LU500526B1 - Image Optimization Method for Neonatal Lung Ultrasonography - Google Patents

Image Optimization Method for Neonatal Lung Ultrasonography Download PDF

Info

Publication number
LU500526B1
LU500526B1 LU500526A LU500526A LU500526B1 LU 500526 B1 LU500526 B1 LU 500526B1 LU 500526 A LU500526 A LU 500526A LU 500526 A LU500526 A LU 500526A LU 500526 B1 LU500526 B1 LU 500526B1
Authority
LU
Luxembourg
Prior art keywords
image
ultrasound
scale value
neonatal lung
gray
Prior art date
Application number
LU500526A
Other languages
German (de)
Inventor
Ruxin Qiu
Guo Guo
Xiaoling Ren
Jing Liu
Original Assignee
Beijing Chaoyang Distr Maternal And Child Healthcare Hospital
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 Beijing Chaoyang Distr Maternal And Child Healthcare Hospital filed Critical Beijing Chaoyang Distr Maternal And Child Healthcare Hospital
Priority to LU500526A priority Critical patent/LU500526B1/en
Application granted granted Critical
Publication of LU500526B1 publication Critical patent/LU500526B1/en

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
    • A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0858—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving measuring tissue layers, e.g. skin, interfaces
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0866—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving foetal diagnosis; pre-natal or peri-natal diagnosis of the baby
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5269—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving detection or reduction of artifacts

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biomedical Technology (AREA)
  • Veterinary Medicine (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Gynecology & Obstetrics (AREA)
  • Pregnancy & Childbirth (AREA)
  • Vascular Medicine (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

A method for adjusting and image optimization for neonatal lung ultrasonography, transmitting ultrasound; using a focus adjuster to adjust the number of ultrasound focus points; screening out the measured neonatal lung ultrasound images with a evaluation score within a preset score range; the generated ultrasound image data of the neonatal lung is stored by the image storage module and the memory is used, and the neonatal lung ultrasound image is displayed on the screen through the display module. The present invention can accurately present the tissue structure of the ultrasound; at the same time, the measured ultrasound image of neonatal lung is input into the trained Faster R-CNN model to obtain the evaluation score corresponding to the measured ultrasound image of the neonatal lung; the obtained ultrasound images of the neonatal lung ultrasound images whose evaluation scores are within the preset score range are ideal ultrasound images.

