WO2018000359A1 - 一种增强超声造影图像的方法、系统及超声造影成像设备 - Google Patents

一种增强超声造影图像的方法、系统及超声造影成像设备 Download PDF

Info

Publication number
WO2018000359A1
WO2018000359A1 PCT/CN2016/087979 CN2016087979W WO2018000359A1 WO 2018000359 A1 WO2018000359 A1 WO 2018000359A1 CN 2016087979 W CN2016087979 W CN 2016087979W WO 2018000359 A1 WO2018000359 A1 WO 2018000359A1
Authority
WO
WIPO (PCT)
Prior art keywords
contrast
information
threshold
tissue
image
Prior art date
Application number
PCT/CN2016/087979
Other languages
English (en)
French (fr)
Inventor
储霞
唐杰
桑茂栋
朱磊
张明博
Original Assignee
北京深迈瑞医疗电子技术研究院有限公司
中国人民解放军总医院
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 北京深迈瑞医疗电子技术研究院有限公司, 中国人民解放军总医院 filed Critical 北京深迈瑞医疗电子技术研究院有限公司
Priority to PCT/CN2016/087979 priority Critical patent/WO2018000359A1/zh
Priority to CN201680039006.7A priority patent/CN108135566B/zh
Publication of WO2018000359A1 publication Critical patent/WO2018000359A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/90Dynamic range modification of images or parts thereof
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10132Ultrasound image

