WO2023103975A1 - 医学影像的移动检测和校正方法、系统和计算机可读介质 - Google Patents

医学影像的移动检测和校正方法、系统和计算机可读介质 Download PDF

Info

Publication number
WO2023103975A1
WO2023103975A1 PCT/CN2022/136636 CN2022136636W WO2023103975A1 WO 2023103975 A1 WO2023103975 A1 WO 2023103975A1 CN 2022136636 W CN2022136636 W CN 2022136636W WO 2023103975 A1 WO2023103975 A1 WO 2023103975A1
Authority
WO
WIPO (PCT)
Prior art keywords
images
interest
medical image
frame
region
Prior art date
Application number
PCT/CN2022/136636
Other languages
English (en)
French (fr)
Inventor
柯纪纶
颜若芳
郑媚方
Original Assignee
柯纪纶
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 柯纪纶 filed Critical 柯纪纶
Priority to AU2022405390A priority Critical patent/AU2022405390A1/en
Priority to CN202280080685.8A priority patent/CN118339582A/zh
Priority to CA3239117A priority patent/CA3239117A1/en
Priority to EP22903397.2A priority patent/EP4446973A1/en
Priority to IL313372A priority patent/IL313372A/en
Priority to KR1020247022351A priority patent/KR20240116526A/ko
Publication of WO2023103975A1 publication Critical patent/WO2023103975A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
    • A61B6/03—Computed tomography [CT]
    • A61B6/032—Transmission computed tomography [CT]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves 
    • A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves  involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
    • A61B6/03—Computed tomography [CT]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
    • A61B6/03—Computed tomography [CT]
    • A61B6/037—Emission tomography
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/503—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of the heart
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5258—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise
    • A61B6/5264—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise due to motion
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/58—Testing, adjusting or calibrating thereof
    • A61B6/582—Calibration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing

Definitions

  • the present disclosure relates to a medical imaging technology, in particular to a method, system and computer-readable medium for motion detection and correction of medical images.
  • nuclear medicine is an indispensable item due to its ability to provide functional assessment.
  • nuclear medicine examinations take a long time, and the movement of the patient's body or organs during the scanning process often easily leads to blurred and inaccurate images and wrong judgments by doctors.
  • MPI myocardial perfusion imaging
  • Figure 1 due to the contraction and beating behavior of the patient's heart itself (as shown in the legend 101, which respectively show a cardiac cycle from left to right, the slices of the heart in dimensions such as its short axis, vertical long axis, and horizontal long axis morphological schematic diagram) and the ups and downs of human respiration (as shown in Figure 102, which describe from left to right in a respiratory cycle stage, the schematic diagram of the section shape of the heart in its short axis, vertical long axis and horizontal long axis etc. dimensions), Under the influence of both, the resulting MPI image of the heart (as shown in legend 103) is often blurred and difficult to diagnose.
  • the solution to the above problems is to install additional monitoring equipment to monitor the heartbeat and respiratory displacement of the human body, and then correct the medical images (eg, MPI images) according to the signals measured by the monitoring equipment.
  • this method of correction not only consumes detection resources, but also may cause poor monitoring equipment settings, tracking errors, or failure to integrate with scanning equipment (for example, single photon emission computerized tomography (SPECT) equipment on the market is generally There is no optional respiratory monitoring equipment) and so on.
  • scanning equipment for example, single photon emission computerized tomography (SPECT) equipment on the market is generally There is no optional respiratory monitoring equipment
  • the present disclosure provides a system for detection and correction of medical image movement, including: a management platform for providing a user interface to submit instructions for performing optimization processing on medical images of target organs, the medical images including list mode data; and an optimization device, configured to perform optimization processing of the medical image according to the instruction, wherein the optimization device performing the optimization process includes: dividing the list mode data corresponding to the medical image into frames with a fixed time dimension, To image each frame into a frame image; mark a region of interest (volume of interest, VOI) in each frame image, wherein each region of interest contains a target organ; calculate the motion curve of the target organ according to each region of interest in each frame image; The motion curve reconstructs the medical image into an optimized medical image; and displays the optimized medical image on a user interface.
  • a management platform for providing a user interface to submit instructions for performing optimization processing on medical images of target organs, the medical images including list mode data
  • an optimization device configured to perform optimization processing of the medical image according to the instruction, wherein the optimization device performing the optimization process includes:
  • the optimization device includes a deep learning module, and the optimization device executes marking the region of interest in each frame of image including: identifying the target organ in each frame of image by the deep learning module The binary segmentation area; through the deep learning module, each binary segmentation area is blurred to generate a soft mask; through the deep learning module, each soft mask is applied to each frame image; through the deep learning module based on Each soft mask uses an initial ellipsoid to fit the target organ in each frame of image; and the deep learning module expands each initial ellipsoid outward by a preset distance according to its radius to generate an ellipsoid representing each region of interest.
  • the optimization device performing calculation of the motion curve of the target organ according to each region of interest of each frame of image includes: dividing each region of interest into a first sub-region of interest and a second sub-region of interest; Extract the first centroid (center of mass, COM) of each first sub-region of interest and the three-dimensional coordinates of the second centroid of each second sub-region of interest as the description value of each frame image; each description value of each frame image is expressed as Principal component analysis is used for dimensionality reduction, and the maximum feature of each description value after dimensionality reduction is used as the movement/rotation signal of the target organ; and each frame of image is grouped and filtered according to each movement/rotation signal to calculate the motion curve.
  • the target organ is the heart
  • each region of interest is divided into the first sub-region of interest and the second sub-region of interest along the long axis of the heart and in the direction of the short axis.
  • the motion curve is drawn relative to any one of the head-tail axis, left-right axis, and ventral-dorsal axis of the human body
  • the optimization device performs reconstruction of the medical image into an optimized medical image according to the motion curve, including : Select the reference object from each frame of image; use the motion curve as a reference to perform motion compensation of each frame of image, wherein, the motion compensation of each frame of image includes: relative to the reference object, each frame of image except the reference object All the pixels contained in the former are adjusted according to their three-dimensional coordinates in any one of the head-tail axis, left-right axis and ventral-dorsal axis; and the adjustment operation of each frame of image is repeated until the adjustment operation of each frame of image until the correlation coefficient between the integration and the reference object reaches the maximum value; and integrating the motion-compensated frames of images to reconstruct an optimized medical image.
  • the adjustment operation includes but not limited
  • the optimization device further includes executing each The gating of frame images, wherein the gating is used to integrate those with similar temporal and/or positional relationships in each frame of images into a preset number of gated set images, and wherein the time and/or Or the positional relationship is defined by the cyclical phases of human breathing and/or heartbeat.
  • the motion curve is drawn relative to any one of the head-tail axis, left-right axis, and ventral-dorsal axis of the human body
  • the optimization device reconstructs the medical image into an optimized medical image according to the motion curve, including : Select the reference object from the images of each gated group; use the motion curve as a reference to perform motion compensation for the images of each gated group, wherein, the motion compensation for images of each gated group includes: relative to the reference object, performing an adjustment operation on any one of the cranial-caudal axis, the left-right axis, and the ventral-dorsal axis according to their three-dimensional coordinates in the image except for the reference object; and repeatedly performing the adjustment operation of each gated group image, Until the correlation coefficient between the integration of the adjusted gated group images and the reference object reaches the maximum value; and integrating the motion compensated gated group images to reconstruct an optimized medical image.
  • the fixed time dimension is in units of 100 milliseconds to 500 milliseconds.
  • the disclosed system further includes: a scanning device for photographing the target organ to obtain a medical image, wherein the scanning device is a single photon emission computed tomography device, a positron emission tomography (positron emission tomography) , PET) equipment, magnetic resonance imaging (magnetic resonance imaging, MRI) equipment, any one of computer tomography (computer tomography, CT) equipment; medical image storage and transmission system (picture archiving and communication system, PACS), with for storing the medical image and the optimized medical image; and an office report computer for accessing or displaying the medical image and the optimized medical image.
  • a scanning device for photographing the target organ to obtain a medical image
  • the scanning device is a single photon emission computed tomography device, a positron emission tomography (positron emission tomography) , PET) equipment, magnetic resonance imaging (magnetic resonance imaging, MRI) equipment, any one of computer tomography (computer tomography, CT) equipment; medical image storage and transmission system (picture archiving and
  • the present disclosure further provides a method for motion detection and correction of a medical image, including: acquiring a medical image about a target organ, wherein the medical image includes list mode data; list mode data corresponding to the medical image Divide into frames with a fixed time dimension to image each frame into a frame image; mark the region of interest in each frame of image, where each region of interest contains the target organ; calculate the motion curve of the target organ according to each region of interest in each frame of image and reconstructing the medical image into an optimized medical image according to the motion curve.
  • marking the region of interest in each frame of image includes: using a deep learning module to identify a binary segmented region containing a target organ in each frame of image; The meta-segmentation area is blurred to generate a soft mask; through the deep learning module, each soft mask is applied to each frame image; through the deep learning module, the initial ellipsoid is used to fit each frame image based on each soft mask the target organ; and expanding each initial ellipsoid outward by a preset distance according to its radius through the deep learning module to generate ellipsoids representing each region of interest.
  • calculating the motion curve of the target organ according to each region of interest of each frame of image includes: dividing each region of interest into a first sub-region of interest and a second sub-region of interest; The three-dimensional coordinates of the first centroid of the sub-region of interest and the second centroid of each second sub-region of interest are used as the description value of each frame of image; the dimensionality reduction of each description value of each frame of image is performed by principal component analysis, and the dimensionality reduction The maximum feature of each description value is used as the movement/rotation signal of the target organ; and each frame of images is grouped and filtered according to each movement/rotation signal to calculate the motion curve.