Description

DESCRIPTION Image Optimization Method for Neonatal Lung Ultrasonography
TECHNICAL FIELD The invention belongs to the technical field of ultrasound imaging, and in particular relates to an image optimization method of the neonatal lung ultrasonography.
BACKGROUND Ultrasound imaging uses ultrasonic sound beams to scan the human body and obtain images of internal organs through the reception and processing of reflected signals. There are many commonly used ultrasonic instruments: Type A (amplitude modulation type) indicates the strength of the reflected signal by the amplitude of the wave and it displays a kind of "echograph". Type M (spot scanning type) represents the spatial position from shallow to deep in the vertical direction, and the time in the horizontal direction, which is displayed as a curve diagram of the movement of the light spot at different times. The two types above are one-dimensional displays, and the application range is limited. Type B (brightness modulation type) is an ultrasonic section imager, abbreviated as "B- ultrasound". The light spots with different brightness indicate the strength of the received signal. When the probe moves along the horizontal position, the light spot on the display screen also moves in the horizontal direction synchronously, connecting the light spot trajectory into a cross-sectional view scanned by the ultrasonic sound beam. For two- dimensional imaging. As for the type D, it is made according to the principle of ultrasonic Doppler. The type C uses a scanning method similar to that of a TV, showing a cross- sectional sound image perpendicular to the sound beam. In recent years, ultrasound imaging technology has continued to develop, such as gray-scale display and color display, real-time imaging, ultrasound holography, penetrating ultrasound imaging, ultrasound parallel tomography, three-dimensional imaging, and intracavitary ultrasound imaging. However, the coherence of the existing neonatal lung ultrasound images will form speckle noise. Speckle noise will cover the neonatal lung ultrasound images and become the main feature of the neonatal lung ultrasound images. Therefore, speckle noise will greatly reduce the quality of neonatal lung ultrasound images. At the same time, the quality of ultrasound images of novice sonographers has always been checked and evaluated by experienced sonographers, and the evaluation results have errors and evaluation are of low efficiency.
In summary, the existing technology has the following problem that the coherence of the existing neonatal lung ultrasound images will form speckle noise, and the speckle noise will cover the neonatal lung ultrasound images and become the main feature of the neonatal lung ultrasound images. Because of the image characteristics, speckle noise will greatly reduce the quality of neonatal lung ultrasound images. At the same time, the quality of ultrasound images from novice doctors has always been checked and evaluated by experienced doctors, and the evaluation results have errors and evaluation efficiency are relatively low.
SUMMARY In view of the problems existing in the prior art, the present invention provides an image optimization method for neonatal lung ultrasonography.
The present invention is realized by an image optimization method for neonatal lung ultrasonography. The image optimization method includes:
In the first step, the ultrasound transmitter module emits ultrasound; the ultrasound collector 1s used to collect the signals reflected by the ultrasound beam which scans the neonatal lung; through the depth adjustment module in the central control module, the ultrasound adjustment button is used to adjust the ultrasound scanning depth to 4 -5 cm; in the second step, the focus adjustment module uses the focus adjuster to adjust the number of ultrasound focus points to 1-2, and adjusts its position to make the first focus point close to the pleural line; the ultrasound image generation module uses the ultrasound signal processing program to process the collected ultrasound signals to generate an ultrasound image of the neonatal lung; use the denoising module to accurately present the structure to be ultrasound; at the same time, the image evaluation module inputs the measured ultrasound image of the neonatal lung into the trained Faster R-CNN model to obtain corresponding evaluation scores of the neonatal lung ultrasound images; The third step is to screen out the ultrasound images whose evaluation scores are within the preset score range among the measured ultrasound images of the neonatal lung and the screened images are ideal ultrasound images, which can improve the accuracy and efficiency of acquired ultrasound images that meet the quality standards; The fourth step is to use the memory to store the generated the neonatal lung ultrasound image data through the image storage module, and display the neonatal lung images on the screen through the display module.
Further, the denoising method of image optimization method for neonatal lung ultrasonography is as follows:
(1) Constructing a total variation image denoising model of the neonatal lung ultrasound images, and determining the initial gray-scale value of the real image and the initial gray- scale value of the speckle image; (2) Based on the minimization model of the neonatal lung ultrasound image, iteratively estimating the initial gray-scale value of the real image and the initial gray-scale value of the speckle image; when the estimated gray-scale value of the real image and the estimated value of the speckle image meet the preset convergence condition, the estimated gray-scale value of the real image and the estimated gray-scale value of the speckle image are obtained; (3) Generating a denoised ultrasound image of neonatal lung according to the estimated gray-scale value of the real image.
Further, the minimization model of the neonatal lung ultrasound image iteratively estimates the initial gray-scale value of the real image speckle image and the followings steps are included: Based on the minimization model of the neonatal lung ultrasound image and the initial gray-scale value of the real image as the constant of the minimization model; the gray- scale value of the speckle image is estimated to obtain the first estimated gray-scale value of the speckle image; Based on the minimization model and using the initial gray-scale value of the speckle image as the constant of the minimization model, the gray-scale value of the real image is estimated to obtain the first estimated gray-scale value of the real image; The first estimated gray-scale value of the speckle image and the first estimated gray- scale value of the real image are alternately used as the initial constants of the minimization model, and the gray-scale value of the real image and the gray-scale value of the real image are estimated alternately.