Definitions

  • the present invention relates to an ultrasound contrast imaging apparatus, and in particular to a method and system for enhancing an ultrasound contrast image.
  • contrast-enhanced ultrasound imaging can enhance the intensity of ultrasound echo signals and display small blood vessels that cannot be seen by conventional ultrasound.
  • ultrasound imaging is becoming more and more It is widely used clinically, and it is used to examine organs more and more, and its position in clinical diagnosis is getting higher and higher.
  • Ultrasound contrast imaging is an ultrasound imaging after injecting a contrast agent, which is an enhanced ultrasound imaging.
  • the contrast agent is an enhancer that enhances the intensity of the ultrasonic echo signal.
  • Ultrasound contrast agents come in a variety of applications. They are hollow sphere-like microvesicles that are close in size to red blood cells. Microbubbles can reach all blood vessels in the body due to their size and red blood cells. Because of the hollow sphere, the acoustic impedance difference of the microbubbles is large, and the intensity of the reflected echo signal is greatly increased, so that tiny blood vessels or capillaries can be visualized. In most cases, the contrast agent microbubbles are injected through the vein and distributed to the whole body with the blood flow. The contrast microbubbles and red blood cells behave similarly, and the microbubbles are detected or tracked, which can reflect the perfusion of normal and abnormal tissues.
  • the ultrasound contrast imaging technique also It can be called a detection microbubble signal technology.
  • CTR contrast to tissue ratio
  • SNR can also be understood as sensitivity or sensitivity, which is the degree of detection of microbubbles
  • CTR is a characteristic characteristic of contrast images, which is the ratio of contrast agent signal to tissue signal intensity in contrast images.
  • the contrast image is only developed for microbubbles, and no contrast agent is injected.
  • the contrast image has no content, but in reality, due to the destructive limitation of the system, even if there is no contrast agent, there is a small amount of development of the contrast image.
  • tissue residue these tissue residues also exist after contrast agent microbubble perfusion, mixed with microbubble signals, affecting the user's recognition of contrast microbubbles.
  • the inter-turn resolution and spatial resolution are the same as those for conventional ultrasound images, and the higher the resolution, the better.
  • the key to improving the quality of contrast images is to improve SNR and CTR.
  • the SNR is determined by the sensitivity of the probe and the system platform.
  • the high SNR means that smaller microbubbles can be detected.
  • the technical problem to be solved by the present invention is to provide another scheme for enhancing contrast information, thereby improving the CTR of the contrast image.
  • an embodiment provides a method for enhancing an ultrasound contrast image, comprising:
  • calculating a nonlinear parameter comparing the contrast information and the tissue information to obtain a nonlinear parameter distribution map of the one-frame contrast information, wherein the nonlinear parameter is a parameter that measures the magnitude of the nonlinear effect generated by the ultrasonic wave propagating in the medium;
  • the first threshold is divided, and the nonlinear parameter profile is segmented by using the first threshold, the region smaller than the first threshold is a tissue residual region, and the region larger than the first threshold is a contrast agent region, and the first threshold is non- The majority of the points in the linear parameter distribution map that are smaller than the first threshold are corresponding to the tissue residual information in the contrast information frame;
  • a second threshold segmentation wherein the nonlinear parameter profile is segmented by using a second threshold, wherein the region smaller than the second threshold is a tissue residual region, and the region greater than the second threshold is a contrast agent region, and the second threshold is non- The majority of the regions in the linear parameter distribution map that are larger than the second threshold are corresponding to the contrast agent information in the contrast information frame, and the second threshold is greater than the first threshold;
  • Image fusion combining the first image data and the second image data to form enhanced contrast image data.
  • an ultrasound contrast image enhancement system including:
  • an information acquiring unit configured to obtain nonlinear contrast information and linear tissue information according to the ultrasonic echo signal during the contrast imaging process
  • a nonlinear parameter calculation unit configured to compare the contrast information and the tissue information to obtain a nonlinear parameter distribution map of one frame of contrast information, wherein the nonlinear parameter is a measure of a nonlinear effect caused by the propagation of the ultrasonic wave in the medium.
  • a first threshold dividing unit configured to divide the nonlinear parameter distribution map by using a first threshold, wherein a region smaller than the first threshold is a tissue residual region, and a region larger than the first threshold is a contrast agent region, where the first The threshold value causes a majority of the points in the non-linear parameter distribution map that are smaller than the first threshold to belong to the tissue residual information in the corresponding information frame;
  • a first image forming unit configured to perform suppression processing on information belonging to the tissue residual region in the contrast information frame, perform enhancement processing on the information belonging to the contrast agent region in the contrast information frame, and form an image based on the processed result First image data;
  • a second threshold dividing unit configured to divide the nonlinear parameter distribution map by using a second threshold, where a region smaller than the second threshold is a tissue residual region, and a region larger than the second threshold is a contrast agent region, and the second The threshold value is such that a majority of the regions in the nonlinear parameter distribution map that are greater than the second threshold value in the contrast information frame belong to the contrast agent information, and the second threshold value is greater than the first threshold value;
  • a second image forming unit configured to suppress information belonging to a tissue residual region in the contrast information frame Processing, performing enhancement processing on the information belonging to the contrast agent region in the contrast information frame, and forming second image data based on the processed result;
  • an image fusion unit configured to fuse the first image data and the second image data to form an enhanced image
  • an ultrasound contrast imaging apparatus including:
  • a pre-signal processing module that processes the ultrasonic echo signal, and extracts a nonlinear component reflecting the microbubble information of the contrast agent from the ultrasonic echo signal and a linear component reflecting the anatomical feature of the tissue;
  • a first signal processing module configured to process a linear component
  • tissue image generating module configured to process an output of the first signal processing module into tissue image data
  • second signal processing module configured to process the nonlinear component
  • a contrast image generating module configured to process the output of the second signal processing module into the contrast image data
  • an ultrasound contrast image enhancement system for enhancing the contrast information the ultrasound contrast image enhancement system is as described above , to obtain enhanced contrast image data
  • a display module configured to visually display the tissue image data and the enhanced contrast image data.
  • the nonlinear parameter distribution map of the angiographic information frame is divided by a bipolar threshold, and the contrast agent region and the tissue residual region are separated for each division, and the contrast agent region is enhanced to treat the tissue residual region. Inhibiting the treatment, thereby achieving a large degree of enhancement to the region of the contrast agent, a large degree of suppression of the region which is absolutely residual tissue, and a degree of enhancement and suppression of the region between the contrast agent and the tissue residue is slightly weaker. Finally, the results of the two classification and weight adjustment are combined to enhance the contrast agent, inhibit the contrast agent, and achieve conservative contrast image enhancement for the contrast agent and tissue residue. effect.
  • FIG. 1 is a first structural schematic view of an ultrasound contrast imaging apparatus
  • FIG. 2 is a second schematic structural view of an ultrasound contrast imaging apparatus
  • FIG. 3 is a third structural schematic diagram of an ultrasound contrast imaging apparatus
  • FIG. 4 is a schematic structural diagram of an ultrasound contrast image enhancement system in an embodiment
  • FIG. 5 is a flow chart of an ultrasound contrast image enhancement system in an embodiment
  • FIG. 6 is a non-linear parametric distribution diagram represented by a histogram method in an embodiment
  • the ultrasound contrast imaging apparatus includes a probe 10, a front signal processing module 13, a first signal processing module 16, a second signal processing module 17, a tissue image generating module 18, and a contrast image generating module.
  • the ultrasound contrast image enhancement system 20 and the display module 100 are The ultrasound contrast image enhancement system 20 and the display module 100.
  • the probe 10 is configured to emit ultrasonic waves to the human tissue 11 and receive ultrasonic echoes reflected back from the tissue 11, which may be present in the signal receiving module 12.
  • the pre-signal processing module 13 is configured to process the ultrasonic echo signals to extract useful information.
  • the pre-signal processing module 13 mainly includes an amplifier, a beam synthesizer, and a filter to extract more useful, different-component signals. In an ultrasound contrast imaging apparatus, the pre-signal processing module 13 needs to extract two different signals.
  • One is a linear component 14, reflecting the tissue anatomical features; the other is a nonlinear component 15, reflecting the contrast agent microbubble information.
  • the first signal processing module 16 is for processing linear components
  • the second signal processing module 17 is for processing non-linear components.
  • the processing of the first signal processing module 16 and the second signal processing module 17 are similar, mainly including demodulation, filtering, downsampling, entropy, Log, dynamic range conversion, etc., but the selection of parameters is different.
  • the tissue image generation module 18 is configured to process the output of the first signal processing module 16 into tissue image data; the contrast image generation module 19 is configured to process the output of the second signal processing module 17 into contrast image data.
  • the ultrasound contrast image enhancement system 20 is configured to enhance the contrast information to obtain enhanced contrast image data; the display module 100 is configured to visually display the tissue image data and the enhanced contrast image data. Most of the contrast images are displayed next to a tissue image that is used to locate the lesion.
  • angiography usually emits multiple pulse waves of different phases or amplitudes at the same position.
  • Tissue and contrast images are imaged with parts of the multiple transmitted waveforms, so tissue images and contrast The image shows the same engraved information at the same position, and the tissue image shows the echo.
  • the linear characteristic, the contrast image shows the nonlinear characteristics in the echo.
  • the linear information of the tissue and the nonlinear information of the contrast agent microbubble are first obtained, and then the nonlinear information and the linear information are compared to obtain a nonlinear parameter value, and then the contrast image is subjected to two according to the nonlinear parameter value.
  • the sub-category is divided into two types: tissue residue and contrast agent. The two types are enhanced or suppressed according to different weights. The weight coefficients are different according to the classification. Finally, the results of the two classifications and weight adjustments are merged. .
  • the ultrasound contrast image enhancement system 20 may perform enhancement processing on the contrast information of the image domain output by the tissue image generation module 18 and the contrast image generation module 19, as shown in FIG. 1, the ultrasound contrast image enhancement system.
  • the input end of the 18 is respectively connected to the tissue image generating module 18 and the contrast image generating module 19, and the output end is connected to the display module 100, and the contrast image is obtained according to the tissue image data and the contrast image data output by the tissue image generating module 18 and the contrast image generating module 19.
  • the data is subjected to enhancement processing, and the enhancement result is output to the display module 100.
  • the ultrasound contrast image enhancement system 20 can also enhance the contrast information of the radio frequency domain output by the front signal processing module 13. As shown in FIG.
  • the ultrasound contrast image enhancement system 20 is connected between the front signal processing module 13 and the second signal processing module 17, and receives the nonlinear component 15 and the linear component 14 output from the front signal processing module 13, according to the nonlinear component.
  • the linear component 14 and the linear component 14 perform enhancement processing on the nonlinear component, and output the enhancement result to the second signal processing module 17.
  • the ultrasound contrast image enhancement system 18 can also enhance the contrast information at any of the second signal processing modules.
  • the input end of the ultrasound contrast image enhancement system 20 is connected to any of the first signal processing module 16 and the second signal processing module 17, for example, after demodulation, after enveloping, after Log transformation, etc.
  • the output of the ultrasound contrast image enhancement system 20 is coupled to the contrast image generation module 19.
  • the structure diagram of the ultrasound contrast image enhancement system 20 is as shown in FIG. 4, and includes an information acquisition unit 201, a nonlinear parameter calculation unit 202, a first threshold division unit 203, a first image forming unit 204, and a second threshold division unit. 205. Second image forming unit 206 and image fusion unit 207.
