2是本发明乳腺筛査图像自动处理方法优选实施例的流程图。
[0018] 本发明目的实现、 功能特点及优点将结合实施例， 参照附图做进一步说明。
实施该发明的最佳实施例
本发明的最佳实施方式
[0019] 为更进一步阐述本发明为达成上述目的所采取的技术手段及功效， 以下结合附 图及较佳实施例， 对本发明的具体实施方式、 结构、 特征及其功效进行详细说 明。 应当理解， 此处所描述的具体实施例仅仅用以解释本发明， 并不用于限定 本发明。
[0020] 参照图 1所示， 图 1是本发明乳腺筛査图像自动处理系统优选实施例的应用环境 示意图。 在本实施例中， 所述乳腺筛査图像自动处理系统 10安装并运行于医疗 终端设备 1中。 所述医疗终端设备 1通过通信网络 3与医疗信息系统平台 2建立通 信连接。 所述医疗终端设备 1可以设置在大型医院或社区医疗工作站等医疗检査 机构内， 该医疗终端设备 1可以为一种计算机、 云服务器等具有数据处理和通信 功能的计算装置。 所述医疗信息系统平台 2可以是一种云平台或云平台中的一台 服务器， 可以设置在医院或社区医疗工作站等医疗检査机构内。 该医疗信息系 统平台 2包括乳腺图像数据库 20， 该乳腺图像数据库 20存储有女性乳腺体检和普 査吋的乳腺筛査图像。 所述通信网络 3可以是一种包括局域网、 广域网的网际网 络， 或者是一种包括 GSM、 GPRS、 CDMA的无线传输网络。
[0021] 在本实施例中， 所述医疗终端设备 1包括， 但不仅限于， 乳腺筛査图像自动处 理系统 10、 输入单元 11、 存储单元 12、 处理单元 13以及输出单元 14。 所述输入 单元 11、 存储单元 12和输出单元 14均通过数据总线连接至处理单元 13， 并能通 过处理单元 13与所述乳腺筛査图像自动处理系统 10进行信息交互。
[0022] 所述输入单元 11可以为键盘或手写触摸屏等输入设备， 用于供用户输入用户信 息， 例如， 可以在输入单元 11上输入用户的乳腺筛査编号或者用户名等信息。 所述存储单元 12可以为一种只读存储单元 ROM, 电可擦写存储单元 EEPROM或 快闪存储单元 FLASH等存储器。 所述处理单元 13可以为一种中央处理器 （CPU ) 、 微处理器、 微控制器 （MCU) 、 数据处理芯片、 或者具有数据处理功能的 信息处理单元。 所述输出单元 14可以为一种用于显示乳腺筛査报告的显示屏， 也可以为一种用于打印乳腺筛査报告的打印机。
[0023] 在本实施例中， 所述乳腺筛査图像自动处理系统 10包括， 但不局限于， 乳腺图 像获取模块 101、 乳腺图像处理模块 102、 纹理特征提取模块 103以及乳腺图像输 出模块 104。 本发明所称的模块是指一种能够被所述医疗终端设备 1的处理单元 1 3执行并且能够完成固定功能的一系列计算机程序指令段， 其存储在所述医疗终 端设备 1的存储单元 12中。
[0024] 所述乳腺图像获取模块 101用于通过输入单元 11接收用户输入的乳腺筛査编号 ， 以及根据用户的乳腺筛査编号从乳腺图像数据库 20中获取该用户的乳腺筛査 图像。 在本实施例中， 用户可以通过输入单元 11 (例如键盘或者手写触摸屏） 输入用户的乳腺筛査编号， 所述乳腺图像数据库 20存储不同用户进行乳腺体检 和普査吋产生的乳腺筛査图像。