Further, the preset convergence conditions for the image optimization of neonatal lung ultrasonography include: Determining the absolute value of the first difference value between the estimated gray- scale value of the real image obtained in this iteration and the estimated gray-scale value of the real image obtained in the previous iteration; Determining the absolute value of the second difference value between the estimated gray-scale value of the speckle image obtained in this iteration and the estimated gray- scale value of the speckle image obtained in the previous iteration; Determining the first ratio between the absolute value of the first difference value and the estimated gray-scale value of the real image obtained in this iteration; Determining the second ratio between the absolute value of the second difference value and the estimated gray- scale value of the speckle image obtained in this iteration; When the first ratio and the second ratio are both less than the preset threshold, it suggests that the estimated gray-scale value of the real image meets the preset convergence condition; otherwise, iterating continues until the convergence condition is satisfied.
Further, the image evaluation includes the following steps: 1) Acquiring an ultrasound image of the neonatal lung through an ultrasound image generation module; performing enhancement processing on the ultrasound image of the neonatal lung, and generating ultrasound image training samples;
2) Formulating the evaluation score of the ultrasound image training samples according to the evaluation protocol, where the evaluation protocol is that if one of the multiple lung structures in the ultrasound image training sample is clear, one point will be added; 3) Training the pre-established Faster R-CNN model by using the ultrasound image training samples and the evaluation scores corresponding to the samples; 4) Inputting the measured ultrasound image of the neonatal lung into the trained Faster R- CNN model, and obtaining an evaluation score corresponding to the measured ultrasound image of the neonatal lung; 5) Screening out the measured ultrasound images of the neonatal lung whose evaluation scores are within the preset score range and the screened images are regarded as ideal images.
Another object of the present invention is to provide a neonatal lung ultrasonography which adopts the image optimization method for neonatal lung ultrasonography described in the invention.
The advantages and beneficial effects of the present invention are: the present invention uses the scanned neonatal lung ultrasound image as the original input through the denoising module, and iteratively estimates the initial gray-scale value of the real image and the initial gray-scale value of the speckle image; when the estimated gray- scale value of the real image and the estimated gray- scale value of the speckle image meet the convergence condition, the denoised neonatal lung ultrasound images of are generated, so that the denoised ultrasound image can retain the details and edge parts the original images and good homogeneity can be shown and the denoised ultrasound image of the neonatal lung can accurately show the tissue structure of the ultrasound; at the same time,
the neonatal lung ultrasound images measured by the image evaluation module are input to the Faster R-CNN model after training, and the evaluation score corresponding to the measured ultrasound image of the neonatal lung 1s obtained; the measured ultrasound image of the neonatal lung whose evaluation scores are within the preset score range the is screened out; the screened ultrasound images are regarded ideal; the accuracy and efficiency of acquired ultrasound images that meet the quality standards can be improved.
BRIEF DESCRIPTION OF THE FIGURES Fig. 1 1s a flowchart of an image optimization method for neonatal lung ultrasonography provided by an embodiment of the present invention.
2 is a schematic structural diagram for neonatal lung ultrasonography provided by an embodiment of the present invention; In the Figure: 1. Ultrasonic transmitter module; 2. Ultrasound signal acquisition module;
3. Central control module; 4. Depth adjustment module; 5. Focus adjustment module; 6. Ultrasound image generation module; 7. Denoising module; 8. Image Optimization module; 9. Image evaluation module; 10. Image storage module; 11. Display module.
DESCRIPTION OF THE INVENTION In order to further understand the content, features, and effects of the present invention, the following embodiments are exemplified and described in detail below with accompanying figures.
In view of the problems existing in the prior art, the present invention provides an image optimization method for neonatal lung ultrasonography. The present invention will be described in detail below with reference to the accompanying figures.
As shown in Figure 1, the image optimization method for neonatal lung ultrasonography provided by the present invention includes the following steps: S101: The ultrasound transmitter module emits ultrasound; the ultrasound collector 1s used to collect the signals reflected by the ultrasound beam which scans the neonatal lung; S102: Through the depth adjustment module in the central control module, the ultrasound adjustment button is used to adjust the ultrasound scan depth, so that the scan depth is 4 - cm; S103: The focus adjustment module uses the focus adjuster to adjust the number of ultrasound focus points to 1-2, and adjusts its position so that the first focus point is close to the pleural line; S104: The ultrasound image generation module uses the ultrasound signal processing program to process the collected ultrasound signals to generate an ultrasound image of the neonatal lung; S105: Using the denoising model to remove speckle noise from the neonatal lung ultrasound image through the denoising module; using the optimization program to optimize the neonatal lung ultrasound image through the image optimization module; S106: Using the evaluation program to evaluate the neonatal lung ultrasound image through the image evaluation module; S107: Using the memory to store the generated ultrasound image data of the neonatal lung through the image storage module; using the display module to display the ultrasound image.