  • the target organ is the heart, and dividing each region of interest into a first sub-region of interest and a second sub-region of interest is performed along the long axis of the heart and in the direction of the short axis.
  • the motion curve is drawn relative to any one of the head-tail axis, left-right axis, and ventral-dorsal axis of the human body, and reconstructing the medical image into an optimized medical image according to the motion curve includes: Select the reference object in the image; use the motion curve as a reference to perform motion compensation for each frame of image, wherein, the motion compensation for each frame of image includes: relative to the reference object, the motion compensation of each frame of image except the reference object All the included pixels are adjusted according to their three-dimensional coordinates in any one of the cranial-caudal axis, the left-right axis, and the ventral-dorsal axis; and the adjustment operation of each frame image is repeated until the integration and reference of each frame image after the adjustment operation until the correlation coefficient between the objects reaches the maximum value; and integrating the motion-compensated frames of images to reconstruct an optimized medical image.
  • the disclosed method further comprising performing gating of each frame of image after calculating the motion curve of the target organ according to each centroid, and each region of interest of each frame of image does not accurately correspond to the motion curve , wherein the gating is used to integrate those with similar time and/or positional relationships in each frame of images into a preset number of gated group images, and wherein the time and/or positional relationship is based on human respiration and/or heartbeat Cycle phases are defined.
  • the motion curve is drawn relative to any one of the head-tail axis, left-right axis, and ventral-dorsal axis of the human body, and reconstructing the medical image into an optimized medical image according to the motion curve includes: Select the reference object in the control group image; use the motion curve as a reference to perform motion compensation for each gated group image, wherein, the motion compensation for each gated group image includes: relative to the reference object, remove performing adjustment operations on any one of the cranial-caudal axis, the left-right axis, and the ventral-dorsal axis according to their three-dimensional coordinates; Until the correlation coefficient between the integration of the adjusted gated group images and the reference object reaches the maximum value; and integrating the motion compensated gated group images to reconstruct an optimized medical image.
  • the medical image is obtained by shooting the target organ through a scanning device, and wherein the scanning device is a single photon emission computed tomography device, a positron emission tomography device, a magnetic resonance imaging device and any of the computed tomography devices.
  • the scanning device is a single photon emission computed tomography device, a positron emission tomography device, a magnetic resonance imaging device and any of the computed tomography devices.
  • the fixed time dimension is in units of 100 milliseconds to 500 milliseconds.
  • the present disclosure further provides a computer-readable storage medium, which is applied to a computer and stores instructions to execute at least one method for motion detection and correction of medical images described above.
  • the method, system, and computer-readable medium for motion detection and correction of medical images disclosed in the present disclosure can divide medical images about target organs into multiple frames of images according to list mode data, and analyze the interest in the multiple frames of images. Multiple centroids of the area to calculate the motion curve of the target organ during scanning, and then perform medical image reconstruction optimization based on the motion curve, so that the movement of human organs or lesions can be considered without installing additional monitoring equipment, so as to Motion detection and correction for medical images.
  • Figure 1 depicts an embodiment of performing myocardial perfusion imaging with current techniques
  • FIG. 2 depicts a schematic diagram of the system architecture of motion detection and correction of medical images of the present disclosure
  • FIG. 3 depicts a schematic diagram of an embodiment of a system for motion detection and correction of medical images of the present disclosure
  • FIG. 4 depicts a flow chart of the steps of the method for motion detection and correction of medical images of the present disclosure
  • FIG. 5 depicts a partial embodiment of the method for motion detection and correction of medical images of the present disclosure
  • FIG. 6C describes some embodiments of the method for motion detection and correction of medical images of the present disclosure, and COM in FIG. 6C represents the centroid;
  • FIG. 7 describes a partial implementation of the method for motion detection and correction of medical images of the present disclosure, where COM1 and COM2 represent centroid 1 and centroid 2, respectively;
  • FIG. 8 depicts a partial embodiment of the method for motion detection and correction of medical images of the present disclosure
  • FIG. 9 depicts a partial embodiment of a method for motion detection and correction of medical images of the present disclosure.
  • FIG. 10 depicts a partial embodiment of the method for motion detection and correction of medical images of the present disclosure
  • Figure 11 depicts a partial embodiment of the method of motion detection and correction of medical images of the present disclosure.
  • FIG. 12 depicts a schematic diagram of an embodiment of the method for motion detection and correction of medical images of the present disclosure.
  • FIG. 2 A schematic diagram of the system architecture of the present disclosure for performing motion detection and correction of medical images can be observed from FIG. 2 .
  • the management platform 201 of the present disclosure is used to integrate the processing flow of medical images, including: receiving and transmitting medical images, providing users with access to medical images, and executing according to user needs Optimal processing of medical images, etc.
  • the management platform 201 may be presented through any suitable user interface such as a web page, an application program page, or a man-machine interface, which is not particularly limited herein.
  • the optimization device 202 of the present disclosure is used as a background service for performing corresponding medical image optimization processing (including motion detection and correction) according to instructions submitted by users at the management platform 201 .
  • the optimization device 202 of the present disclosure can be any suitable physical computer system, cloud system, etc., and the optimization device 202 and the management platform 201 can also be realized by an integrated computer system, and there is no special limitation in this disclosure. .
  • the scanning device 203 of the present disclosure may be any detection device that can take medical images, for example, including but not limited to: single photon emission computed tomography equipment, positron emission tomography equipment, magnetic resonance imaging equipment , computerized tomography equipment, etc., used to obtain medical images of the parts to be detected by the patient (for example, including but not limited to: heart, lungs, coronary arteries, liver, stomach, etc.).
  • the medical images obtained by the scanning device 203 of the present disclosure include their corresponding list mode data, so it is helpful for non-real-time (for example, post-referral) retrospective analysis and correction of the captured medical images .
  • the medical image storage and transmission system 204 of the present disclosure can be any storage system used by current hospitals, for storing the medical images obtained by the scanning device 203 and/or optimized by the optimization device 202 optimized medical images.
  • the inter-diagnosis report computer 205 of the present disclosure may be any terminal device used by physicians during the inter-examination, which is used to provide physicians with access to or display medical images stored in the medical image storage and transmission system 204 and/or Optimize medical images.
  • the management platform 201 shown in FIG. Imaging and Communications in Medicine, DICOM communicate with each other, so it can provide a high degree of extensibility of the present disclosure.
  • the present disclosure is not limited to the above-mentioned components; for example, according to operation requirements, any number of the above-mentioned components can be integrated into the same device, or designed so that a single management platform 201 and/or optimization device 202 can The present disclosure has no special limitation on the optimization of the medical images that support multiple scanning devices 203 .
  • FIG. 3 depicts a schematic diagram of a specific embodiment of performing motion detection and correction of medical images in the present disclosure.
  • the present disclosure may perform motion detection and correction on medical images of the heart taken using single photon emission computed tomography (SPECT) equipment.
  • the scanning device 203 e.g., single photon emission computed tomography device
  • the scanning device 203 includes 19 pinhole collimators and 19 CdZnTe (CZT) sensors (e.g., a The CZT gamma camera of the CZT component) is used to scan the heart from a right oblique front view to a left oblique rear view and obtain related SPECT images (ie, medical images).
  • CZT CdZnTe
  • the imaging process of the SPECT image by the scanning device 203 may include: setting dual energy windows ( energy window) to scan the heart; store the list mode and/or frame mode (frame mode) data corresponding to the SPECT image according to the scan result; transmit the list mode and/or frame mode data to the scanning device 203 in the form of standard medical digital imaging and communication A built-in workstation; and the steps of resampling list mode and/or frame mode data along the cardiac short axis, vertical long axis and horizontal long axis for display.
  • the applicable specifications, equipment, or methods of obtaining medical images described by the scanning device 203 in the present disclosure are only examples, and are not intended to limit the content of the present disclosure.
  • the medical image 302 can be directly stored in the medical image storage and transmission system 204 for reference by the inter-diagnosis report computer 205 .
  • the management platform 203 can specify that the medical image 302 received from the scanning device 203 should be immediately detected and corrected by the optimization device 202, including Time dimension segmentation 303 , motion centroid analysis 304 , and deformation model correction 305 are processed to obtain an optimized medical image 306 .
  • the obtained medical image 302 includes list mode data, it is also possible to access the medical image 302 to the medical image storage and transmission system 204 through the management platform 201 when it is found that the medical image 302 is blurred afterwards, and pass The optimization device 202 performs back analysis correction.
  • Fig. 4 describes the flow chart of the steps of performing the motion detection and correction of medical images in the present disclosure (for example, the optimization device 202 executes the processing procedures of 303 to 305 in Fig. 3 above), and the implementation of each step can be implemented through Fig. 5 to Fig. 12 and The instructions below follow step by step.
  • the user interface provided by the management platform 201 can be used to select the medical image that needs to be detected and corrected (for example, the heart image obtained by a patient in one scan).
  • the medical image can be obtained immediately through the management platform 201 when the patient is scanning at the scanning device 203 , or can be accessed through the management platform 201 to the medical image storage and transmission system 204 when needed.
  • the optimization device 202 executes the above-mentioned time dimension segmentation 303 shown in FIG. Frames with a fixed time dimension (e.g., in units of 500 milliseconds), and said frames are projected via back-projection onto an object plane (e.g., the surface of the 19 CZT sensors of the aforementioned scanning device 203, which is parallel to Corresponding image planes, and intersect at the common focus of the multi-pinhole system) for imaging, as shown in the imaging results of each frame in FIG.