  • the information acquisition unit 201 is configured to obtain nonlinear angiographic information and linear tissue information according to the ultrasonic echo signal in the angiographic imaging process.
  • the angiographic information and the linear tissue information are two pieces of information extracted from the same ultrasonic echo signal, the nonlinear information is angiographic information, the linear information is tissue residual information; the nonlinear parameter calculation unit 202 is used to compare the same engraving
  • the angiographic information and the tissue information obtain a nonlinear parameter distribution map of the angiographic information, and the nonlinear parametric distribution map is called a nonlinear parametric distribution map of the angiographic information frame, and the nonlinear parametric is to measure the propagation of the ultrasonic wave in the medium.
  • the parameter of the magnitude of the generated nonlinear effect; in this embodiment, the nonlinear parametric profile is calculated from the ratio or difference of one frame of nonlinear contrast information and one frame of linear tissue information of the same engraving.
  • the first threshold segmentation unit 203 is configured to segment the nonlinear parameter profile by using the first threshold, the region smaller than the first threshold is a tissue residual region, and the region larger than the first threshold is a contrast agent region, and the first threshold is non- The majority of the regions in the linear parameter distribution map that are smaller than the first threshold in the contrast information frame belong to the tissue residual information; the first image forming unit 204 is configured to perform the information belonging to the tissue residual region in the contrast information frame.
  • second threshold dividing unit 205 is configured to segment the nonlinear parameter distribution map by using the second threshold
  • the region smaller than the second threshold is a tissue residual region, and the region larger than the second threshold is a contrast agent region, and the second threshold is such that the majority of the regions in the nonlinear parameter distribution map that are larger than the second threshold are in the contrast information frame.
  • the corresponding point in the middle belongs to the contrast agent information, the second threshold is greater than the first threshold; the second image is formed
  • the element 206 is configured to perform suppression processing on the information belonging to the tissue residual area in the contrast information frame, perform enhancement processing on the information belonging to the contrast medium area in the contrast information frame, and form second image data based on the processed result; image fusion
  • the unit 207 is configured to fuse the first image data and the second image data to form enhanced contrast image data.
  • the ultrasound contrast image enhancement system 20 further includes a noise recognition unit 208 for determining noise in the contrast information before the nonlinear parameter calculation unit calculates the nonlinear parameter, and will frame one frame.
  • the contrast information is divided into a noise signal and a noise-free non-noise contrast signal; the nonlinear parameter calculation unit 202 calculates a nonlinear parameter using a non-noise contrast signal in calculating a nonlinear parameter.
  • FIG. 5 The specific processing flow of the ultrasound contrast image enhancement system is shown in FIG. 5, and includes the following steps:
  • the purpose of denoising is to restore the size of the signal itself, avoiding the effects of TGC (Signal Gain Compensation) and noise on subsequent processing.
  • ATGC analog time gain compensation
  • DTGC digital time gain compensation
  • the noise of snoring is approximately white noise.
  • the noise mean curve is a curve that increases with depth. In order to get the true size of the signal, subtracting the noise mean curve is not only denoising, but also eliminates the effects of ATGC and DTGC on the signal.
  • Steps 21 through 24 are to achieve denoising. Denoising firstly obtains a noise signal.
  • the method of obtaining the noise signal is to suspend the transmission. Steps 21 and 22 are performed before the ultrasound contrast image enhancement system 20, and a frame of nonlinear contrast noise signal and a frame linear organization are buffered.
  • the noise signal that is, the control probe does not emit, and then obtains a frame of nonlinear contrast noise signal and a frame of linear tissue noise signal according to the echo.
  • the specific step of processing is to calculate the mean value of each line of the noise signal of one frame to obtain the longitudinal mean value curve of the noise, and then pass the low-pass filter to obtain a smooth noise mean curve.
  • step 23 the noise average curve of the nonlinear contrast noise signal is subtracted from the contrast information input by the ultrasound contrast image enhancement system 20 to obtain a denoised contrast signal.
  • step 24 the noise average curve of the linear tissue noise signal is subtracted from the tissue information input by the ultrasound contrast image enhancement system 20 to obtain a denoised tissue signal.
  • Step 25 detecting noise in the image after denoising, and dividing the one-frame contrast image into the noise region 26 and the non-noise contrast agent region 27.
  • Noise detection can be divided by a threshold, and less than the threshold is noise. The threshold is determined empirically based on system characteristics.
  • the nonlinear parameter is an important parameter for measuring the magnitude of the nonlinear effect produced by the propagation of the ultrasonic wave in the medium.
  • the nonlinear parameter is the ratio or difference between the multiple coefficient and the linear coefficient in the Taylor equation of the medium state equation.
  • the linear coefficient is A
  • the quadratic coefficient For B the cubic term coefficient is C
  • the nonlinear parameter can be B/A, C/A or (B+C) /A, and the ratio of other combinations of C to A.
  • the nonlinear parameters are calculated by the same engraving The ratio or difference between the contrast signal and the tissue signal is derived.
  • each pixel value corresponds to a pixel point on the display interface, and the pixel value may belong to a signal generated by a contrast agent or may belong to a tissue residual signal.
  • each pixel has a corresponding non-linear parameter. If the pixel value is tissue residual information, the corresponding nonlinear parameter value is small. If the pixel value is contrast agent information, Then, the corresponding nonlinear parameter value is large. Therefore, by calculating the nonlinear parameter of each pixel value in the contrast information frame, the pixel value belonging to the tissue residual information and the pixel value belonging to the contrast agent information can be distinguished.
  • the nonlinear parameters of each pixel in the angiographic information frame can be obtained, and the nonlinear parameters can be obtained by counting
  • the nonlinear parameter distribution map of the angiographic information frame may be represented by a histogram method or a two-dimensional map.
  • FIG. 6 is a non-linear parametric distribution diagram represented by a histogram method, including a large peak and a small peak, and the abscissa of the histogram is the calculated nonlinear parameter value, and the unit is dB, and the ordinate indicates the contrast.
  • the number of points in the information frame that correspond to the non-linear parameter value It has been proved by experiments that the midpoint of the contrast information frame corresponding to most of the large peak region reflects the contrast agent microbubble information, and the midpoint of the contrast information frame corresponding to most of the small peak region reflects the tissue residual information.
  • the nonlinear parametric profile is a two-dimensional map, as shown in FIG. 7, which represents the nonlinear parameter values of each point in the contrast information frame in color, and the dark blue represents the noise.
  • the color of the nonlinear parameter is represented by red, yellow, green, and blue.
  • the nonlinear parameter value of the tissue residue is small, and the nonlinear acoustic parameter of the contrast agent is large, so it can be determined to be smaller than -40dB must be tissue residue, about -20dB must be contrast agent, where -40dB is the smaller threshold, -20dB is the larger threshold.
  • Figure 7 is only a specific example.
  • the nonlinear parameter distribution map is divided by the two-pole threshold, and each division is divided into a contrast agent region and a tissue residual region, and the contrast agent region is enhanced and displayed, and the residual region of the tissue is suppressed.
  • the threshold of the two poles is the minimum and maximum thresholds.
  • the nonlinear parameters are different, and the threshold is different.
  • the threshold is related to the destructiveness of the system.
  • the large threshold is greater than the cancellation value of the system.
  • the small threshold is smaller than the system.
  • the destructive value, the destructive value is simply the difference between the size of the two nonlinear components injected into the contrast agent and the contrast agent. details as follows:
  • the nonlinear parameter profile is segmented by using the first threshold.
  • the left straight line threshold is a first threshold, and the first threshold is such that the nonlinear parameter profile is smaller than the first threshold.
  • the vast majority of the points in the region of the contrast information belong to tissue residual information, so the first threshold is also referred to as the minimum threshold.
  • the first threshold is also referred to as the minimum threshold.
  • the left area smaller than the first threshold is the tissue residual area 32
  • the corresponding point in the tissue residual area 32 is considered to be the corresponding point in the contrast information frame. Belongs to organizational residual information.
  • the area on the right side larger than the first threshold is the contrast agent area 3 1, and the point in the contrast medium area 31 is considered to be information generated by the contrast agent in the corresponding point in the contrast information frame. This is less than the first threshold and almost all of the tissue residue, and greater than the first threshold may belong to the contrast agent.
  • the contrast image is divided into three different regions of noise, contrast agent and tissue residue.
  • Step 35 In order to achieve the purpose of enhancing the contrast agent and suppressing tissue residue, the information belonging to the tissue residual region in the contrast information frame is suppressed, and the information belonging to the contrast agent region in the contrast information frame is enhanced.
  • the above three regions are respectively multiplied or added with different weight coefficients P1, P2 and P3. If multiplied by the weight coefficient, the coefficient P2 of the contrast agent region is greater than 1, and the coefficient P 3 of the tissue residual region should be less than 1.
  • the noise area coefficient P1 can be equal to 1. If you want to suppress noise, P1 can also be less than 1. In the same way, P2 greater than 1, also increases the dynamic range of the contrast signal.
  • P3 ⁇ 0 ⁇ P 2 can enhance the intensity of the contrast signal and weaken the residual strength of the tissue.
  • P1 ⁇ 0, P1 ⁇ 0 also achieve the purpose of suppressing noise. Multiplying or adding different weight coefficients to the above three regions respectively forms a new image, which is called first image data.
  • the nonlinear parameter profile is segmented by using the second threshold.
  • the right straight line threshold is a second threshold, and the second threshold is greater than the second threshold in the nonlinear parameter distribution map.
  • the second threshold is also referred to as the maximal threshold, and the second threshold is greater than the first threshold.
  • the region larger than the second threshold is the contrast agent region 33, that is, the corresponding point in the contrast medium frame in the contrast medium region 33 is considered to belong to the information generated by the contrast agent.
  • the ⁇ is greater than the second threshold, almost all of the contrast agent, and less than the second threshold may belong to the tissue residue.
  • the contrast image is also divided into three different regions of noise, contrast agent and tissue residue.
  • Step 36 In order to achieve the purpose of enhancing the contrast agent and suppressing tissue residue, the information belonging to the tissue residual region in the contrast information frame is suppressed, and the information belonging to the contrast agent region in the contrast information frame is enhanced.
  • the three regions in step 30 are multiplied or added with different weight coefficients P1, P4 and P5, respectively.
  • the coefficient P4 of the contrast agent region is larger than 1, and the coefficient P5 of the tissue residual region should be less than 1, that is, P5 ⁇ 1 ⁇ P4, and the noise region coefficient PI is equal to or smaller than 1.
  • P5 ⁇ 0 ⁇ P4, P1 ⁇ 0 Multiplying or adding different weight coefficients to the above three regions respectively forms a new image called second image data.
  • the degree of enhancement of the pixel points which are absolutely belonging to the contrast agent area is large
  • the degree of suppression of the pixel points which are absolutely belonging to the tissue residual area is large
  • the area between the contrast agent and the tissue residual is enhanced.
  • the degree of inhibition is slightly weaker. If multiplied by the weighting factor, the parameters of enhancement and suppression satisfy P4>P2>1, P3 ⁇ P5 ⁇ 1; if the weighting coefficient is added, the parameters of enhancement and suppression satisfy P4>P2>0, P3 ⁇ P5 ⁇ 0.
  • the embodiment combines the two image data to complement each other, so that the contrast agent is enhanced, the contrast agent is suppressed, and the conservative treatment between the contrast agent and the tissue residue can be achieved.
  • the contrast image enhancement effect
  • the program is instructed to execute related hardware, and the program may be stored in a computer readable storage medium, and the storage medium may include: a read only memory, a random access memory, a magnetic disk or an optical disk, and the like.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • General Physics & Mathematics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Theoretical Computer Science (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)
  • Image Processing (AREA)