[0025] 所述乳腺图像处理模块 102用于将所述乳腺筛査图像进行无失真去除噪声滤波 处理得到无失真的乳腺筛査图像。 在本实施例中， 所述乳腺图像处理模块 102采 用高斯滤波函数将所述乳腺筛査图像进行无失真去除噪声滤波处理清除所述乳 腺筛査图像的杂质得到无失真的乳腺筛査图像， 从而提高对乳腺疾病进行检测 与筛査的准确性。 所述高斯滤波函数为本领域技术人员对图像进行滤波处理的 惯用手段， 本实施不作具体描述。
[0026] 所述乳腺图像处理模块 102还用于将无失真的乳腺筛査图像进行灰度分层处理 得到灰度分层后的乳腺筛査图像， 以增强所述乳腺筛査图像的分层显示效果。 所述灰度分层处理也称谓密度分层处理， 包括将所述乳腺筛査图像按灰度分割 成不同的区域并对每个区域进行色彩赋值处理， 从而使所述乳腺筛査图像达到 分层显示的效果。 在本实施例中， 经过灰度分层处理后的乳腺筛査图像更能够 明显地显示出乳腺筛査图像上的纹理分布情况， 例如乳腺的组织结构、 尺寸大 小及外形轮廓等。
[0027] 所述纹理特征提取模块 103用于比较灰度分层后的乳腺筛査图像与正常乳腺图 像两者的纹理分布差异以从乳腺筛査图像中提取乳腺纹理特征区域。 在本实施 例中， 所述正常乳腺图像存在存储单元 12中， 作为与用户的乳腺筛査图像比较 的参考依据。 所述纹理分布差异包括乳腺的组织结构差异、 尺寸大小差异及外 形轮廓差异。 这种从乳腺筛査图像中提取乳腺纹理特征区域可以辅助医生提高 对乳腺疾病进行检测与筛査的效率及准确性。
[0028] 所述乳腺图像输出模块 104用于在灰度分层后的乳腺筛査图像中标示出乳腺纹 理特征区域， 以及通过输出单元 14输出标示有乳腺纹理特征区域的乳腺筛査图 像供医生对乳腺疾病进行诊断与筛査参考。 所述乳腺图像输出模块 104还用于将 所述标示有乳腺纹理特征区域的乳腺筛査图像显示在输出单元 14的显示屏上， 或者控制所述输出单元 14的打印机打印所述标示有乳腺纹理特征区域的乳腺筛 査图像， 以供医生对乳腺疾病进行诊断和治疗提供依据。
[0029] 本发明还提供了一种乳腺筛査图像自动处理方法， 应用于医疗终端设备中。 如 图 2所示， 图 2是本发明乳腺筛査图像自动处理方法优选实施例的流程图。 在本 实施例中， 参考图 1所示， 所述乳腺筛査图像自动处理方法包括如下步骤：
[0030] 步骤 S21， 通过输入单元接收用户输入的乳腺筛査编号； 在本实施例中， 乳腺 图像获取模块 101通过输入单元 11 (例如键盘或者手写触摸屏） 接收用户输入的 乳腺筛査编号。
[0031] 步骤 S22， 根据用户的乳腺筛査编号从乳腺图像数据库中获取该用户的乳腺筛 査图像； 具体地， 乳腺图像获取模块 101根据用户的乳腺筛査编号从乳腺图像数 据库 20中获取该用户的乳腺筛査图像。 在本实施例中， 用户可以通过输入单元 1 1输入用户的乳腺筛査编号， 所述乳腺图像数据库 20存储不同用户进行乳腺体检 和普査吋产生的乳腺筛査图像。
[0032] 步骤 S23， 将乳腺筛査图像进行无失真去除噪声滤波处理得到无失真的乳腺筛 査图像； 具体地， 乳腺图像处理模块 102将所述乳腺筛査图像进行无失真去除噪 声滤波处理得到无失真的乳腺筛査图像。 在本实施例中， 所述乳腺图像处理模 块 102采用高斯滤波函数将所述乳腺筛査图像进行无失真去除噪声滤波处理清除 所述乳腺筛査图像的杂质， 从而提高对乳腺疾病进行检测与筛査的准确性。