  • an object plane e.g., the surface of the 19 CZT sensors of the aforementioned scanning device 203, which is parallel to Corresponding image planes, and intersect at the common focus of the multi-pinhole system
  • each frame can be cut in a fixed time dimension at any value from 100 milliseconds to 500 milliseconds; however, according to the computing power or job requirements of the optimization device 202, the frames can also be divided into other suitable fixed time Dimension cutting, which is not particularly limited in the present disclosure.
  • Displacement information such as position, rotation angle and/or axis (for example, long axis and short axis of the heart) in the frame image.
  • the ROI 501 including the target organ is spherical; however, according to the shape of the target organ or operation requirements, the ROI 501 may also be marked in other suitable shapes, which is not particularly limited in the present disclosure.
  • the annotation of the myocardial region can be performed by the physician through a specific application program (for example, the automated software built in the optimization device 202), and its operation
  • the steps may include: marking a 3D ellipsoid passing through the center of the myocardium in the frame image (for example, obtained at the aforementioned step S20); obtaining three-dimensional coordinates including the center of the sphere from the 3D ellipsoid; Short axis, vertical long axis and horizontal long axis (such as axial) radius, three-axis rotation angle, the long and short ratio and angle of the heart base plane and other 12 spherical parameters; and according to the above spherical parameters, by the specific application program
  • the smooth inner and outer surfaces are fitted inwardly and outwardly from the 3D ellipsoid marked by the physician, and the frame
  • the labeling of the region of interest 501 in step S30 can also be implemented by a deep learning module (for example, constructing a deep learning module in the optimization device 202 ).
  • the deep learning module includes main structures such as convolutional layers, deconvolutional layers, leaky linear rectification activation layers, residual connections, beating connections, etc., and thousands of sets of frame images of myocardial regions marked by physicians (for example, The aforementioned annotators through specific applications) are used as training data for deep learning modules.
  • the deep learning module that has completed the training and can accurately predict the myocardial region when inputting unlabeled frame images can be used for labeling the ROI 501 in step S30.
  • the trained deep learning module can first receive the frame image generated at step S20 (as shown in FIG. cutaway screen). In some embodiments of the present disclosure, the deep learning module will determine the region 601 of the heart in the frame image (as shown in FIG. The range where the rotation angle is zero)), and then identify the binary segmented region 603 including the myocardial region (the region indicated by 603 in FIG. 6B ).
  • the deep learning module blurs the binary segmented region 603 to generate a soft mask, and further applies the soft mask to the original frame image, thereby excluding the image activity of the region other than the myocardial region in the frame image.
  • the deep learning module uses an ellipsoid to fit the myocardial region 602 on the frame image (the ellipsoid section corresponding to the solid line 602 in FIG. 6B ), and expands the radius of the ellipsoid outward. 2 cm to generate a new ellipsoid representing the region of interest 604 (the region corresponding to the solid line indicated by 604 in FIG. 6C ).
  • the optimization device 202 performs the motion centroid analysis 304 shown in FIG. 3 for the target organ within the region of interest marked in step S30 .
  • the ROI in each frame of images is divided into two. Reference can be made to the difference between Figure 6C and Figure 7 to see an implementation of segmenting a region of interest.
  • the region of interest 604 marked in each frame of image in FIG. 6C is divided into two parts along the long axis of the heart in step S40 to form sub-regions of interest 701 and 702 about the heart as shown in FIG.
  • FIG. 7 The left side of FIG. 7 is a slice view of the heart in its short-axis dimension, so the observed sub-regions of interest 701 and 702 are presented in an overlapping manner. Then, through the obtained two sub-regions of interest 701, 702 of the target organ, their respective centroids 703, 704 can be calculated, so as to assist in tracking the movement of the target organ in each frame of image, and then obtain the motion curve of the target organ during shooting .
  • the region of interest 604 is divided into two sub-regions of interest 701 and 702 as shown in FIG. 7 , and then motion centroid analysis is performed through the obtained two centroids 703 and 704
  • the advantages are shown in Figure 8.
  • 801 to 803 respectively represent when the target organ (for example, the heart) translates (801), translates and rotates (802), rotates (803), etc.
  • the effect of observing the above-mentioned cases 801 to 803 with a single centroid 605 and two centroids 703, 704 is then indicated.
  • the manner of analyzing the center of mass of the present disclosure is not limited except for the above-mentioned description;
  • the short axis (according to the long axis direction), or according to the requirements (for example, according to the shape characteristics of the target organ), divides the region of interest 604 into more than two sub-regions of interest, and observes the target organ with more than two centroids Displacement and rotation cases.
  • step S40 when the region of interest 604 is divided into sub-regions of interest 701, 702, calculate the three-dimensional coordinates corresponding to the centroids 703, 704 in each frame image (for example, in the form of "(x1, y1, z1), (x2, y2, z2)"), as the description value of each frame image.
  • principal component analysis principal component analysis, PCA
  • PCA principal component analysis
  • each frame of image can also be filtered according to the movement/rotation signal of the target organ, so as to filter out too high-frequency noise signals, and then calculate the motion curve of the target organ during shooting.
  • the motion curve of the target organ during shooting is shown in FIG.
  • the motion curves of the abdomen and back (Y axis, the schematic diagram of the middle layer in Figure 9), head and tail (Z axis, the schematic diagram of the lower layer of Figure 9), wherein, the curve shown in 901 in Figure 9 (indicated by a dotted line) represents the centroid in each frame of image
  • the relative positions of 703 or 704 on the above-mentioned X-axis, Y-axis and Z-axis change, and the curve shown in 902 (indicated by a short dotted line) represents the position of the center of mass 703 or 704 on the above-mentioned X-axis, Y-axis and Z-axis calculated according to the curve 901 rotation signal, and the curve shown in 903 (indicated by a solid line) represents the more precise position change of the centroid 703 or 704 obtained after grouping and accumulating each frame of image.
  • the optimization device 202 may first judge whether the movement degree of the target organ is too large according to the aforementioned motion curve of the target organ, and then If it is too large (for example, when the amplitude of the motion curve of the target organ is greater than 50 mm), the radiologist may request the patient to perform another scan and repeat steps S10 to S40 to obtain a new motion curve of the target organ. If there is no need to re-scan, the subsequent steps can be continued to correct the deformation model of the medical image.
  • the above steps S10 to S50 are used to complete the motion detection procedure of the medical image, and according to the result of the motion detection, the optimization device 202 can continue to perform the deformation model correction 305 shown in FIG. 3 on the medical image, such as the following steps S60 to S80 instruction of.
  • the optimization device 202 confirms whether the position of the region of interest in each frame of image accurately corresponds to the motion curve before step S60, if each When the ROI of the frame image does not accurately correspond to the motion curve, gating in step S70 may be performed. On the contrary, each frame of image can be directly subjected to the optimal reconstruction of the medical image in step S80. In some embodiments of the present disclosure, regardless of whether the region of interest of each frame of image has an accurate corresponding motion curve, directly performing the gating of each frame of image in step S70 can help to further grasp the movement of the target organ, and reduce the cost of step S80. Time spent on optimal reconstruction of medical images.
  • the gating described in step S70 is used to perform the close temporal and/or positional relationship among the frames of images according to the motion curve of the target organ (such as the heart) calculated in step S40. Integrate to form gated group images.
  • Fig. 10 depicts the implementation of organizing each frame of medical images of the heart into eight gating groups in the present disclosure, wherein the eight gating groups in the upper row are the gating groups observed with the vertical long axis of the heart Group images, and the eight gated groups in the lower row are images of the gated group observed with the short axis of the heart.
  • the eight gated group images can also be corresponding to respiration and/or heartbeat
  • Each cycle stage of each cycle as shown in Figure 10, from left to right is the respiratory cycle stage of inhalation/expiration, and each gated group image is relative to the gated group image of the inspiratory end stage (the leftmost gated group Image) has an iterative displacement relationship observed on the head-tail axis, left-right axis and/or ventral-dorsal axis of the human body, so step S70 is also called respiratory gating or heartbeat gating.
  • FIG. 10 schematically depicts an embodiment in which all frame images from a medical image about the heart are combined into eight gating groups according to the stages of the human respiratory cycle (that is, each gating group includes frames 12.5% of the total number of images); however, according to the needs of medical image correction and optimization effects or operational needs, the number of gating groups can also be increased or decreased, or an additional group of gating group images can be formed by considering the human cardiac cycle, or a comprehensive Considering the respiration and heartbeat cycle phases to form a group of gated images, which are not specifically limited in this paper.
  • step S80 for each frame of image that has been accurately positioned (that is, each frame of image of the medical image that has not been processed in step S70) or the gate that is gated by step S70
  • the reconstruction optimization of medical images is performed on the images of the control group.
  • the execution of medical image reconstruction optimization is mainly based on the reference objects in each frame of images or gated group images (for example, it is judged to be in the end of inspiratory phase, the isovolumic systolic period of the cardiac cycle, or a frame of images or gated objects that meet the above two group image) as the standard for motion compensation of the remaining frames of images or gated group images, and the above-mentioned motion curves relative to the head-tail axis, left-right axis and/or ventral-dorsal axis of the human body are used for adjustment (including rotation, Displacement, scaling, deformation and other adjustment operations) frame images or gated group images, and the correlation coefficient between the reconstructed optimized medical image and the reference object is used to observe the degree of completion of the reconstruction optimization of the medical image.
  • the optimization device 202 selects a frame of image at the end of inspiratory stage from the frame images or gated group images corresponding to the medical image or the gated group image as the reference object, and under the reference motion curve, relative to the reference object, all the pixels contained in the remaining frame images or the gated group image are arranged according to their three-dimensional coordinates on the head and tail axis, left and right sides of the human body axis and/or ventral-dorsal axis (that is, the aforementioned adjustment operations such as rotation, displacement, scaling, etc.), and then integrate the adjusted images of each frame or gated group of images to reconstruct an optimized medical image, and iteratively perform all the adjustments.