As shown in Figure 2, the neonatal lung ultrasonography provided by the embodiment of the present invention includes: an ultrasound transmitter module 1, an ultrasound signal acquisition module 2, a central control module 3, a depth adjustment module 4, a focus adjustment module 5, and an ultrasound image generation module 6, denoising module 7, image optimization module 8, image evaluation module 9, image storage module 10, and display module 11. The ultrasonic transmitter module 1 is connected to the ultrasonic signal acquisition module 2 and is used to transmit ultrasonic waves through an ultrasonic generator; The ultrasonic signal acquisition module 2 is connected with the ultrasonic transmitter module 1 and the central control module 3, and is used to collect the signals reflected by the ultrasound beam which scans the neonatal lung through the ultrasonic receiver; Central control module 3, and ultrasonic transmitter module 1, ultrasonic signal acquisition module 2, depth adjustment module 4, focus adjustment module 5, ultrasonic image generation module 6, denoising module 7, image optimization module 8, image evaluation module 9, image storage Module 10 and display module 11 are connected to control the normal operation of each module through the host; The depth adjustment module 4, connected with the central control module 3, is used to adjust the ultrasonic scanning depth through the ultrasonic adjustment button, so that the scanning depth is 4-5 cm; The focus adjustment module 5, connected to the central control module 3, is used to adjust the number of ultrasound focus points to 1-2 through a focus adjuster, and adjust its position so that the first focus point is close to the pleural line;
The ultrasound image generation module 6 is connected to the central control module 3, and is used to process the collected ultrasound signals through an ultrasound signal processing program to generate an ultrasound image of the neonatal lung; The denoising module 7, connected to the central control module 3, is used to remove speckle noise from the neonatal lung ultrasound image through the denoising model; The image optimization module 8, connected to the central control module 3, is used to optimize the ultrasound image of the neonatal lung through the optimization program; The image evaluation module 9, which is connected to the central control module 3, is used to evaluate the neonatal lung ultrasound images through the evaluation program; The image storage module 10 is connected to the central control module 3 and is used to store the generated ultrasound image data of the neonatal lung through the memory; The display module 11 is connected to the central control module 3 and is used for displaying the generated ultrasound image of the neonatal lung through the display.
The denoising steps of denoising module 6 are as follows: (1) Constructing a total variation image denoising model of the neonatal lung ultrasound images, and determining the initial gray-scale value of the real image and the initial gray- scale value of the speckle image; (2) Based on the minimization model of the neonatal lung ultrasound image, iteratively estimating the initial gray-scale value of the real image and the initial gray-scale value of the speckle image.
When the estimated gray-scale value of the real image and the estimated value of the speckle image meet the preset convergence condition, the estimated gray-scale value of the real image and the estimated gray-scale value of the speckle image are obtained;
(3) Generating a denoised neonatal lung ultrasound image according to the estimated gray-scale value of the real image.
The minimization model of the neonatal lung ultrasound image 1teratively estimates the initial gray-scale value of the real image speckle image and the followings steps are included: Based on the minimization model of the neonatal lung ultrasound image and the initial gray-scale value of the real image as the constant of the minimization model, the gray- scale value of the speckle image is estimated to obtain the first estimated gray-scale value of the speckle image; Based on the minimization model and using the initial gray-scale value of the speckle image as the constant of the minimization model, the gray-scale value of the real image is estimated to obtain the first estimated gray-scale value of the real image; The first estimated gray-scale value of the speckle image and the first estimated gray- scale value of the real image are alternately used as the initial constants of the minimization model, and the gray-scale value of the real image and the gray-scale value of the real image are estimated alternately.
The preset convergence conditions include: Determining the absolute value of the first difference value between the estimated gray- scale value of the real image obtained in this iteration and the estimated gray-scale value of the real image obtained in the previous iteration; Determining the absolute value of the second difference value between the estimated gray-scale value of the speckle image obtained in this iteration and the estimated gray- scale value of the speckle image obtained in the previous iteration;
Determining the first ratio between the absolute value of the first difference value and the estimated gray-scale value of the real image obtained in this iteration; Determining the second ratio between the absolute value of the second difference value and the estimated gray- scale value of the speckle image obtained in this iteration; When the first ratio and the second ratio are both less than the preset threshold, it suggests that the estimated gray-scale value of the real image meets the preset convergence condition. otherwise, iterating continues until the convergence condition is satisfied.
The image evaluation steps of evaluation module 9 are as follows: 1) Acquiring an ultrasound image of the neonatal lung through an ultrasound image generation module; performing enhancement processing on the ultrasound image of the neonatal lung, and generating ultrasound image training samples; 2) Formulating the evaluation score of the ultrasound image training samples according to the evaluation protocol, where the evaluation protocol is that if one of the multiple lung structures in the ultrasound image training sample is clear one point will be added; 3) Training the pre-established Faster R-CNN model by using the ultrasound image training samples and the evaluation scores corresponding to the samples; 4) Inputting the measured ultrasound image of the neonatal lung into the trained Faster R- CNN model, and obtaining an evaluation score corresponding to the measured ultrasound image of the neonatal lung; 5) Screening out the measured ultrasound images of the neonatal lung whose evaluation scores are within the preset score range and the screened images are regarded as ideal images.
The descriptions above are only preferred embodiments of the present invention, and are not intended to limit the present invention in any form.
Any simple modification, equivalent changes and modifications made to the embodiments above based on the technical essence of the present invention should fall within the scope of the technical solution of the invention.