[0033] 步骤 S24， 将无失真的乳腺筛査图像进行灰度分层处理得到灰度分层后的乳腺 筛査图像； 具体地， 乳腺图像处理模块 102将所述无失真的乳腺筛査图像进行灰 度分层处理得到灰度分层后的乳腺筛査图像， 以增强所述乳腺筛査图像的分层 显示效果。 所述灰度分层处理也称谓密度分层处理， 包括将所述乳腺筛査图像 按灰度分割成不同的区域并对每个区域进行色彩赋值处理， 从而使所述乳腺灰 度图像达到分层显示的效果。 在本实施例中， 经过灰度分层处理后的乳腺筛査 图像更能够明显地显示出乳腺筛査图像上的纹理分布情况， 例如乳腺的组织结 构、 尺寸大小及外形轮廓等。
[0034] 步骤 S25， 比较灰度分层后的乳腺筛査图像与正常乳腺图像两者的纹理分布差 异以从乳腺筛査图像中提取乳腺纹理特征区域； 具体地， 纹理特征提取模块 103 将灰度分层后的乳腺筛査图像与正常乳腺图像两者的纹理分布差异进行比较以 从乳腺筛査图像中提取乳腺纹理特征区域。 在本实施例中， 所述纹理分布差异 包括乳腺的组织结构差异、 尺寸大小差异及外形轮廓差异。
[0035] 步骤 S26， 在灰度分层后的乳腺筛査图像中标示出乳腺纹理特征区域； 具体地 ， 乳腺图像输出模块 104在灰度分层后的乳腺筛査图像中标示出乳腺纹理特征区 域， 从而辅助医生提高对乳腺疾病进行检测与筛査的效率及准确性。
[0036] 步骤 S27， 通过输出单元输出标示有乳腺纹理特征区域的乳腺筛査图像供医生 对乳腺进行诊断与筛査参考； 具体地， 乳腺图像输出模块 104通过输出单元 14输 出标示有乳腺纹理特征区域的乳腺筛査图像供医生对乳腺疾病进行诊断与筛査 参考。
[0037] 在本实施例中， 在步骤 S27之后还可以包括步骤： 将所述标示有乳腺纹理特征 区域的乳腺筛査图像显示在输出单元 14的显示屏上， 或者控制所述输出单元 14 的打印机打印所述标示有乳腺纹理特征区域的乳腺筛査图像。 具体地， 所述乳 腺图像输出模块 104将所述标示有乳腺纹理特征区域的乳腺筛査图像显示在输出 单元 14的显示屏上， 或者控制所述输出单元 14的打印机打印所述标示有乳腺纹 理特征区域的乳腺筛査图像， 以供医生对乳腺疾病进行诊断和筛査提供依据。
[0038] 本发明提供的乳腺筛査图像自动处理系统及方法通过对乳腺筛査图像进行去除 噪音及灰度分层处理， 从处理后的乳腺筛査图像中提取乳腺纹理特征区域并在 乳腺筛査图像中标示出乳腺纹理特征区域， 以供医生对乳腺疾病进行诊断与筛 査提供参考， 从而辅助医生提高对乳腺疾病进行检测与筛査的效率及准确性， 提高了乳腺筛査的社会效率。
[0039] 以上仅为本发明的优选实施例， 并非因此限制本发明的专利范围， 凡是利用本 发明说明书及附图内容所作的等效结构或等效功能变换， 或直接或间接运用在 其他相关的技术领域， 均同理包括在本发明的专利保护范围内。
工业实用性
[0040] 相较于现有技术， 本发明所述乳腺筛査图像自动处理系统及方法采用上述技术 方案， 达到了如下技术效果： 通过对乳腺筛査图像进行去除噪音及灰度分层处 理， 从处理后的乳腺筛査图像中提取乳腺纹理特征区域并在乳腺筛査图像中标 示出乳腺纹理特征区域， 以供医生对乳腺疾病进行诊断与筛査提供参考， 从而 辅助医生提高对乳腺疾病进行检测与筛査的效率及准确性， 提高了乳腺筛査的 社会效率。