  • the reconstruction optimization of the above medical image is performed until the most similar between the optimized medical image and the reference object is reached (for example, the correlation coefficient reaches the maximum value or the root mean square error reaches the minimum value), which means that the motion compensation of the medical image has been completed, and it can be Achieve the motion compensation effect from the left picture (without reconstruction optimization) to the right picture (complete reconstruction optimization) as shown in Figure 11.
  • the motion compensation performed in step S80 is performed by integrating the motion compensation program into the maximum a posteriori expectation maximization (MAPEM) algorithm, which can be expressed as the following mathematical formula form:
  • MAPEM maximum a posteriori expectation maximization
  • p k represents the projection result of the kth frame image or gated group image of this medical image
  • i represents the index value of each pixel in the projection result
  • a k represents the kth frame image of this medical image or the system matrix for modeling the gated group image (that is, the new system matrix after the original system matrix is compensated and adjusted according to the movement, rotation, translation, or deformation information of the k-th frame image or the gated group image)
  • j represents The adjusted index value of each pixel in the projection result, Represents the partial derivative of the median square root prior energy function
  • ⁇ represents the prior adjustable factor
  • x current represents the reconstruction result of the medical image estimated by the current iterative cycle (that is, the optimized medical image of the current iterative cycle)
  • x next represents the next The reconstruction result of the estimated medical image in the iterative cycle (ie, the optimized medical image for the next iterative cycle).
  • the number of iterations for the optimal reconstruction of the preset medical image is based on 70 (that is, the maximum value of next is set to 70); however, more or more
  • the small number of iterations is not particularly limited in the present disclosure.
  • FIG. 12 depicts a schematic diagram of the effect after the optimization device 202 performs the above steps S10 to S80 on the medical image.
  • the optimization device 202 analyzes the medical image about the target organ (for example, the heart) into frame images (for example, as shown in legend 1201 corresponding to a certain respiratory cycle stage frame images and/or frame images corresponding to a certain heartbeat cycle stage as shown in the legend 1202), and then consider the motion curve of the target organ to compare each frame image (or the gated group image after gated) relative to the human body
  • the head-tail axis, left-right axis and/or ventral-dorsal axis are adjusted (for example, the frame image shown in legend 1203 is adjusted corresponding to a certain breathing cycle stage and/or the frame image shown in legend 1204 is adjusted corresponding to a certain heartbeat cycle stage), and then A reconstructed optimized medical image 1205 is obtained.
  • the optimization device 202 can directly display the reconstructed optimized medical image 1205 on the user interface of the management platform 201, or store the reconstructed optimized medical image 1205 in the medical image storage and transmission system 204 through the management platform 201, and then provide Physicians consult through the report computer 205 between consultations.
  • the present disclosure further provides a computer-readable medium, which is applied to a computer or computing device having a processor and/or memory, and stores instructions so that the computer or computing device can use the processor (for example, CPU, GPU, etc.) and and/or a memory, executing the above-mentioned method for motion detection and correction of medical images through instructions.
  • a computer or computing device having a processor and/or memory, and stores instructions so that the computer or computing device can use the processor (for example, CPU, GPU, etc.) and and/or a memory, executing the above-mentioned method for motion detection and correction of medical images through instructions.