Abstract

一种增强超声造影图像的方法、系统和超声造影成像设备,根据从超声回波信号中获得的非线性的造影信息和线性的组织信息,计算非线性参量,得到非线性参量分布图,分别采用两极阈值对非线性参量分布图进行分割,每次划分都分出造影剂区域(31,33)和组织残留区域(32,34),对造影剂区域(31,33)进行增强处理,对组织残留区域(32,34)抑制处理,从而可实现对绝对属于造影剂区域(31,33)进行大程度的增强,对绝对属于组织残留的区域进行大程度的抑制,对介于造影剂和组织残留的区域增强和抑制的程度稍微弱些,最后再将两次分类和权重调整后的结果进行融合处理,使对造影剂的增强、对造影剂的抑制,对介于造影剂和组织残留的保守处理,都能达到比较理想的造影图像增强效果。

Description

说明书
发明名称:一种增强超声造影图像的方法、 系统及超声造影成像设 备
技术领域
[0001] 本发明涉及超声造影成像设备, 具体涉及增强超声造影图像的方法及系统。
[0002]
[0003] 背景技术
[0004] 超声造影成像与常规超声成像相比, 能增强超声回波信号强度, 显示常规超声 无法看见的细小血管, 随着临床对超声检査的要求越来越高, 超声造影成像越 来越多地被临床应用, 且用来检査器官也越来越广, 其在临床诊断中的地位越 来越高。
[0005] 超声造影成像是注入造影剂后再超声成像, 是一种增强超声成像, 造影剂是一 种能增强超声回波信号强度的增强剂。 超声造影剂的种类多样, 应用较多的是 一种大小和红细胞接近的空心球体样的微泡, 由于大小和红细胞接近所以微泡 能到达全身所有血管。 由于是空心球样所以微泡的声阻抗差大, 极大的增加反 射回波信号强度, 使微小血管或毛细血管能显像。 大多情况下造影剂微泡通过 静脉注入, 随血流分布到全身, 造影微泡与红细胞行为相似, 检测或者跟踪微 泡, 能反映出正常和异常组织的血流灌注情况, 超声造影成像技术也可以称之 为检测微泡信号技术。
[0006] 衡量造影图像的好坏, 一般会从以下几个指标考虑, SNR(signal noise
ratio) CTR(contrast to tissue ratio)、 吋间分辨力、 空间分辨力。 SNR也可以理解 为灵敏度或敏感性, 就是对微泡的检测程度; CTR是造影图像特有的特征, 是造 影图像中造影剂信号和组织信号强度之比。 理论上造影图像只是对微泡显影, 没有注入造影剂吋造影图像是没有内容的, 但是实际情况下由于系统的相消性 限制, 即使在没有注入造影剂吋造影图像也有少量显影, 该少量显影称为组织 残留, 这些组织残留在造影剂微泡灌注后也存在, 与微泡信号混在一起, 影响 用户对造影剂微泡的识别判断。 CTR越大, 组织残留越小, 微泡强度和组织残留 强度的差距越大, 纯造影剂信号的动态范围更大, 造影图像的层次更丰富, 微 泡的灌注和效果过程显示的更清楚, 图像的对比分辨力也更好。 吋间分辨力和 空间分辨力和常规超声图像的要求一样, 分辨力越高越好。
[0007] 随着超声造影的使用越来越广, 临床对造影图像的要求也越来越高, 不再是能 看见微泡显影, 而是希望能显示更多细节。 这就要求提高造影图像质量以能够 显示更多的细节。
[0008] 在超声造影图像中, 提高造影图像质量的关键是提高 SNR和 CTR。 SNR是由探 头灵敏度和系统平台决定的, SNR高, 意味着能检测出更微小的微泡, 而微泡越 小信号幅度越小, 有些小微泡的信号幅度和组织残留幅度相当甚至更小, 因此 , 即使 SNR很高, 但如果 CTR不够高, 这些小微泡反映出的信号将淹没在组织残 留信息中, 在造影图像中也观察不到。 因此提高造影图像质量最关键的是要尽 量提高 CTR。
[0009]
[0010] 发明内容
[0011] 本发明要解决的技术问题是提供另一种对造影信息进行增强的方案, 从而提高 造影图像的 CTR。
[0012] 根据第一方面, 一种实施例中提供一种增强超声造影图像的方法,包括:
[0013] 获取信息, 根据造影成像过程中的超声回波信号获得非线性的造影信息和线性 的组织信息;
[0014] 计算非线性参量, 比较所述造影信息和组织信息得到一帧造影信息的非线性参 量分布图, 所述非线性参量是衡量超声波在媒质中传播吋产生的非线性效应大 小的参量;
[0015] 第一阈值分割, 采用第一阈值对非线性参量分布图进行分割, 小于第一阈值的 区域为组织残留区域, 大于第一阈值的区域为造影剂区域, 所述第一阈值使得 非线性参量分布图中小于第一阈值的区域中绝大多数在该造影信息帧中对应的 点属于组织残留信息;
[0016] 形成第一图像, 对该造影信息帧中属于组织残留区域的信息进行抑制处理, 对 该造影信息帧中属于造影剂区域的信息进行增强处理, 并基于处理后的结果形 成第一图像数据;
[0017] 第二阈值分割, 采用第二阈值对非线性参量分布图进行分割, 小于第二阈值的 区域为组织残留区域, 大于第二阈值的区域为造影剂区域, 所述第二阈值使得 非线性参量分布图中大于第二阈值的区域中绝大多数在该造影信息帧中对应的 点属于造影剂信息, 所述第二阈值大于第一阈值;
[0018] 形成第二图像, 对该造影信息帧中属于组织残留区域的信息进行抑制处理, 对 该造影信息帧中属于造影剂区域的信息进行增强处理, 并基于处理后的结果形 成第二图像数据;
[0019] 图像融合, 将第一图像数据和第二图像数据进行融合, 形成增强后的造影图像 数据。
[0020] 根据第二方面, 一种实施例中提供一种超声造影图像增强系统,包括:
[0021] 信息获取单元, 用于根据造影成像过程中的超声回波信号获得非线性的造影信 息和线性的组织信息;
[0022] 非线性参量计算单元, 用于比较所述造影信息和组织信息得到一帧造影信息的 非线性参量分布图, 所述非线性参量是衡量超声波在媒质中传播吋产生的非线 性效应大小的参量;
[0023] 第一阈值分割单元, 用于采用第一阈值对非线性参量分布图进行分割, 小于第 一阈值的区域为组织残留区域, 大于第一阈值的区域为造影剂区域, 所述第一 阈值使得非线性参量分布图中小于第一阈值的区域中绝大多数在该造影信息帧 中对应的点属于组织残留信息;
[0024] 第一图像形成单元, 用于对该造影信息帧中属于组织残留区域的信息进行抑制 处理, 对该造影信息帧中属于造影剂区域的信息进行增强处理, 并基于处理后 的结果形成第一图像数据;
[0025] 第二阈值分割单元, 用于采用第二阈值对非线性参量分布图进行分割, 小于第 二阈值的区域为组织残留区域, 大于第二阈值的区域为造影剂区域, 所述第二 阈值使得非线性参量分布图中大于第二阈值的区域中绝大多数在该造影信息帧 中对应的点属于造影剂信息, 所述第二阈值大于第一阈值;
[0026] 第二图像形成单元, 用于对该造影信息帧中属于组织残留区域的信息进行抑制 处理, 对该造影信息帧中属于造影剂区域的信息进行增强处理, 并基于处理后 的结果形成第二图像数据;
[0027] 图像融合单元, 用于将第一图像数据和第二图像数据进行融合, 形成增强后的
[0028] 根据第三方面, 一种实施例中提供一种超声造影成像设备, 包括:
[0029] 探头, 用于向组织发射超声波并接收组织反射回的超声回波;
[0030] 前信号处理模块, 对超声回波信号进行处理, 从超声回波信号中提取反映造影 剂微泡信息的非线性成分和反映组织解剖特征的线性成分;
[0031] 第一信号处理模块, 用于对线性成分进行处理;
[0032] 组织图像生成模块, 用于将第一信号处理模块的输出处理成组织图像数据; [0033] 第二信号处理模块, 用于对非线性成分进行处理;
[0034] 造影图像生成模块, 用于将第二信号处理模块的输出处理成造影图像数据; [0035] 用于对造影信息进行增强的超声造影图像增强系统, 超声造影图像增强系统如 上述所述, 以得到增强后的造影图像数据;
[0036] 显示模块, 用于对组织图像数据和增强后的造影图像数据进行可视化显示。
[0037] 本发明实施例中, 对造影信息帧的非线性参量分布图采用两极阈值分割, 每次 划分都分出造影剂区域和组织残留区域, 对造影剂区域进行增强处理, 对组织 残留区域抑制处理, 从而可实现对绝对属于造影剂区域进行大程度的增强, 对 绝对属于组织残留的区域进行大程度的抑制, 对介于造影剂和组织残留的区域 增强和抑制的程度稍微弱些, 最后再将两次分类和权重调整后的结果进行融合 处理, 使对造影剂的增强、 对造影剂的抑制, 对介于造影剂和组织残留的保守 处理, 都能达到比较理想的造影图像增强效果。
[0038]
[0039] 附图说明