Claims

权利要求书
一种乳腺筛査图像自动处理系统， 应用于医疗终端设备中， 该医疗终 端设备通过通信网络连接至医疗信息系统平台， 其特征在于， 所述乳 腺筛査图像自动处理系统包括： 乳腺图像获取模块， 用于根据用户的 乳腺筛査编号从所述医疗信息系统平台的乳腺图像数据库中获取用户 的乳腺筛査图像； 乳腺图像处理模块， 用于将所述乳腺筛査图像进行 无失真去除噪声滤波处理得到无失真的乳腺筛査图像， 以及将无失真 的乳腺筛査图像进行灰度分层处理得到灰度分层后的乳腺筛査图像； 纹理特征提取模块， 用于比较灰度分层后的乳腺筛査图像与正常乳腺 图像两者的纹理分布差异以从所述乳腺筛査图像中提取乳腺纹理特征 区域； 乳腺图像输出模块， 用于在灰度分层后的乳腺筛査图像中标示 出所述乳腺纹理特征区域， 以及通过所述医疗终端设备的输出单元输 出所述标示有乳腺纹理特征区域的乳腺筛査图像。
如权利要求 1所述的乳腺筛査图像自动处理系统， 其特征在于， 所述 乳腺图像获取模块还用于通过所述医疗终端设备的输入单元接收用户 输入的乳腺筛査编号。
如权利要求 1所述的乳腺筛査图像自动处理系统， 其特征在于， 所述 乳腺图像输出模块还用于将所述标示有乳腺纹理特征区域的乳腺筛査 图像显示在所述输出单元的显示屏上， 以及控制所述输出单元的打印 机打印所述标示有乳腺纹理特征区域的乳腺筛査图像。
如权利要求 1所述的乳腺筛査图像自动处理系统， 其特征在于， 所述 乳腺图像数据库存储有不同用户在乳腺体检和普査吋产生的乳腺筛査 图像， 所述正常乳腺图像存储在所述医疗终端设备的存储单元中， 所 述纹理分布差异包括乳腺的组织结构差异、 尺寸大小差异及外形轮廓 差异。
如权利要求 1至 4任一项所述的乳腺筛査图像自动处理系统， 其特征在 于， 所述灰度分层处理包括将所述乳腺筛査图像按灰度分割成不同的 区域并对每个区域进行色彩赋值处理。 一种乳腺筛査图像自动处理方法， 应用于医疗终端设备中， 该医疗终 端设备通过通信网络连接至医疗信息系统平台， 其特征在于， 所述乳 腺筛査图像自动处理方法包括步骤： 根据用户的乳腺筛査编号从所述 医疗信息系统平台的乳腺图像数据库中获取用户的乳腺筛査图像； 将 所述乳腺筛査图像进行无失真去除噪声滤波处理得到无失真的乳腺筛 査图像； 将无失真的乳腺筛査图像进行灰度分层处理得到灰度分层后 的乳腺筛査图像； 比较灰度分层后的乳腺筛査图像与正常乳腺图像两 者的纹理分布差异以从所述乳腺筛査图像中提取乳腺纹理特征区域； 在灰度分层后的乳腺筛査图像中标示出所述乳腺纹理特征区域； 通过 所述医疗终端设备的输出单元输出所述标示有乳腺纹理特征区域的乳 腺筛査图像。
如权利要求 6所述的乳腺筛査图像自动处理方法， 其特征在于， 该方 法还包括步骤： 通过所述医疗终端设备的输入单元接收用户输入的乳 腺筛査编号。
如权利要求 6所述的乳腺筛査图像自动处理方法， 其特征在于， 该方 法还包括步骤： 将所述标示有乳腺纹理特征区域的乳腺筛査图像显示 在所述输出单元的显示屏上； 控制所述输出单元的打印机打印所述 标示有乳腺纹理特征区域的乳腺筛査图像。
如权利要求 6所述的乳腺筛査图像自动处理方法， 其特征在于， 所述 乳腺图像数据库存储有不同用户在乳腺体检和普査吋产生的乳腺筛査 图像， 所述正常乳腺图像存储在所述医疗终端设备的存储单元中， 所 述纹理分布差异包括乳腺的组织结构差异、 尺寸大小差异及外形轮廓 差异。
如权利要求 6至 9任一项所述的乳腺筛査图像自动处理方法， 其特征在 于， 所述灰度分层处理包括将所述乳腺筛査图像按灰度分割成不同的 区域并对每个区域进行色彩赋值处理。
PCT/CN2016/109736 2016-05-13 2016-12-13 乳腺筛查图像自动处理系统及方法 Ceased WO2017193581A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610318856.2 2016-05-13
CN201610318856.2A CN105975785A (zh) 2016-05-13 2016-05-13 乳腺筛查图像自动处理系统及方法

Publications (1)

Publication Number Publication Date
WO2017193581A1 true WO2017193581A1 (zh) 2017-11-16

Family

ID=56992641

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/109736 Ceased WO2017193581A1 (zh) 2016-05-13 2016-12-13 乳腺筛查图像自动处理系统及方法

Country Status (2)

Country Link
CN (1) CN105975785A (zh)
WO (1) WO2017193581A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119833106A (zh) * 2024-12-10 2025-04-15 中国人民解放军空军军医大学 一种基于智能分析的乳腺磁共振图像的处理诊断方法