  • the method, system and computer-readable medium for motion detection and correction of medical images in the present disclosure can be used to divide the medical images of target organs into multi-frame images according to the list mode data, and analyze the multi-frame images Multiple centroids of the region of interest in order to calculate the motion curve of the target organ during the scan, and then perform reconstruction optimization of medical images based on the motion curve, so that the human organs or lesions can be considered without installing additional monitoring equipment Motion to perform motion detection and correction on medical images.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • General Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Pathology (AREA)
  • Veterinary Medicine (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Optics & Photonics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Cardiology (AREA)
  • Epidemiology (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Primary Health Care (AREA)
  • Pulmonology (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Processing (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

一种医学影像的移动检测和校正的方法、系统和计算机可读介质，包括将关于目标器官的医学影像依照列表模式数据分割为多帧影像，以及分析该多帧影像中兴趣区域的多个质心，以计算得到该目标器官在扫描期间的运动曲线，并基于该运动曲线执行医学影像的重建优化，故能在不需安装额外监控设备的情况下，考虑人体器官或病灶的移动，以对医学影像进行移动检测和校正。

Description

医学影像的移动检测和校正方法、系统和计算机可读介质 技术领域
本公开涉及一种医学成像技术，尤其涉及医学影像的移动检测和校正的方法、系统和计算机可读介质。
背景技术
在精准医疗领域中，核医学由于具有可提供功能性评估的特点，属于不可或缺的项目。然而，核医学的检查时间长，且病人的身体或器官在扫描过程中的移动，常容易造成影像模糊失准及医师判断错误的问题。
以核医学的心肌灌注成像(myocardial perfusion imaging，MPI)为例，其为使用率最高，但最易受病人移动影响的检查项目。如图1所示，由于病人心脏本身的收缩跳动行为(如图例101所示，其由左至右分别示意在一心动周期，心脏在其短轴、垂直长轴和水平长轴等维度的切面形态示意图)以及人体呼吸的起伏情形(如图例102所示，其由左到右分别描述在一呼吸循环阶段，心脏在其短轴、垂直长轴和水平长轴等维度的切面形态示意图)，在这两者的影响下，所得心脏的MPI影像(如图例103所示)常常是模糊且不易诊断的。
在现行技术中，针对上述问题的解决方式为装设额外的监控设备，以监控人体的心脏跳动和呼吸位移，进而根据监控设备所测得的信号对医学影像(如，MPI影像)进行校正。然而，这种校正的方式不仅耗费检测资源，且可能发生监控设备设置不良、追踪错误或无法与扫描设备整合(例如，市面上单光子发射计算机断层扫描(single photon emission computerized tomography，SPECT)设备一般并无可选配的呼吸监测设备)等情况。
因此，如何在不需安装额外监控设备下，考虑人体器官或病灶的移动，以对医学影像进行移动检测和校正，已成为本领域亟需解决的课题之一。
发明内容
为解决上述问题，本公开提供一种医学影像的移动检测及校正的系统，包括：管理平台，用于提供使用者界面，以对关于目标器官的医学影像提交执行优化处理的指令，该医学影像包括列表模式数据；以及优化设备，用于根据该指令执行医学影像的优化处理，其中，该优化设备执行优化处理包括：将该医学影像所对应的列表模式数据分割为具有固定时间维度的帧，以将各帧成像为帧影像；在各帧影像中标注兴趣区域(volume of interest，VOI)，其中，各兴趣区域包含目标器官；根据各帧影像的各兴趣区域计算目标器官的运动曲线；根据该运动曲线将医学影像重建为优化医学影像；以及在使用者界面显示该优化医学影像。
在本公开系统的至少一个实施方案中，该优化设备包括深度学习模组，且该优化设备 执行在各帧影像中标注兴趣区域包括：通过该深度学习模组在各帧影像中识别包含目标器官的二元分割区域；通过该深度学习模组将各二元分割区域模糊化以生成软遮罩；通过该深度学习模组将各软遮罩套用于各帧影像；通过该深度学习模组基于各软遮罩以初始椭圆球面拟合各帧影像中的目标器官；以及通过该深度学习模组将各初始椭圆球面依其半径向外扩张预设距离以生成代表各兴趣区域的椭圆球面。
在本公开系统的至少一个实施方案中，该优化设备执行根据各帧影像的各兴趣区域计算目标器官的运动曲线包括：将各兴趣区域分割为第一子兴趣区域和第二子兴趣区域；分别提取各第一子兴趣区域的第一质心(center of mass，COM)和各第二子兴趣区域的第二质心的三维坐标以作为各帧影像的描述值；将各帧影像的各描述值以主成分分析法进行降维，并以降维后的各描述值的最大特征作为目标器官的移动/旋转信号；以及根据各移动/旋转信号对各帧影像分组并过滤以计算运动曲线。
在本公开系统的至少一个实施方案中，该目标器官为心脏，且将各兴趣区域分割为第一子兴趣区域和第二子兴趣区域是沿着心脏的长轴，以短轴方向进行。
在本公开系统的至少一个实施方案中，该运动曲线相对于人体头尾轴、左右轴和腹背轴中的任一者绘制，且优化设备执行根据该运动曲线将医学影像重建为优化医学影像包括：自各帧影像中选定基准对象；以该运动曲线为参考进行各帧影像的动作补偿，其中，各帧影像的动作补偿包括：相对于基准对象，将各帧影像中除基准对象外的各者所包含的所有像素依其三维座标在头尾轴、左右轴和腹背轴的任一者中执行调整作业；以及重复执行各帧影像的调整作业，直到经过该调整作业的各帧影像的集成与基准对象间的相关系数到达最大值为止；以及将经过动作补偿的各帧影像集成以重建为优化医学影像。在本公开的至少一个具体实施例中，该调整作业包括但不限于旋转、位移、缩放、形变或其中两者以上的任意组合；在本公开的一些具体实施例中，该调整作业为平移、旋转或其组合。
在本公开系统的至少一个实施方案中，在根据各质心计算目标器官的运动曲线之后，且各帧影像的各兴趣区域未准确对应运动曲线的情况下，该优化设备执行优化处理进一步包括执行各帧影像的门控(gating)，其中，该门控用于将各帧影像中时间和/或位置关系相近者集成为预设数量的门控组(gated set)影像，且其中该时间和/或位置关系以人体呼吸和/或心跳的循环阶段所定义。
在本公开系统的至少一个实施方案中，该运动曲线相对于人体头尾轴、左右轴和腹背轴中的任一者绘制，且该优化设备根据该运动曲线将医学影像重建为优化医学影像包括：自各门控组影像中选定基准对象；以该运动曲线为参考进行各门控组影像的动作补偿，其中，各门控组影像的动作补偿包括：相对于基准对象，将各门控组影像中除基准对象外的各者所包含的所有像素依其三维座标在头尾轴、左右轴和腹背轴的任一者中执行调整作业；及重复执行各门控组影像的调整作业，直到经过调整作业的各门控组影像的集成与基准对象间的相关系数到达最大值为止；以及将经过动作补偿的各门控组影像集成以重建为优化医学影像。
在本公开系统的至少一个实施方案中，该固定时间维度以100毫秒至500毫秒为单位。
在本公开系统的至少一个实施方案中，进一步包括：扫描设备，用于拍摄目标器官以获得医学影像，其中，该扫描设备为单光子发射计算机断层扫描设备、正电子发射断层扫描(positron emission tomography，PET)设备、磁共振成像(magnetic resonance imaging，MRI)设备、计算机断层扫描(computer tomography，CT)设备中的任一者；医学影像存储与传输系统(picture archiving and communication system，PACS)，用于存储该医学影像和该优化医学影像；以及诊间报告计算机，用于调阅或展示该医学影像和该优化医学影像。
本公开进一步提供一种医学影像的移动检测和校正的方法，包括：取得关于目标器官的医学影像，其中，该医学影像包括列表模式(list mode)数据；将该医学影像所对应的列表模式数据分割为具有固定时间维度的帧，以将各帧成像为帧影像；在各帧影像中标注兴趣区域，其中，各兴趣区域包含目标器官；根据各帧影像的各兴趣区域计算目标器官的运动曲线；以及根据该运动曲线将医学影像重建为优化医学影像。
在本公开方法的至少一个实施方案中，在各帧影像中标注兴趣区域包括：通过深度学习模组在各帧影像中识别包含目标器官的二元分割区域；通过该深度学习模组将各二元分割区域模糊化以生成软遮罩；通过该深度学习模组将各软遮罩套用于各帧影像；通过该深度学习模组基于各软遮罩以初始椭圆球面拟合各帧影像中的目标器官；以及通过该深度学习模组将各初始椭圆球面依其半径向外扩张预设距离以生成代表各兴趣区域的椭圆球面。
在本公开方法的至少一个实施方案中，根据各帧影像的各兴趣区域计算目标器官的运动曲线包括：将各兴趣区域分割为第一子兴趣区域和第二子兴趣区域；分别提取各第一子兴趣区域的第一质心和各第二子兴趣区域的第二质心的三维坐标以作为各帧影像的描述值；将各帧影像的各描述值以主成分分析法进行降维，并以降维后的各描述值的最大特征作为目标器官的移动/旋转信号；以及根据各移动/旋转信号对各帧影像分组并过滤以计算该运动曲线。
在本公开方法的至少一个实施方案中，该目标器官为心脏，且将各兴趣区域分割为第一子兴趣区域和第二子兴趣区域是沿着该心脏的长轴，以短轴方向进行。
在本公开方法的至少一个实施方案中，该运动曲线相对于人体头尾轴、左右轴和腹背轴中的任一者绘制，且根据该运动曲线将医学影像重建为优化医学影像包括：自各帧影像中选定基准对象；以该运动曲线为参考进行各帧影像的动作补偿，其中，各帧影像的该动作补偿包括：相对于基准对象，将各帧影像中除基准对象外的各者所包含的所有像素依其三维座标在头尾轴、左右轴和腹背轴的任一者中执行调整作业；以及重复执行各帧影像的调整作业，直到经过调整作业的各帧影像的集成与基准对象间的相关系数到达最大值为止；以及将经过动作补偿的各帧影像集成以重建为优化医学影像。
在本公开方法的至少一个实施方案中，进一步包括在根据各质心计算目标器官的运动曲线之后，且各帧影像的各兴趣区域未准确对应该运动曲线的情况下，执行各帧影像的门 控，其中，该门控用于将各帧影像中时间和/或位置关系相近者集成为预设数量的门控组影像，且其中，该时间和/或位置关系以人体呼吸及/或心跳的循环阶段所定义。