[0040] 图 1为超声造影成像设备的第一种结构示意图;
[0041] 图 2为超声造影成像设备的第二种结构示意图;
[0042] 图 3为超声造影成像设备的第三种结构示意图;
[0043] 图 4为一种实施例中超声造影图像增强系统的结构示意图; [0044] 图 5为一种实施例中超声造影图像增强系统的流程图;
[0045] 图 6为一种实施例中采用直方图方式表示的非线性参量分布图;
[0046] 图 7为另一种实施例中采用二维方式表示的非线性参量分布图。
[0047]
[0048] 具体实施方式
[0049] 下面通过具体实施方式结合附图对本发明作进一步详细说明。
[0050] 如图 1所示, 超声造影成像设备包括探头 10、 前信号处理模块 13、 第一信号处 理模块 16、 第二信号处理模块 17、 组织图像生成模块 18、 造影图像生成模块 19
、 超声造影图像增强系统 20和显示模块 100。
[0051] 探头 10用于向人体组织 11发射超声波并接收组织 11反射回的超声回波, 超声回 波可存在信号接收模块 12中。
[0052] 前信号处理模块 13用于对超声回波信号进行处理, 提取有用的信息。 前信号处 理模块 13主要包含放大器、 波束合成器和滤波器, 提取更多有用的、 不同成分 的信号。 在超声造影成像设备中, 前信号处理模块 13需要提取两种不同的信号
, 一种是线性成分 14, 反映组织解剖特征; 一种是非线性成分 15, 反映造影剂 微泡信息。
[0053] 第一信号处理模块 16用于对线性成分进行处理, 第二信号处理模块 17用于对非 线性成分进行处理。 第一信号处理模块 16和第二信号处理模块 17的处理类似, 主要包括解调、 滤波、 降采样、 求包络、 求 Log、 动态范围变换等, 只是参数的 选择不同。
[0054] 组织图像生成模块 18用于将第一信号处理模块 16的输出处理成组织图像数据; 造影图像生成模块 19用于将第二信号处理模块 17的输出处理成造影图像数据。
[0055] 超声造影图像增强系统 20用于对造影信息进行增强, 以得到增强后的造影图像 数据; 显示模块 100用于对组织图像数据和增强后的造影图像数据进行可视化显 示。 大多数造影图像旁会同吋显示一幅组织图像, 用于定位病灶位置。 为了得 到微泡的非线性特征, 造影成像通常会在同一位置发射多次相位或者幅度不同 的脉冲波, 组织和造影图像都是用这多次发射波形中的部分进行成像, 所以组 织图像和造影图像显示的是同一吋刻同一位置的信息, 组织图像显示的是回波 的线性特性, 造影图像显示的是回波中的非线性特性。
[0056] 在采用造影增强技术对造影图像进行增强吋, 需要通过阈值将图像划分成组织 残留区域和造影剂区域, 然后对划归为造影剂区域的点进行增强, 但在对造影 增强的继续研究中, 发明人认识到组织残留区域和造影剂区域并没有明确的边 界, 任何阈值都无法正确的区分组织残留区域和造影剂区域。 因此, 在本发明 实施例中, 首先得到组织的线性信息和造影剂微泡的非线性信息, 再比较非线 性信息和线性信息得到非线性参量值, 接着根据非线性参量值对造影图像进行 两次分类, 每次都分成组织残留和造影剂两类, 对两类按照不同的权重增强或 抑制, 根据分类的不同, 权重系数也不同; 最后将两次分类和权重调整后的结 果进行融合处理。
[0057] 在具体实施例中, 超声造影图像增强系统 20可以对组织图像生成模块 18和造影 图像生成模块 19输出的图像域的造影信息进行增强处理, 如图 1所示, 超声造影 图像增强系统 18的输入端分别连接组织图像生成模块 18和造影图像生成模块 19 , 输出端连接到显示模块 100, 根据组织图像生成模块 18和造影图像生成模块 19 输出的组织图像数据和造影图像数据对造影图像数据进行增强处理, 并将增强 结果输出到显示模块 100。 超声造影图像增强系统 20还可以对前信号处理模块 13 输出的射频域的造影信息进行增强处理。 如图 2所示, 超声造影图像增强系统 20 连接在前信号处理模块 13和第二信号处理模块 17之间, 接收前信号处理模块 13 输出的非线性成分 15和线性成分 14, 根据非线性成分 15和线性成分 14对非线性 成分进行增强处理, 并将增强结果输出到第二信号处理模块 17。 超声造影图像 增强系统 18还可以对第二信号处理模块中任一环节的造影信息进行增强处理。 如图 3所示, 超声造影图像增强系统 20的输入端分别连接第一信号处理模块 16和 第二信号处理模块 17的任一环节, 例如解调后信号, 求包络后, Log变换后等, 超声造影图像增强系统 20的输出端连接造影图像生成模块 19。
[0058] 超声造影图像增强系统 20的结构图如图 4所示, 包括信息获取单元 201、 非线性 参量计算单元 202、 第一阈值分割单元 203、 第一图像形成单元 204、 第二阈值分 割单元 205、 第二图像形成单元 206和图像融合单元 207。 信息获取单元 201用于 根据造影成像过程中的超声回波信号获得非线性的造影信息和线性的组织信息 , 即造影信息和线性的组织信息是从同一超声回波信号中提取出的两个信息, 非线性信息为造影信息, 线性信息为组织残留信息; 非线性参量计算单元 202用 于比较同一吋刻的造影信息和组织信息得到一帧造影信息的非线性参量分布图 , 该非线性参量分布图称为该造影信息帧的非线性参量分布图, 所述非线性参 量是衡量超声波在媒质中传播吋产生的非线性效应大小的参量; 本实施例中, 非线性参量分布图根据同一吋刻的一帧非线性造影信息和一帧线性组织信息的 比值或差值计算得出。 第一阈值分割单元 203用于采用第一阈值对非线性参量分 布图进行分割, 小于第一阈值的区域为组织残留区域, 大于第一阈值的区域为 造影剂区域, 所述第一阈值使得非线性参量分布图中小于第一阈值的区域中绝 大多数在该造影信息帧中对应的点属于组织残留信息; 第一图像形成单元 204用 于对该造影信息帧中属于组织残留区域的信息进行抑制处理, 对该造影信息帧 中属于造影剂区域信息进行增强处理, 并基于处理后的结果形成第一图像数据 ; 第二阈值分割单元 205用于采用第二阈值对非线性参量分布图进行分割, 小于 第二阈值的区域为组织残留区域, 大于第二阈值的区域为造影剂区域, 所述第 二阈值使得非线性参量分布图中大于第二阈值的区域中绝大多数在该造影信息 帧中对应的点属于造影剂信息, 所述第二阈值大于第一阈值; 第二图像形成单 元 206用于对该造影信息帧中属于组织残留区域的信息进行抑制处理, 对该造影 信息帧中属于造影剂区域的信息进行增强处理, 并基于处理后的结果形成第二 图像数据; 图像融合单元 207用于将第一图像数据和第二图像数据进行融合, 形 成增强后的造影图像数据。
[0059] 在另一实施例中, 超声造影图像增强系统 20还包括噪声识别单元 208, 噪声识 别单元 208用于在非线性参量计算单元计算非线性参量之前判断造影信息中的噪 声, 将一帧造影信息分为噪声信号和除去噪声的非噪声造影信号; 非线性参量 计算单元 202在计算非线性参量吋采用非噪声造影信号计算非线性参量。
[0060] 超声造影图像增强系统的具体处理流程如图 5所示, 包括以下步骤:
[0061] 一、 去噪
[0062] 去噪的目的是还原信号自身大小, 避免 TGC (信号增益补偿) 及噪声对后面处 理带来影响。 一般超声系统中用 ATGC (analog time gain compensation,前端模拟 信号增益补偿) 、 DTGC (digital time gain compensation,后端数字域增益补偿) 用来补偿信号传播过程中深度方向的衰减来补偿信号沿传播方向的信号衰减, 如果没有 ATGC和 DTGC的作用, 探头空打吋的噪声是近似为白噪声, 加上了 AT GC和 DTGC的影响后, 噪声均值曲线是随深度递增的曲线。 为了得到信号真实 的大小, 减去噪声均值曲线不仅仅是去噪, 也消除了 ATGC和 DTGC对信号的影 响。
[0063] 步骤 21到 24是实现去噪。 去噪首先要获得噪声信号, 本实施例中获得噪声信号 的方法是暂停发射, 在超声造影图像增强系统 20前的环节执行步骤 21和 22, 缓 存一帧非线性造影噪声信号和一帧线性组织噪声信号, 即控制探头不发射, 然 后根据回波得到一帧非线性造影噪声信号和一帧线性组织噪声信号。 处理的具 体步骤是计算一帧噪声信号每行的均值得到噪声纵向均值曲线, 再经过低通滤 波得到光滑的噪声均值曲线。
[0064] 在步骤 23中, 将超声造影图像增强系统 20输入的造影信息减去非线性造影噪声 信号的噪声均值曲线, 得到去噪后的造影信号。