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105975785A (zh) * 2016-05-13 2016-09-28 深圳市前海安测信息技术有限公司 乳腺筛查图像自动处理系统及方法
CN107049248A (zh) * 2017-03-25 2017-08-18 深圳市前海安测信息技术有限公司 基于医疗云平台的乳腺筛查影像分析系统及方法
CN109363699B (zh) * 2018-10-16 2022-07-12 杭州依图医疗技术有限公司 一种乳腺影像病灶识别的方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080152204A1 (en) * 2006-12-22 2008-06-26 Zhimin Huo Enhanced display of medical images
CN101404062A (zh) * 2008-11-14 2009-04-08 南京大学 一种基于决策树的数字乳腺图像自动筛查方法
CN103020466A (zh) * 2012-12-26 2013-04-03 海信集团有限公司 一种计算机诊断方法及电子设备
CN105975785A (zh) * 2016-05-13 2016-09-28 深圳市前海安测信息技术有限公司 乳腺筛查图像自动处理系统及方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080152204A1 (en) * 2006-12-22 2008-06-26 Zhimin Huo Enhanced display of medical images
CN101404062A (zh) * 2008-11-14 2009-04-08 南京大学 一种基于决策树的数字乳腺图像自动筛查方法
CN103020466A (zh) * 2012-12-26 2013-04-03 海信集团有限公司 一种计算机诊断方法及电子设备
CN105975785A (zh) * 2016-05-13 2016-09-28 深圳市前海安测信息技术有限公司 乳腺筛查图像自动处理系统及方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119833106A (zh) * 2024-12-10 2025-04-15 中国人民解放军空军军医大学 一种基于智能分析的乳腺磁共振图像的处理诊断方法

Also Published As

Publication number Publication date
CN105975785A (zh) 2016-09-28

Similar Documents

Publication Publication Date Title
TWI667996B (zh) 乳房腫瘤輔助檢測模型及乳房腫瘤輔助檢測系統
Pranata et al. Deep learning and SURF for automated classification and detection of calcaneus fractures in CT images
JP6231885B2 (ja) 頸部を検査する医療機器
CN108464840A (zh) 一种乳腺肿块自动检测方法及系统
CN108241865B (zh) 基于超声图像的多尺度多子图肝纤维化多级量化分期方法
CN105513077A (zh) 一种用于糖尿病性视网膜病变筛查的系统
CN111785363A (zh) 一种基于ai引导的慢性病辅助诊断系统
CN104299242B (zh) 基于ngc‑acm的荧光造影眼底图像提取方法
WO2019047365A1 (zh) 基于医疗云平台的影像大数据分析系统及方法
Kim et al. Asbestosis diagnosis algorithm combining the lung segmentation method and deep learning model in computed tomography image
CN106408024A (zh) 针对dr片的肺叶轮廓提取方法
CN110390678B (zh) 一种结直肠癌ihc染色图像的组织类型分割方法
CN118397015A (zh) 一种基于深度学习的超声图像斑块分割方法、系统及设备
Kajla et al. Analysis of x-ray images with image processing techniques: A review
WO2018176717A1 (zh) 基于医疗云平台的乳腺筛查影像分析系统及方法
CN108038875A (zh) 一种肺部超声图像识别方法和装置
Kumar et al. A Novel Approach for Breast Cancer Detection by Mammograms
JP2006325638A (ja) 異常陰影候補の検出方法及び医用画像処理システム
CN101208042A (zh) 异常阴影候选检测方法、异常阴影候选检测装置
CN117457142B (zh) 用于报告生成的医学影像处理系统及方法
CN113688942A (zh) 基于深度学习的头侧片腺样体图像自动评估方法和装置
Lange et al. Computer-aided-diagnosis (CAD) for colposcopy
CN113114952B (zh) 一种评价乳腺随访图像质量的方法、系统及装置
CN102521517A (zh) 乳腺肿瘤判别分析系统及方法
CN111544021B (zh) 一种基于胸部ct的covid-19疾病进展评价方法及装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16901536

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC, EPO FORM 1205A DATED 10.04.19

122 Ep: pct application non-entry in european phase

Ref document number: 16901536

Country of ref document: EP

Kind code of ref document: A1