在本公开方法的至少一个实施方案中，该运动曲线相对于人体头尾轴、左右轴和腹背轴中的任一者绘制，且根据该运动曲线将医学影像重建为优化医学影像包括：自各门控组影像中选定基准对象；以该运动曲线为参考进行各门控组影像的动作补偿，其中，各门控组影像的动作补偿包括：相对于基准对象，将各门控组影像中除基准对象外的各者所包含的所有像素依其三维座标于该头尾轴、该左右轴及该腹背轴的任一者中执行调整作业；及重复执行各门控组影像的调整作业，直到经过调整作业的各门控组影像的集成与基准对象间的相关系数到达最大值为止；以及将经过动作补偿的各门控组影像集成以重建为优化医学影像。
在本公开方法的至少一个实施方案中，该医学影像为通过扫描设备拍摄目标器官而得，并且其中，该扫描设备为单光子发射计算机断层扫描设备、正电子发射断层扫描设备、磁共振成像设备及计算机断层扫描设备中的任一者。
在本公开方法的至少一个实施方案中，该固定时间维度以100毫秒至500毫秒为单位。
本公开进一步提供一种计算机可读存储介质，应用于计算机中，且存储有指令，以执行上述至少一种医学影像的移动检测和校正的方法。
综上所述，本公开的医学影像的移动检测和校正的方法、系统和计算机可读介质可将关于目标器官的医学影像依照列表模式数据分割为多帧影像，并分析该多帧影像中兴趣区域的多个质心，以计算得到目标器官在扫描期间的运动曲线，遂基于该运动曲线执行医学影像的重建优化，故能在不需安装额外监控设备下，考虑人体器官或病灶的移动，以对医学影像进行移动检测和校正。
附图说明
本案公开的具体实施例将搭配下列图式详述，这些说明显示在下列图式中：
图1描述以现行技术执行心肌灌注成像的实施方案；
图2描述本公开医学影像的移动检测和校正的系统架构示意图；
图3描述本公开医学影像的移动检测和校正的系统的实施例示意图；
图4描述本公开医学影像的移动检测和校正的方法的步骤流程图；
图5描述本公开医学影像的移动检测和校正的方法的部分实施方案；
图6A至图6C描述本公开医学影像的移动检测和校正的方法的部分实施方案，图6C中的COM表示质心；
图7描述本公开医学影像的移动检测和校正的方法的部分实施方案，COM1和COM2分别表示质心1和质心2；
图8描述本公开医学影像的移动检测和校正的方法的部分实施方案；
图9描述本公开医学影像的移动检测和校正的方法的部分实施方案；
图10描述本公开医学影像的移动检测和校正的方法的部分实施方案；
图11描述本公开医学影像的移动检测和校正的方法的部分实施方案；并且
图12描述本公开医学影像的移动检测和校正的方法的实施例示意图。
主要组件符号说明
101～103图例
201管理平台
202优化设备
203扫描设备
204医学影像存储与传输系统
205诊间报告计算机
300病患
301γ光子
302医学影像
303时间维度分割
304运动质心分析
305形变模型校正
306优化医学影像
501兴趣区域
601区域
602心肌区域
603二元分割区域
604兴趣区域
605、605’质心
701、702子兴趣区域
703、703’质心
704、704’质心
801、802、803情况
804、805效果
901、902、903曲线
1201、1202、1203、1204图例
1205优化医学影像
S10～S80步骤。
具体实施方式
以下通过特定的实施例说明本公开的实施方式，本领域技术人员可由本文所描述的内 容轻易地了解本案的其他优点及功效。本公开所附图式所绘示的结构、比例、大小等均仅用于配合说明书所描述的内容，以供本领域技术人员了解与阅读，非用于限定本公开可实施的限定条件，故任何修饰、改变或调整，在不影响本公开所能产生的功效和所能达成的目的下，均应仍落在本公开所描述的技术内容能够涵盖的范围内。
由图2可观察本公开执行医学影像的移动检测和校正的系统架构示意图。
在至少一个实施例中，本公开的管理平台201用于整合对医学影像的处理流程，包括：对医疗影像的接收和传输、提供使用者对医疗影像的调阅、及根据使用者的需求执行医疗影像的优化处理等。在一些实施例中，管理平台201可通过任意合适的网页、应用程序页面、人机界面等使用者界面呈现，在本文中并无特别限定。
在至少一个实施例中，本公开的优化设备202用于根据使用者在管理平台201处所提交的指令执行对应的医疗影像优化处理(包括，移动检测和校正)的后台服务。在一些实施例中，本公开的优化设备202可以是任意合适的实体计算机系统、云端系统等，且优化设备202也可与管理平台201以整合的计算机系统实现，在本公开中也无特别限定。
在至少一个实施例中，本公开的扫描设备203可以是任意可拍摄医学影像的检测设备，例如，包括但不限于：单光子发射计算机断层扫描设备、正电子发射断层扫描设备、磁共振成像设备、计算机断层扫描设备等，用于对病患所欲检测的部位(例如，包括但不限于：心脏、肺部、冠状动脉、肝、胃等)进行医学影像的取得。在一些实施例中，本公开的扫描设备203所取得的医学影像包括其对应的列表模式数据，故有助于对所拍摄的医学影像进行非即时地(例如，事后调阅)回朔分析校正。
在至少一个实施例中，本公开的医学影像存储与传输系统204可为任何当前医院所采用的存储用系统，用于存储上述扫描设备203处所取得的医学影像和/或经优化设备202优化处理的优化医学影像。
在至少一个实施例中，本公开的诊间报告计算机205可以是医师用于诊间的任意终端装置，用于提供医师调阅或展示医学影像存储与传输系统204所存储的医学影像和/或优化医学影像。
在本公开的至少一个实施例中，图2所示管理平台201、优化设备202、扫描设备203、医学影像存储与传输系统204和诊间报告计算机205配置为以标准医学数字成像与通信(Digital Imaging and Communications in Medicine，DICOM)向彼此进行通信传输，故能提供本公开高度的扩充性。另外，本公开并不仅限于上述各元件；举例而言，依据作业需求，可将上述各元件中的任意多者整合在同一装置中，或是设计使单个管理平台201和/或优化设备202可支援多台扫描设备203的医学影像的优化等，本公开对此均无特别限制。
图3描述本公开中执行医学影像的移动检测和校正的具体实施例示意图。
具体而言，在至少一个实施例中，本公开可以对使用单光子发射计算机断层扫描设备(SPECT)拍摄心脏的医学影像进行移动检测和校正。举例来说，本公开的扫描设备203(如，单光子发射计算机断层扫描设备)包括装设有19个针孔式准直仪和19个CdZnTe (CZT)传感器(例如，包含32×32像素的CZT元件)的CZT伽马相机，用于从右斜前视角向左斜后视角扫描心脏并获得相关的SPECT影像(即，医学影像)。此外，扫描设备203对SPECT影像的成像过程可以包括：设定分别以非对称式(例如，-14％至23％)和对称式(例如，-9％至9％)设置的双能窗(energy window)进行心脏的扫描；依据扫描结果存储SPECT影像对应的列表模式和/或帧模式(frame mode)数据；以标准医学数字成像和通信形式传送列表模式和/或帧模式数据至扫描设备203内建的工作站(workstation)；以及将列表模式和/或帧模式数据沿着心脏的短轴、垂直长轴和水平长轴进行重新采样以进行显示等步骤。然而，本公开中描述扫描设备203所适用的规格、装备、或取得医学影像的方式仅作为例示，并非旨在限定本公开的内容。
在图3所示的流程示意图中，在拍摄期间自(进行心肌灌注成像的)病患300释放的γ光子301经扫描设备203进行成像(如前述成像过程)后所得的医学影像如元件符号302所示。此时，医学影像302可直接存储至医学影像存储与传输系统204，以供诊间报告计算机205调阅。然而，在医学影像302可能因病患300在扫描期间移动而造成模糊的情况下，可通过管理平台203指定对接收自扫描设备203的医学影像302即时通过优化设备202进行移动检测和校正，包括时间维度分割303、运动质心分析304和形变模型校正305等处理程序，进而获得优化医学影像306。进一步地，当所取得医学影像302包括列表模式数据的情况下，也可在事后发现医学影像302有模糊情形时再通过管理平台201向医学影像存储与传输系统204存取此医学影像302，并通过优化设备202进行回朔分析校正。
图4描述本公开中执行医学影像的移动检测和校正(如优化设备202执行上述图3中303至305的处理程序)的步骤流程图，其各步骤的实施方案可通过图5至图12及以下说明逐步了解。
在本公开的至少一个实施例中，在步骤S10处，可通过管理平台201提供的使用者界面选取希望进行移动检测和校正的医学影像(例如，某病患在一次扫描下所获得的关于心脏的一组压力态(stress)影像和休息态(rest)影像)。此时，所述医学影像可在病患在扫描设备203执行扫描时即时通过管理平台201取得，或在需要时通过管理平台201向医学影像存储与传输系统204存取。
在本公开的至少一个实施例中，在步骤S20处，考量医学影像所拍摄目标器官(例如，心脏)在拍摄期间于不同呼吸及/或心跳循环阶段(或病患自身的移动)中的位移情形(即，事件)，优化设备202执行上述图3所示的时间维度分割303，将医学影像对应的列表模式数据所包含的事件(例如，利用前述的双能窗所捕捉的事件)切割为具有固定时间维度(例如，以500毫秒为单位)的帧，并将所述帧经由反投影(back-projection)至物体平面(例如，前述扫描设备203的19个CZT传感器的表面，其平行于对应的影像平面，且相交于多针孔系统的共同焦点)进行成像，如图5各帧的成像结果(后续也称为帧影像)所示。在一些实施例中，各帧可以100毫秒至500毫秒中的任意值为单位进行固定时间维度的切割；然而，根据优化设备202的运算能力或作业需求，也可将帧以其他合适的固定 时间维度切割，本公开对此并无特别限制。
在本公开的至少一个实施例中，在步骤S30处，如图5所示，优化设备202在各帧影像中标注包含目标器官(例如，心脏)的兴趣区域501，从而得知目标器官在各帧影像中的位置、旋转角度和/或轴(例如，心脏的长轴和短轴)等位移信息。在一些实施例中，包含目标器官的兴趣区域501为球形；然而，根据目标器官的形状或作业需求，也可将兴趣区域501以其他合适的形状进行标注，本公开对此并无特别限制。
在本公开的一些实施例中，以在帧影像中心脏的心肌区域标注兴趣区域为例，心肌区域的标注可由医师通过特定应用程序(例如，优化设备202内建构的自动化软件)进行，其实操步骤可包括：在帧影像(例如，前述步骤S20处获得者)中标注通过心肌中心的3D椭圆球面；自所述3D椭圆球面获得包含球面中心的三维座标、在三轴(即，心脏的短轴、垂直长轴和水平长轴等轴向)的半径、三轴旋转角度、心脏基平面的长短比例和角度等12个球面参数的数值；以及根据上述球面参数，由所述特定应用程序通过3D主动轮廓模型自医师标注的3D椭圆球面分别向内和向外拟合出平滑的内、外面，所完成3D椭圆球面的拟合的帧影像即为包含已定义心肌区域的兴趣区域的帧影像。