[0065] 在步骤 24中, 将超声造影图像增强系统 20输入的组织信息减去线性组织噪声信 号的噪声均值曲线, 得到去噪后的组织信号。
[0066] 二、 检测噪声
[0067] 步骤 25, 检测去噪后造影图像中的噪声, 将一帧造影图像划分成噪声区域 26和 非噪声造影剂区域 27。 噪声检测可用阈值分割的方法, 小于阈值的就是噪声, 阈值是根据系统特性经验确定的。
[0068] 三、 计算非线性参量
[0069] 本实施例中, 非线性参量是衡量超声波在媒质中传播吋产生的非线性效应大小 的重要参量。
[0070] 步骤 28中, 比较同一吋刻的一帧造影信息和一帧组织信息, 得到该造影信息帧 的非线性参量分布图。 理论上, 非线性参量为媒质物态方程泰勒展幵式中多次 项系数与线性系数的比值或差值, 例如, 媒质物态方程泰勒展幵式中, 线性系 数为 A, 二次项系数为 B, 三次项系数为 C, 非线性参量可以是 B/A、 C/A或 (B+C) /A, 及 C其他组合与 A的比值。 本实施例中, 非线性参量通过计算同一吋刻 造影信号和组织信号的比值或差值得出。 对于一帧造影信息, 其由 MxN个像素 值组成, 每个像素值对应显示界面上的一个像素点, 该像素值可能属于造影剂 产生的信号, 也可能属于组织残留信号。 对于一帧造影信息而言, 每个像素点 都会有对应的一个非线性参量, 如果该像素值是组织残留信息, 则其对应的非 线性参量值较小, 如果该像素值是造影剂信息, 则其对应的非线性参量值较大 , 因此, 通过计算造影信息帧中各像素值的非线性参量, 即可将属于组织残留 信息的像素值和属于造影剂信息的像素值区分幵。 根据同一吋刻的一帧非线性 造影信息和一帧线性组织信息的非线性参量即可得出该造影信息帧中各像素点 的非线性参量, 对这些非线性参量进行统计, 则可得到该造影信息帧的非线性 参量分布图, 得出的非线性参量分布图可以采用直方图方式表示, 也可以采用 二维图方式表示。
[0071] 四、 两极阈值划分
[0072] 如图 6所示为采用直方图方式表示的非线性参量分布图, 包括一个大峰和一个 小峰, 直方图的横坐标就是计算出的非线性参量值, 单位是 dB, 纵坐标表示造 影信息帧中与该非线性参量值对应的点的数量。 经试验证明, 大峰区域中绝大 部分区域所对应的造影信息帧中点反映的是造影剂微泡信息, 小峰区域中绝大 部分区域所对应的造影信息帧中点反映的是组织残留信息。
[0073] 在确定两极阈值的另一实施例中, 非线性参量分布图为二维图, 如图 7所示, 其以色彩表示造影信息帧中各点非线性参量值, 深蓝色的表示噪声区域, 非线 性参量值由大到小采用的颜色表示为红、 黄、 绿、 蓝, 因组织残留的非线性参 量值较小, 造影剂的非线性声参量值较大, 故可以确定约小于 -40dB的一定是组 织残留, 约大于 -20dB的一定是造影剂, 这里 -40dB就是较小阈值, -20dB就是较 大阈值。 图 7只是一个具体的例子, 实际中, 需要根据系统平台, 通过较多非线 性参量二维图, 找出确定为组织残留的点的非线性参量值大小, 其最大的值就 是较小阈值, 找出确定为造影剂的点的非线性参量值的大小, 其最小值就是较 大阈值。
[0074] 本实施例中, 通过两极阈值来划分非线性参量分布图, 每次划分都分为造影剂 区域和组织残留区域, 对造影剂区域进行增强显示, 对组织残留区域抑制显示 。 两极阈值就是极小、 极大两个阈值, 非线性参量不同, 阈值的选择也不同, 阈值的大小和系统的相消性有关, 大阈值要大于系统的相消值, 小阈值要小于 系统的相消值, 相消值简单讲就是注入造影剂最强吋刻和没有注入造影剂吋两 个非线性成分大小的差值。 具体如下:
[0075] 在步骤 29中, 采用第一阈值对非线性参量分布图进行分割, 如图 6中, 左边的 直线阈值为第一阈值, 该第一阈值使得非线性参量分布图中小于第一阈值的区 域中绝大多数在该造影信息帧中对应的点属于组织残留信息, 因此第一阈值也 称为极小阈值。 划分后可认为: 在如图 6所示的非线性参量分布图中, 小于第一 阈值的左边的区域为组织残留区域 32, 组织残留区域 32中的点在造影信息帧中 的对应点被认为属于组织残留信息。 大于第一阈值的右边的区域为造影剂区域 3 1, 造影剂区域 31中的点在造影信息帧中的对应点被认为可能属于造影剂产生的 信息。 此吋的小于第一阈值的几乎都是组织残留, 而大于第一阈值的可能属于 造影剂。 再结合噪声区域 26, 这样就把造影图像分成噪声、 造影剂和组织残留 三个不同的区域。
[0076] 步骤 35, 为了达到增强造影剂、 抑制组织残留的目的, 对该造影信息帧中属于 组织残留区域的信息进行抑制处理, 对该造影信息帧中属于造影剂区域的信息 进行增强处理。 例如对上述三个区域分别乘以或加上不同的权重系数 Pl、 P2和 P 3, 如果是乘以权重系数, 则造影剂区域的系数 P2大于 1, 组织残留区域的系数 P 3应小于 1, 噪声区域系数 P1可以等于 1, 如果想抑制噪声, P1也可以小于 1。 同 吋, P2大于 1, 也增加了造影信号的动态范围。 如果是加上权重系数, 则 P3<0<P 2, 这样能增强造影信号强度, 减弱组织残留强度, P1≤0, P1<0也达到抑制噪声 的目的。 对上述三个区域分别乘以或加上不同的权重系数后再形成一个新的图 像, 称为第一图像数据。
[0077] 在步骤 30中, 采用第二阈值对非线性参量分布图进行分割, 如图 6中, 右边直 线阈值为第二阈值, 该第二阈值使得非线性参量分布图中大于第二阈值的区域 中绝大多数在该造影信息帧中对应的点属于造影剂信息, 因此, 第二阈值也称 为极大阈值, 且第二阈值大于第一阈值。 划分后可认为: 在如图 6所示的非线性 参量分布图中, 小于第二阈值的区域为组织残留区域 34, 组织残留区域 34中的 点在造影信息帧中的对应点被认为可能属于组织残留信息。 大于第二阈值的区 域为造影剂区域 33, 即造影剂区域 33中的点在造影信息帧中的对应点被认为属 于造影剂产生的信息。 此吋的大于第二阈值的几乎都是造影剂, 而小于第二阈 值的则可能属于组织残留。 再结合噪声区域 26, 也将造影图像分成噪声、 造影 剂和组织残留三个不同的区域。
[0078] 步骤 36, 为了达到增强造影剂、 抑制组织残留的目的, 对该造影信息帧中属于 组织残留区域的信息进行抑制处理, 对该造影信息帧中属于造影剂区域的信息 进行增强处理。 例如对步骤 30中的三个区域分别乘以或加上不同的权重系数 P1 、 P4和 P5。 同样如果乘以权重系数, 则造影剂区域的系数 P4大于 1, 组织残留区 域的系数 P5应小于 1, 即 P5<1<P4, 而噪声区域系数 PI等于或小于 1。 如果加上权 重系数, 则 P5<0<P4, P1≤0。 对上述三个区域分别乘以或加上不同的权重系数后 再形成一个新的图像, 称为第二图像数据。
[0079] 为了达到增强造影图像的效果, 对绝对属于造影剂区域的像素点增强的程度大 , 对绝对属于组织残留区域的像素点抑制的程度大, 对介于造影剂和组织残留 的区域增强和抑制的程度稍微弱些。 如果乘以权重系数, 增强和抑制的参数满 足 P4>P2>1, P3<P5<1 ; 如果加上权重系数, 增强和抑制的参数满足 P4>P2>0, P 3<P5<0。
[0080] 五、 图像融合
[0081] 因为任何阈值都无法完全正确的区分造影剂和组织残留, 所以第一图像数据和 第二图像数据都有些极端, 作为最终的增强图像都有些不正确。 因此在接下来 的步骤 37中将两种图像进行融合处理, 融合的方法有很多, 本实施例提出了一 种简单的融合方法, 将第一图像数据乘以权重系数 K, 第二图像数据乘以权重系 数 1-Κ, 0<Κ<1,再组合显示。 为了达到增强造影剂的目的, 1-K≥K, K≤0.5。
[0082] 本实施例最后通过对两个图像数据进行融合处理, 相互补充, 使对造影剂的增 强、 对造影剂的抑制, 对介于造影剂和组织残留的保守处理, 都能达到比较理 想的造影图像增强效果。
[0083]
[0084] 本领域技术人员可以理解, 上述实施方式中各种方法的全部或部分步骤可以通 过程序来指令相关硬件完成, 该程序可以存储于一计算机可读存储介质中, 存 储介质可以包括: 只读存储器、 随机存储器、 磁盘或光盘等。
[0085]
[0086] 以上应用了具体个例对本发明进行阐述, 只是用于帮助理解本发明, 并不用以 限制本发明。 对于本领域的一般技术人员, 依据本发明的思想, 可以对上述具 体实施方式进行变化。
技术问题
问题的解决方案
发明的有益效果