Claims (6)

CLAIMS :
1. An image optimization method for neonatal lung ultrasonography, is characterized by including the following steps: in the first step, the ultrasound transmitter module emits ultrasound; the ultrasound collector is used to collect the signals reflected by the ultrasound beam which scans the neonatal lung; through the depth adjustment module in the central control module, the ultrasound adjustment button is used to adjust the ultrasound scan depth, so that the scan depth is 4 -5 cm; in the second step, the focus adjustment module uses the focus adjuster to adjust the number of ultrasound focus points to 1-2, and adjusts its position so that the first focus point is close to the pleural line; the ultrasound image generation module uses the ultrasound signal processing program to process the collected ultrasound signals to generate an ultrasound image of the neonatal lung; use the denoising module to accurately present the tissue structure to be ultrasound; at the same time, the image evaluation module inputs the measured ultrasound image of the neonatal lung into the trained Faster R-CNN model to obtain the corresponding evaluation scores of the neonatal lung ultrasound images; the third step is to screen out the ultrasound images whose evaluation scores are within the preset score range among the measured ultrasound images of the neonatal lung; the screened are ideal ultrasound images and can improve the accuracy and efficiency of ultrasound images that meet the quality standards;
the fourth step 1s to use the memory to store the generated neonatal lung ultrasound image data through the image storage module, and display the neonatal lung ultrasound image on the screen through the display module.
2. The image optimization method for neonatal lung ultrasonography according to claim 1, is characterized by the following denoising method: (1) constructing a total variation image denoising model of the neonatal lung ultrasound images, and determining the initial gray-scale value of the real image and the initial gray- scale value of the speckle image; (2) based on the minimization model of the neonatal lung ultrasound image, iteratively estimating the initial gray-scale value of the real image and the initial gray-scale value of the speckle image; when the estimated gray-scale value of the real image and the estimated value of the speckle image meet the preset convergence condition, the estimated gray-scale value of the real image and the estimated gray-scale value of the speckle image are obtained; (3) generating a denoised ultrasound image of neonatal lung according to the estimated gray-scale value of the real image.
3. The image optimization method for neonatal lung ultrasonography according to claim 2, is characterized in that the minimization model of the neonatal lung ultrasound image iteratively estimates the initial gray-scale value of the real image speckle image and the followings steps are included: based on the minimization model of the neonatal lung ultrasound image and the initial gray-scale value of the real image as the constant of the minimization model, the gray-
scale value of the speckle image is estimated to obtain the first estimated gray-scale value of the speckle image; based on the minimization model and using the initial gray-scale value of the speckle image as the constant of the minimization model, the gray-scale value of the real image is estimated to obtain the first estimated gray-scale value of the real image; the first estimated gray-scale value of the speckle image and the first estimated gray-scale value of the real image are alternately used as the initial constants of the minimization model, and the gray-scale value of the real image and the gray-scale value of the real image are estimated alternately.
4. The image optimization method for neonatal lung ultrasonography according to claim 3, 1s characterized by the preset convergence conditions which include: determining the absolute value of the first difference value between the estimated gray- scale value of the real image obtained in this iteration and the estimated gray-scale value of the real image obtained in the previous iteration; determining the absolute value of the second difference value between the estimated gray-scale value of the speckle image obtained in this iteration and the estimated gray- scale value of the speckle image obtained in the previous iteration; determining the first ratio between the absolute value of the first difference value and the estimated gray-scale value of the real image obtained in this iteration; determining the second ratio between the absolute value of the second difference value and the estimated gray- scale value of the speckle image obtained in this iteration;
when the first ratio and the second ratio are both less than the preset threshold, it suggests that the estimated gray-scale value of the real image meets the preset convergence condition; otherwise, iterating continues until the convergence condition is satisfied.
5. The image optimization method for neonatal lung ultrasonography according to claim 1, 1s characterized by the following image evaluation: 1) acquiring an ultrasound image of the neonatal lung through an ultrasound image generation module; performing enhancement processing on the ultrasound image of the neonatal lung, and generating ultrasound image training samples; 2) formulating the evaluation score of the ultrasound image training samples according to the evaluation protocol, where the evaluation protocol is that if one of the multiple lung structures in the ultrasound image training sample is clear one point will be added; 3) training the pre-established Faster R-CNN model by using the ultrasound image training samples and the evaluation scores corresponding to the samples; 4) inputting the measured ultrasound image of the neonatal lung into the trained Faster R- CNN model, and obtaining an evaluation score corresponding to the measured ultrasound image of the neonatal lung; 5) screening out the measured ultrasound images of the neonatal lung whose evaluation scores are within the preset score range and the screened images are regarded as ideal images.
6. A neonatal lung ultrasonography which applies the image optimization method for neonatal lung ultrasonography according to any one of claims 1 to 5.
LU500526A 2021-08-11 2021-08-11 Image Optimization Method for Neonatal Lung Ultrasonography LU500526B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
LU500526A LU500526B1 (en) 2021-08-11 2021-08-11 Image Optimization Method for Neonatal Lung Ultrasonography