在本公开的一些实施例中，步骤S30对于兴趣区域501的标注也可以深度学习模组(例如，在优化设备202内建构深度学习模组)实现。在本实施例中，深度学习模组包括卷积层、反卷积层、泄露线性整流活化层、残余连接、跳动连接等主要结构，而上千组由医师标注心肌区域的帧影像(例如，前述通过特定应用程序标注者)用作深度学习模组的训练数据。完成训练且能在输入未经标注的帧影像时从中准确预测心肌区域的深度学习模组即可用于步骤S30对于兴趣区域501的标注。
图6A至图6C是将深度学习模组应用于标示兴趣区域501的阶段示意图。在本公开的至少一个实施例中，训练完成的深度学习模组首先可接收步骤S20处所产生的帧影像(如图6A中所示心脏在其短轴、垂直长轴和水平长轴等维度的切面画面)。在本公开的一些实施例中，深度学习模组会在帧影像中确定心脏的所在区域601(如图6A中所示以实线区域601圈选出距离心脏分割区域至少5厘米的正椭圆(旋转角度为零)的范围)，进而识别出包含心肌区域的二元分割区域603(如图6B中603所标示的区域)。接着，深度学习模组将该二元分割区域603进行模糊化以生成软遮罩，并进一步将该软遮罩套用于原始的帧影像，从而排除该帧影像中心肌区域以外区域的影像活性。最后，基于软遮罩，深度学习模组以一椭圆球面拟合帧影像上的心肌区域602(如图6B中602所标示实线对应的椭圆球面截线)，并将椭圆球面半径向外扩张2厘米以生成代表兴趣区域604的新椭圆球面(如图6C中604所标示的实线对应区域)。
在本公开的至少一个实施例中，在步骤S40处，优化设备202针对步骤S30所标注的兴趣区域内的目标器官执行上述图3所示的运动质心分析304。在本公开的一些实施例中，首先为准确观察目标器官(例如，心脏)在各帧影像中的位移和旋转情形，各帧影像中的兴趣区域将被分割为二份。可参考图6C与图7之间的差异，以观察分割兴趣区域的实施 方式。举例而言，图6C所标注各帧影像的兴趣区域604在此步骤S40中沿着心脏长轴分割为二份，以形成如图7所示关于心脏的子兴趣区域701、702，其中，由于图7左侧为心脏在其短轴维度的切面示图，故观察到的子兴趣区域701、702以重合方式呈现。接着，通过所得的目标器官的二个子兴趣区域701、702，可计算其各自的质心703、704，从而辅助追踪目标器官在每个帧影像的移动情形，进而获得目标器官在拍摄期间的运动曲线。
相较于图6C所示仅有一个质心605的兴趣区域604，图7所示将兴趣区域604分割为二个子兴趣区域701、702后再通过所获得的二个质心703、704进行运动质心分析的优点如图8所示。其中，801至803分别代表当目标器官(例如，心脏)平移(801)、平移且旋转(802)、旋转(803)等情况(即，从实线图案至虚线图案的变化)，而804、805则标示分别以单一质心605和二个质心703、704观察上述情况801至803的效果。由此可知，虽然在目标器官仅发生平移801的情况无论使用单一质心605或二个质心703、704均能清楚追踪(例如，质心605移动至605’的变化和质心703、704移动至703’、704’的变化)，但对于目标器官发生平移且旋转802或旋转803等情况，仅使用单一质心605所观察的追踪效果明显不如使用二个质心703、704所观察的追踪效果，尤其当目标器官仅发生旋转803的情况时，质心605在旋转前后的变化(与605’重合)并不明显，造成无法追踪目标器官旋转的情形。
然而，本公开对于运动质心分析的方式并不受限于除上述的说明；举例而言，本领域一般技术人员应能理解自兴趣区域604分割子兴趣区域701、702的方式也可沿着心脏短轴(依长轴方向)进行、或依照需求(例如，根据目标器官的形状特征)，将兴趣区域604分割为多于二个子兴趣区域，并以多于二个质心的方式观察目标器官的位移和旋转情形。
紧接在步骤S40之后，当兴趣区域604被分割为子兴趣区域701、702后，计算每个帧影像中质心703、704对应的三维座标(例如，以“(x1,y1,z1)、(x2,y2,z2)”表示)，以做为各帧影像的描述值。在本公开的一些实施例中，可使用主成分分析法(principle component analysis，PCA)对各帧影像的描述值进行降维，并从降维后的描述值中取得最大特征，以作为目标器官在各帧影像中的移动/旋转信号。在本公开的一些实施例中，还可根据目标器官的移动/旋转信号将各帧影像进行过滤，以将过于高频的杂信号滤除，进而计算得目标器官在拍摄期间的运动曲线。
在本公开的至少一个实施例中，目标器官在拍摄期间的运动曲线如图9所示，其描述检测目标器官(心脏)在拍摄期间相对于人体左右(X轴，图9上层的示意图)、腹背(Y轴，图9中层的示意图)、头尾(Z轴，图9下层的示意图)的运动曲线，其中，图9中901所示曲线(以点虚线表示)代表各帧影像中的质心703或704在上述X轴、Y轴与Z轴的相对位置变化，902所示曲线(以短线条虚线表示)代表根据901曲线所计算质心703或704在上述X轴、Y轴与Z轴的旋转信号，而903所示曲线(以实线表示)代表将各帧影像分组累加后所得质心703或704更为精确的位置变化。
在本公开的至少一个实施例中，在步骤S50处，为确保后续步骤对医学影像的校正的 效益，优化设备202可先根据前述目标器官的运动曲线判断目标器官的移动程度是否过大，在过大的情形(例如，目标器官的运动曲线的振幅大于50毫米时)下，放射师可要求病患重新进行一次扫描并重复进行步骤S10至S40，以取得目标器官新的运动曲线。若在不须重新扫描的情况下，则可接续后续的步骤，以进行医学影像的形变模型校正。
以上步骤S10至S50用于完成对医学影像的移动检测的程序，而依据移动检测的结果，优化设备202可接续对医学影像执行上述图3所示的形变模型校正305，如以下步骤S60至S80的说明。
鉴于前述运动曲线为所有帧影像中兴趣区域的位置(以质心为准)的平均数据，优化设备202先于步骤S60处确认各帧影像中兴趣区域的位置是否准确对应所述运动曲线，若各帧影像的兴趣区域并未准确对应运动曲线时，则可进行步骤S70的门控(gating)。反之，则可将各帧影像直接进行步骤S80的医学影像的优化重建。在本公开的一些实施例中，无论各帧影像的兴趣区域有无准确对应运动曲线，直接将各帧影像接续进行步骤S70的门控可帮助进一步掌握目标器官的移动情形，并降低步骤S80的医学影像的优化重建所耗费的时间。
在本公开的至少一个实施例中，步骤S70所述的门控用于根据步骤S40所计算目标器官(如，心脏)的运动曲线，而将各帧影像中时间和/或位置关系相近者进行集成，以形成门控组影像。
图10描述本公开中将关于心脏的医学影像的各帧影像整理为八个门控组的实施方案，其中，上排的八个门控组为以心脏的垂直长轴面向所观察的门控组影像，而下排的八个门控组则为以心脏的短轴面向所观察的门控组影像。另外，考量所述运动曲线可能对应至人体呼吸和/或心跳的规律运动，在扫描期间未出现病患剧烈运动的情形下，还可将此八个门控组影像对应于呼吸及/或心跳的各循环阶段，如图10所示从左到右为吸气/呼气的呼吸循环阶段，每张门控组影像相对于吸气结束阶段的门控组影像(最左侧的门控组影像)在人体头尾轴、左右轴和/或腹背轴上观察均有迭代的位移关系，故此步骤S70也称为呼吸门控或心跳门控。
在本公开的一些实施例中，图10示意性地描述自关于心脏的医学影像的所有帧影像依据人体呼吸循环阶段组合为八个门控组的实施方案(即，每个门控组包括帧影像总数的12.5％)；然而，根据医学影像的校正优化效果的需求或操作需要，也可将门控组数量增加或减少，或是另外考量人体心动周期组成额外一组门控组影像，或综合考量呼吸与心跳循环阶段组成一组门控组影像，其在本文中并无特别限定。
在本公开的至少一个实施例中，在步骤S80处，用于根据已定位准确的各帧影像(即，未经过步骤S70处理的医学影像的各帧影像)或经步骤S70的门控的门控组影像执行医学影像的重建优化。医学影像重建优化的执行主要以各帧影像或门控组影像中的基准对象(例如，判断处于吸气结束阶段、心动周期的等容收缩期、或符合上述二者的一帧影像或门控组影像)为标准进行其余帧影像或门控组影像的动作补偿，而上述相对于人体头尾轴、 左右轴和/或腹背轴的运动曲线用于在进行上述动作补偿时调整(包括旋转、位移、缩放、形变等调整作业)帧影像或门控组影像的参考，且经重建的优化医学影像与基准对象间的相关系数用于观察此医学影像的重建优化的完成程度。
举例来说，当考虑人体呼吸循环阶段进行关于心脏的医学影像的重建优化时，优化设备202自所述医学影像对应的帧影像或门控组影像中选定处于吸气结束阶段的一帧影像或门控组影像作为基准对象，并在参考运动曲线下，相对于所述基准对象，将其余帧影像或门控组影像所包含的所有像素，依其三维座标在人体头尾轴、左右轴和/或腹背轴中进行调整(即，前述的旋转、位移、缩放等调整作业)，进而将调整后的各帧影像或门控组影像集成以重建为优化医学影像，并迭代地执行所述医学影像的重建优化，直到优化医学影像与基准对象间达到最相似(例如，相关系数到达最大值或均方根误差达到最小值)为止，即代表此医学影像的动作补偿已完全，遂能达到如图11所示左侧图(未进行重建优化)到右侧图(完成重建优化)的动作补偿效果。
在本公开的一些实施例中，步骤S80所执行的动作补偿为通过将动作补偿程序融入最大后验期望最大(maximum a posteriori expectation maximization，MAPEM)算法的方式进行，其可表达为以下数学式的形式：
[数学式1]
Figure PCTCN2022136636-appb-000001
其中，p k代表关于此医学影像的第k个帧影像或门控组影像的投影结果，i代表所述投影结果中各像素的指数值，a k代表关于此医学影像的第k个帧影像或门控组影像的建模用系统矩阵(即，将原系统矩阵根据第k个帧影像或门控组影像的移动、旋转、平移、或形变信息补偿调整后的新系统矩阵)，j代表所述投影结果中各像素经调整后的指数值，
Figure PCTCN2022136636-appb-000002
代表中值方根先验能量函数的偏导数，β代表先验的可调整因子，x current代表当前迭代循环所估计医学影像的重建结果(即，当前迭代循环的优化医学影像)，而x next代表下个迭代循环所估计医学影像的重建结果(即，用于下个迭代循环的优化医学影像)。在本公开的一些实施例中，预设医学影像的优化重建的迭代次数以70次为准(即，next的最大值设定为70)；然而，也可视作业需求设定更多或更少的迭代次数，在本公开中并无特别限定。
图12描述优化设备202对医学影像执行上述步骤S10至S80后的效果示意图。举例而言，在本公开的至少一实施例中，优化设备202通过将关于目标器官(例如，心脏)的医学影像解析为帧影像的形式进行分析(例如，图例1201所示对应某一呼吸循环阶段的帧影像和/或图例1202所示对应某一心跳循环阶段的帧影像)，接着考量目标器官的运动 曲线，以将各帧影像(或进行门控后的门控组影像)相对于人体头尾轴、左右轴和/或腹背轴进行调整(例如，图例1203所示对应某一呼吸循环阶段调整的帧影像和/或图例1204所示对应某一心跳循环阶段调整的帧影像)，进而得到重建的优化医学影像1205。到此，优化设备202可直接将重建的优化医学影像1205显示在管理平台201的使用者界面，或是通过管理平台201将重建的优化医学影像1205存储至医学影像存储与传输系统204，进而供医师通过诊间报告计算机205调阅。