Claims

权利要求书
[权利要求 1] 一种增强超声造影图像的方法,其特征在于包括:
获取信息, 根据造影成像过程中的超声回波信号获得非线性的造影信 息和线性的组织信息;
计算非线性参量, 比较所述造影信息和组织信息得到一帧造影信息的 非线性参量分布图, 所述非线性参量是衡量超声波在媒质中传播吋产 生的非线性效应大小的参量;
第一阈值分割, 采用第一阈值对非线性参量分布图进行分割, 小于第 一阈值的区域为组织残留区域, 大于第一阈值的区域为造影剂区域, 所述第一阈值使得非线性参量分布图中小于第一阈值的区域中绝大多 数在该造影信息帧中对应的点属于组织残留信息; 形成第一图像, 对该造影信息帧中属于组织残留区域的信息进行抑制 处理, 对该造影信息帧中属于造影剂区域的信息进行增强处理, 并基 于处理后的结果形成第一图像数据;
第二阈值分割, 采用第二阈值对非线性参量分布图进行分割, 小于第 二阈值的区域为组织残留区域, 大于第二阈值的区域为造影剂区域, 所述第二阈值使得非线性参量分布图中大于第二阈值的区域中绝大多 数在该造影信息帧中对应的点属于造影剂信息, 所述第二阈值大于第 一阈值;
形成第二图像, 对该造影信息帧中属于组织残留区域的信息进行抑制 处理, 对该造影信息帧中属于造影剂区域的信息进行增强处理, 并基 于处理后的结果形成第二图像数据;
图像融合, 将第一图像数据和第二图像数据进行融合, 形成增强后的 造影图像数据。
[权利要求 2] 如权利要求 1所述的方法, 其特征在于, 所述增强处理包括乘以大于 1 的权重系数或加上大于 0的权重系数, 所述抑制处理包括乘以小于 1的 权重系数或加上小于 0的权重系数。
[权利要求 3] 如权利要求 2所述的方法, 其特征在于, 形成第一图像数据过程中进 行抑制处理的权重系数小于形成第二图像数据过程中进行抑制处理的 权重系数, 形成第二图像数据过程中进行增强处理的权重系数大于形 成第一图像数据过程中进行增强处理的权重系数。
[权利要求 4] 如权利要求 1所述的方法, 其特征在于, 在计算非线性参量之前还包 括: 判断造影信息中的噪声, 将一帧造影信息分为噪声信号和除去噪 声的非噪声造影信号; 在计算非线性参量吋采用非噪声造影信号计算 非线性参量。
[权利要求 5] 如权利要求 1所述的方法, 其特征在于, 所述非线性参量分布图根据 同一吋刻的一帧非线性造影信息和一帧线性组织信息的比值或差值计 算得出。
[权利要求 6] 如权利要求 1至 5中任一项所述的方法, 其特征在于, 所述图像融合包 括: 将第一图像数据和第二图像数据分别乘以第一权重系数和第二权 重系数, 所述第一权重系数与第二权重系数的和为 1且第二权重系数 大于或等于第一权重系数。
[权利要求 7] —种超声造影图像增强系统,其特征在于包括:
信息获取单元, 用于根据造影成像过程中的超声回波信号获得非线性 的造影信息和线性的组织信息;
非线性参量计算单元, 用于比较所述造影信息和组织信息得到一帧造 影信息的非线性参量分布图, 所述非线性参量是衡量超声波在媒质中 传播吋产生的非线性效应大小的参量;
第一阈值分割单元, 用于采用第一阈值对非线性参量分布图进行分割 , 小于第一阈值的区域为组织残留区域, 大于第一阈值的区域为造影 剂区域, 所述第一阈值使得非线性参量分布图中小于第一阈值的区域 中绝大多数在该造影信息帧中对应的点属于组织残留信息; 第一图像形成单元, 用于对该造影信息帧中属于组织残留区域的信息 进行抑制处理, 对该造影信息帧中属于造影剂区域的信息进行增强处 理, 并基于处理后的结果形成第一图像数据;
第二阈值分割单元, 用于采用第二阈值对非线性参量分布图进行分割 , 小于第二阈值的区域为组织残留区域, 大于第二阈值的区域为造影 剂区域, 所述第二阈值使得非线性参量分布图中大于第二阈值的区域 中绝大多数在该造影信息帧中对应的点属于造影剂信息, 所述第二阈 值大于第一阈值;
第二图像形成单元, 用于对该造影信息帧中属于组织残留区域的信息 进行抑制处理, 对该造影信息帧中属于造影剂区域的信息进行增强处 理, 并基于处理后的结果形成第二图像数据;
图像融合单元, 用于将第一图像数据和第二图像数据进行融合, 形成 增强后的造影图像数据。
[权利要求 8] 如权利要求 7所述的系统, 其特征在于, 所述增强处理包括乘以大于 1 的权重系数或加上大于 0的权重系数, 所述抑制处理包括乘以小于 1的 权重系数或加上小于 0的权重系数。
[权利要求 9] 如权利要求 8所述的系统, 其特征在于, 第一图像形成单元进行抑制 处理的权重系数小于第二图像形成单元进行抑制处理的权重系数, 第 二图像形成单元进行增强处理的权重系数大于第一图像形成单元进行 增强处理的权重系数。
[权利要求 10] 如权利要求 7所述的系统, 其特征在于, 还包括噪声识别单元, 所述 噪声识别单元用于在非线性参量计算单元计算非线性参量之前判断造 影信息中的噪声, 将一帧造影信息分为噪声造影和除去噪声的非噪声 造影信号; 非线性参量计算单元在计算非线性参量吋采用非噪声信号 计算非线性参量。
[权利要求 11] 如权利要求 7所述的系统, 其特征在于, 所述非线性参量计算单元根 据同一吋刻的一帧非线性造影信息和一帧线性组织信息的比值或差值 计算该造影信息帧的非线性参量分布图。
[权利要求 12] 如权利要求 7至 11中任一项所述的系统, 其特征在于, 所述图像融合 单元将第一图像数据和第二图像数据分别乘以第一权重系数和第二权 重系数后进行融合, 所述第一权重系数与第二权重系数的和为 1且第 二权重系数大于或等于第一权重系数。
[权利要求 13] 如权利要求 7至 11中任一项所述的系统, 其特征在于, 所述非线性参 量分布图为直方图或二维图。
[权利要求 14] 一种超声造影成像设备, 其特征在于包括:
探头, 用于向组织发射超声波并接收组织反射回的超声回波; 前信号处理模块, 对超声回波信号进行处理, 从超声回波信号中提取 反映造影剂微泡信息的非线性成分和反映组织解剖特征的线性成分; 第一信号处理模块, 用于对线性成分进行处理; 组织图像生成模块, 用于将第一信号处理模块的输出处理成组织图像 数据;
第二信号处理模块, 用于对非线性成分进行处理; 造影图像生成模块, 用于将第二信号处理模块的输出处理成造影图像 数据;
如权利要求 7至 13中任一项所述的超声造影图像增强系统, 用于对造 影信息进行增强, 以得到增强后的造影图像数据; 显示模块, 用于对组织图像数据和增强后的造影图像数据进行可视化 显示。
[权利要求 15] 如权利要求 14所述的超声造影成像设备, 其特征在于, 所述超声造影 图像增强系统对前信号处理模块输出的射频域的造影信息进行增强处 理。
[权利要求 16] 如权利要求 14所述的超声造影成像设备, 其特征在于, 所述超声造影 图像增强系统对组织图像生成模块和造影图像生成模块输出的图像域 的造影信息进行增强处理。
[权利要求 17] 如权利要求 14所述的超声造影成像设备, 其特征在于, 所述超声造影 图像增强系统对第二信号处理模块中任一环节的造影信息进行增强处 理。
PCT/CN2016/087979 2016-06-30 2016-06-30 一种增强超声造影图像的方法、系统及超声造影成像设备 WO2018000359A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2016/087979 WO2018000359A1 (zh) 2016-06-30 2016-06-30 一种增强超声造影图像的方法、系统及超声造影成像设备
CN201680039006.7A CN108135566B (zh) 2016-06-30 2016-06-30 一种增强超声造影图像的方法、系统及超声造影成像设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/087979 WO2018000359A1 (zh) 2016-06-30 2016-06-30 一种增强超声造影图像的方法、系统及超声造影成像设备