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
LU500526A LU500526B1 (en) 2021-08-11 2021-08-11 Image Optimization Method for Neonatal Lung Ultrasonography

Publications (1)

Publication Number Publication Date
LU500526B1 true LU500526B1 (en) 2022-02-11

Family

ID=80215555

Family Applications (1)

Application Number Title Priority Date Filing Date
LU500526A LU500526B1 (en) 2021-08-11 2021-08-11 Image Optimization Method for Neonatal Lung Ultrasonography

Country Status (1)

Country Link
LU (1) LU500526B1 (en)

Similar Documents

Publication Publication Date Title
US8834374B2 (en) Setting an optimal image parameter in an ultrasound system
US9901323B2 (en) Aberration correction using channel data in ultrasound imaging system
JP4575738B2 (en) Ultrasonic image boundary extraction method, ultrasonic image boundary extraction device, and ultrasonic imaging device
US20160074017A1 (en) Ultrasound optimization method and ultrasonic medical device therefor
US20200146656A1 (en) Methods for ultrasound system independent attenuation coefficient estimation
US11819363B2 (en) Systems and methods to improve resolution of ultrasound images with a neural network
CN108685596A (en) Estimated using the tissue property of ultrasonic medical imaging
US11593933B2 (en) Systems and methods for ultrasound image quality determination
KR20190080121A (en) Ultrasound imaging apparatus and control method for the same
US20120238877A1 (en) Ultrasound diagnostic apparatus and ultrasound image producing method
WO2012029417A1 (en) Ultrasound diagnostic device and evaluation calculation method
US20210298717A1 (en) Methods and systems for performing fetal weight estimations
LU500526B1 (en) Image Optimization Method for Neonatal Lung Ultrasonography
CN110338842A (en) A kind of image optimization method of newborn's lungs ultrasonic image-forming system
US20230061869A1 (en) System and methods for beamforming sound speed selection
JP5823184B2 (en) Ultrasonic diagnostic apparatus, medical image processing apparatus, and medical image processing program
US11506771B2 (en) System and methods for flash suppression in ultrasound imaging
US20230065683A1 (en) System and methods for beamforming sound speed selection
JP7432426B2 (en) Ultrasonic diagnostic equipment, signal processing equipment, and signal processing programs
US20240046467A1 (en) Method and system for evaluating ultrasound data for the purpose of ultrasound attenuation estimation in a medium
JP7345374B2 (en) Ultrasound diagnostic equipment
US20230267618A1 (en) Systems and methods for automated ultrasound examination
EP4014884A1 (en) Apparatus for use in analysing an ultrasound image of a subject
JP2023104734A (en) Ultrasonic diagnostic device and image processing device
JP2019177093A (en) Ultrasonic diagnostic device, medical image processing device, and medical image processing program

Legal Events

Date Code Title Description
FG Patent granted

Effective date: 20220211