Publications (1)

Publication Number Publication Date
WO2018000359A1 true WO2018000359A1 (zh) 2018-01-04

Family

ID=60785845

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/087979 WO2018000359A1 (zh) 2016-06-30 2016-06-30 一种增强超声造影图像的方法、系统及超声造影成像设备

Country Status (2)

Country Link
CN (1) CN108135566B (zh)
WO (1) WO2018000359A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112837254A (zh) * 2021-02-25 2021-05-25 普联技术有限公司 一种图像融合方法、装置、终端设备及存储介质

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111200699B (zh) * 2018-11-19 2022-04-26 瑞昱半导体股份有限公司 影像调整方法
CN116416125A (zh) * 2019-06-26 2023-07-11 图码思(成都)科技有限公司 一种针对图像序列的图像拼接方法及终端
WO2021232192A1 (zh) * 2020-05-18 2021-11-25 深圳迈瑞生物医疗电子股份有限公司 超声造影成像方法、装置和存储介质
CN113538299B (zh) * 2021-09-13 2022-01-11 深圳瀚维智能医疗科技有限公司 超声图像去噪方法、装置、设备及计算机可读存储介质
CN113768547B (zh) * 2021-09-14 2024-03-22 南京超维景生物科技有限公司 冠状动脉成像方法及装置、存储介质及电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN86107496A (zh) * 1985-10-31 1987-06-10 通用电气公司 用于超声波成象的自适应时间增益补偿系统
US20030097069A1 (en) * 2001-11-21 2003-05-22 Ge Medical Systems Global Technology Company, Llc Computationally efficient noise reduction filter for enhancement of ultrasound images
US20030236459A1 (en) * 2002-06-20 2003-12-25 Loftman Rickard C. Automatic gain compensation for multiple mode or contrast agent imaging
CN101987023A (zh) * 2009-07-31 2011-03-23 深圳迈瑞生物医疗电子股份有限公司 一种超声成像增益补偿及图像优化方法及其装置和系统
CN103845077A (zh) * 2012-12-05 2014-06-11 深圳迈瑞生物医疗电子股份有限公司 超声图像增益优化方法及超声成像增益自动优化装置
CN104720850A (zh) * 2013-12-23 2015-06-24 深圳迈瑞生物医疗电子股份有限公司 一种超声造影成像方法及造影图像的区域检测、显像方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6508767B2 (en) * 2000-02-16 2003-01-21 Koninklijke Philips Electronics N.V. Ultrasonic harmonic image segmentation
US6371914B1 (en) * 2000-04-13 2002-04-16 Bracco Research S.A. Single-shot phase cancellation ultrasound contrast imaging
CN105631867B (zh) * 2015-12-25 2019-08-23 中国科学院深圳先进技术研究院 一种全自动超声造影图像分割方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN86107496A (zh) * 1985-10-31 1987-06-10 通用电气公司 用于超声波成象的自适应时间增益补偿系统
US20030097069A1 (en) * 2001-11-21 2003-05-22 Ge Medical Systems Global Technology Company, Llc Computationally efficient noise reduction filter for enhancement of ultrasound images
US20030236459A1 (en) * 2002-06-20 2003-12-25 Loftman Rickard C. Automatic gain compensation for multiple mode or contrast agent imaging
CN101987023A (zh) * 2009-07-31 2011-03-23 深圳迈瑞生物医疗电子股份有限公司 一种超声成像增益补偿及图像优化方法及其装置和系统
CN103845077A (zh) * 2012-12-05 2014-06-11 深圳迈瑞生物医疗电子股份有限公司 超声图像增益优化方法及超声成像增益自动优化装置
CN104720850A (zh) * 2013-12-23 2015-06-24 深圳迈瑞生物医疗电子股份有限公司 一种超声造影成像方法及造影图像的区域检测、显像方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112837254A (zh) * 2021-02-25 2021-05-25 普联技术有限公司 一种图像融合方法、装置、终端设备及存储介质

Also Published As

Publication number Publication date
CN108135566A (zh) 2018-06-08
CN108135566B (zh) 2020-11-24

Similar Documents

Publication Publication Date Title
WO2018000359A1 (zh) 一种增强超声造影图像的方法、系统及超声造影成像设备
EP3419516B1 (en) Ultrasound blood flow imaging
Narayanan et al. A view on despeckling in ultrasound imaging
Yu et al. Ultrasound speckle reduction by a SUSAN-controlled anisotropic diffusion method
CN102306377B (zh) 一种超声图像去噪的方法和装置
JP6272707B2 (ja) 医療超音波の適応的音圧推定方法
Guo et al. A novel approach to speckle reduction in ultrasound imaging
JP2001299764A (ja) 超音波診断装置
Tay et al. A wavelet thresholding method to reduce ultrasound artifacts
US9186124B2 (en) Ultrasonic diagnostic apparatus, ultrasonic image processing apparatus, and ultrasonic image processing method
JP2020520711A (ja) 超音波を用いた組織微小脈管構造の可視化のためのシステム及び方法
Tasnim et al. Study of speckle noise reduction from ultrasound B-mode images using different filtering techniques
US8047991B2 (en) Automatic identification of orientation in medical diagnostic ultrasound
Huang et al. High-quality ultrafast power Doppler imaging based on spatial angular coherence factor
Abd-Elmoniem et al. Real time adaptive ultrasound speckle reduction and coherence enhancement
WO2015096352A1 (zh) 一种超声造影成像方法及造影图像的区域检测、显像方法
WO2016161574A1 (zh) 一种超声造影成像方法及装置
Shao et al. Artifacts detection-based adaptive filtering to noise reduction of strain imaging
KR100778823B1 (ko) 초음파 영상 처리 시스템 및 방법
Fang et al. Discrimination between newly formed and aged thrombi using empirical mode decomposition of ultrasound B-scan image
Hourani et al. Block-wise ultrasound image deconvolution from fundamental and harmonic images
Kleckler et al. Characterization of Heterogeneous Perfusion in Contrast-Enhanced Ultrasound
Randhawa et al. Investigation of performance of Savitzky-Golay filter for speckle reduction in ultrasound images
Hourani et al. Joint deconvolution of fundamental and harmonic ultrasound images
Matrone et al. Enhanced ultrasound harmonic imaging using the filtered-delay multiply and sum beamformer

Legal Events

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

Ref document number: 16906759

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16906759

Country of ref document: EP

Kind code of ref document: A1