本公开进一步提供一种计算机可读介质，应用于具有处理器和/或存储器的计算机或计算装置中，其存储有指令，使计算机或计算装置可通过处理器(例如，CPU、GPU等)和/或存储器，通过指令执行如上所述医学影像的移动检测和校正的方法。
综上所述，本公开中医学影像的移动检测和校正的方法、系统和计算机可读介质可用于将关于目标器官的医学影像依照列表模式数据分割为多帧影像，并分析所述多帧影像中兴趣区域的多个质心，以计算得到目标器官在扫描期间的运动曲线，遂基于所述运动曲线执行医学影像的重建优化，故能在不需安装额外监控设备下，考虑人体器官或病灶的移动，以对医学影像进行移动检测和校正。
上述实施例仅例示性说明本公开的功效，而非用于限制本公开的范围，任何本领域技术人员均可在不违背本公开的范围下，对上述实施方案进行修饰与改变。因此本公开的权利保护范围，应如权利要求书所列。

Claims (19)

  1. 一种医学影像的移动检测和校正的系统，包括：
    管理平台，用于提供使用者界面，以对目标器官的医学影像提交执行优化处理的指令，其中，所述医学影像对应于列表模式数据；和
    优化设备，用于根据所述指令执行所述医学影像的所述优化处理，其中，所述优化处理包括：
    将所述医学影像所对应的所述列表模式数据分割为具有固定时间维度的多个帧，并将各所述帧成像为多帧影像；
    在各所述帧影像中标注兴趣区域以涵盖所述目标器官；
    根据各所述帧影像的所述兴趣区域计算所述目标器官的运动曲线；
    根据所述运动曲线重建所述医学影像；以及
    将经重建的所述医学影像显示在所述使用者界面。
  2. 如权利要求1所述的系统，其中，所述优化设备包括深度学习模组，且在各所述帧影像中标注兴趣区域包括：
    通过所述深度学习模组在各所述帧影像中识别包含所述目标器官的二元分割区域；
    通过所述深度学习模组将各所述二元分割区域模糊化以生成软遮罩；
    通过所述深度学习模组将各所述软遮罩套用于各所述帧影像；
    通过所述深度学习模组基于各所述软遮罩以初始椭圆球面拟合各所述帧影像中的所述目标器官；以及
    通过所述深度学习模组将各所述初始椭圆球面依其半径向外扩张预设距离以生成代表所述兴趣区域的椭圆球面。
  3. 如权利要求1所述的系统，其中，根据各所述帧影像的所述兴趣区域计算所述目标器官的运动曲线包括：
    将所述兴趣区域分割为第一子兴趣区域和第二子兴趣区域；
    分别提取所述第一子兴趣区域的第一质心和所述第二子兴趣区域的第二质心的三维坐标，以作为所述帧影像的描述值；
    将各所述帧影像的所述描述值以主成分分析法进行降维，并以降维后的各所述描述值的最大特征作为所述目标器官的移动和/或旋转信号；以及
    根据各所述移动和/或旋转信号对各所述帧影像分组并过滤以计算所述运动曲线。
  4. 如权利要求3所述的系统，其中，所述目标器官为心脏，且将各所述兴趣区域分割为所述第一子兴趣区域和所述第二子兴趣区域是沿着所述心脏的长轴，以短轴方向进行。
  5. 如权利要求1所述的系统，其中，所述运动曲线相对于人体的头尾轴、左右轴和腹背轴中的任一者绘制，且根据所述运动曲线重建所述医学影像包括：
    自各所述帧影像中选定基准对象；
    以所述运动曲线为参考进行各所述帧影像的动作补偿，其中，所述动作补偿包括：
    相对于所述基准对象，将各所述帧影像中除所述基准对象外的各者所包含的所有像素依其三维座标在所述头尾轴、所述左右轴和所述腹背轴中的任一者中执行调整作业；以及
    重复执行各所述帧影像的所述调整作业，直到经过所述调整作业的各所述帧影像的集成与所述基准对象间的相关系数到达最大值为止；以及
    将经过所述动作补偿的各所述帧影像集成以重建所述医学影像。
  6. 如权利要求3所述的系统，进一步包括在根据各所述质心计算所述目标器官的运动曲线之后，且各所述帧影像的各所述兴趣区域未准确对应所述运动曲线的情况下，所述优化设备执行所述优化处理另包括执行各所述帧影像的门控，其中，所述门控用于将各所述帧影像中时间和/或位置关系相近者集成为预设数量的门控组影像，且所述时间和/或位置关系为由人体呼吸和/或心跳的循环阶段所定义。
  7. 如权利要求6所述的系统，其中，所述运动曲线为相对于人体的头尾轴、左右轴和腹背轴中的任一者绘制，且根据所述运动曲线重建所述医学影像包括：
    自各所述门控组影像中选定基准对象；
    以所述运动曲线为参考进行各所述门控组影像的动作补偿，其中，各所述门控组影像的所述动作补偿包括：
    相对于所述基准对象，将各所述门控组影像中除所述基准对象外的各者所包含的所有像素依其三维座标在所述头尾轴、所述左右轴和所述腹背轴的任一者中执行调整作业；以及
    重复执行各所述门控组影像的所述调整作业，直到经过所述调整作业的各所述门控组影像的集成与所述基准对象间的相关系数到达最大值为止；以及
    将经过所述动作补偿的各所述门控组影像集成以重建所述医学影像。
  8. 如权利要求1所述的系统，其中，所述固定时间维度以100毫秒至500毫秒为单位。
  9. 如权利要求1所述的系统，进一步包括：
    扫描设备，用于拍摄所述目标器官以获得所述医学影像，其中，所述扫描设备选自由单光子发射计算机断层扫描设备、正电子发射断层扫描设备、磁共振成像设备和计算机断层扫描设备所组成的组中的任一者或其组合；
    医学影像存储与传输系统，用于存储所述医学影像和经重建的所述医学影像；以及
    诊间报告计算机，用于调阅或展示所述医学影像和经重建的所述医学影像。
  10. 一种医学影像的移动检测和校正的方法，包括：
    取得目标器官的医学影像，其中，所述医学影像对应于列表模式数据；
    将所述医学影像所对应的所述列表模式数据分割为具有固定时间维度的多个帧，并将各所述帧成像为多帧影像；
    在各所述帧影像中标注兴趣区域以涵盖所述目标器官；
    根据各所述帧影像的所述兴趣区域计算所述目标器官的运动曲线；以及
    根据所述运动曲线重建所述医学影像。
  11. 如权利要求10所述的方法，其中，在各所述帧影像中标注所述兴趣区域以涵盖所述目标器官包括：
    通过深度学习模组在各所述帧影像中识别包含所述目标器官的二元分割区域；
    通过所述深度学习模组将各所述二元分割区域模糊化以生成软遮罩；
    通过所述深度学习模组将各所述软遮罩套用于各所述帧影像；
    通过所述深度学习模组基于各所述软遮罩以初始椭圆球面拟合各所述帧影像中的所述目标器官；以及
    通过所述深度学习模组将各所述初始椭圆球面依其半径向外扩张预设距离，以生成代表所述兴趣区域的椭圆球面。
  12. 如权利要求10所述的方法，其中，根据各所述帧影像的所述兴趣区域计算所述目标器官的运动曲线包括：
    将所述兴趣区域分割为第一子兴趣区域和第二子兴趣区域；
    分别提取所述第一子兴趣区域的第一质心和所述第二子兴趣区域的第二质心的三维坐标以作为所述帧影像的描述值；
    将各所述帧影像的所述描述值以主成分分析法进行降维，并以降维后的所述描述值的最大特征作为所述目标器官的移动和/或旋转信号；以及
    根据各所述移动和/或旋转信号对各所述帧影像分组并过滤以计算所述运动曲线。
  13. 如权利要求12所述的方法，其中，所述目标器官为心脏，且将所述兴趣区域分割为所述第一子兴趣区域和所述第二子兴趣区域是沿着所述心脏的长轴，以短轴方向进行。
  14. 如权利要求10所述的方法，其中，所述运动曲线相对于人体的头尾轴、左右轴和腹背轴中的任一者绘制，且根据所述运动曲线重建所述医学影像包括：
    自各所述帧影像中选定基准对象；
    以所述运动曲线为参考进行各所述帧影像的动作补偿，其中，所述动作补偿包括：
    相对于所述基准对象，将各所述帧影像中除所述基准对象外的各者所包含的所有像素依其三维座标在所述头尾轴、所述左右轴和所述腹背轴中的任一者中执行调整作业；以及
    重复执行各所述帧影像的所述调整作业，直到经过所述调整作业的各所述帧影像的集成与所述基准对象间的相关系数到达最大值为止；以及
    将经过所述动作补偿的各所述帧影像集成以重建所述医学影像。
  15. 如权利要求12所述的方法，进一步包括在所述根据各所述质心计算所述目标器官的运动曲线之后，且各所述帧影像的各所述兴趣区域未准确对应所述运动曲线的情况下，执行各所述帧影像的门控，其中，所述门控用于将各所述帧影像中时间和/或位置关系相近者集成为预设数量的门控组影像，且所述时间和/或位置关系由人体呼吸及/或心跳的循环阶段所定义。
  16. 如权利要求15所述的方法，其中，所述运动曲线相对于人体的头尾轴、左右轴和腹背轴中的任一者绘制，且根据所述运动曲线重建所述医学影像包括：
    自各所述门控组影像中选定基准对象；
    以所述运动曲线为参考进行各所述门控组影像的动作补偿，其中，各所述门控组影像的所述动作补偿包括：
    相对于所述基准对象，将各所述门控组影像中除所述基准对象外的各者所包含的所有像素依其三维座标在所述头尾轴、所述左右轴和所述腹背轴中的任一者中执行调整作业；及
    重复执行各所述门控组影像的所述调整作业，直到经过所述调整作业的各所述门控组影像的集成与所述基准对象间的相关系数到达最大值为止；以及
    将经过所述动作补偿的各所述门控组影像集成以重建所述医学影像。
  17. 如权利要求10所述的方法，其中，所述医学影像由扫描设备拍摄所述目标器官所得，且所述扫描设备选自由单光子发射计算机断层扫描设备、正电子发射断层扫描设备、磁共振成像设备和计算机断层扫描设备所组成的组中的任一者或其组合。
  18. 如权利要求10所述的方法，其中，所述固定时间维度以100毫秒至500毫秒为单位。
  19. 一种计算机可读存储介质，其应用于计算机中且具有指令，以执行如权利要求10至18中任一项所述医学影像的移动检测和校正的方法。
PCT/CN2022/136636 2021-12-06 2022-12-05 医学影像的移动检测和校正方法、系统和计算机可读介质 WO2023103975A1 (zh)

Priority Applications (6)

Application Number Priority Date Filing Date Title
AU2022405390A AU2022405390A1 (en) 2021-12-06 2022-12-05 Medical image movement detection and correction method and system, and computer readable medium
CN202280080685.8A CN118339582A (zh) 2021-12-06 2022-12-05 医学影像的移动检测和校正方法、系统和计算机可读介质
CA3239117A CA3239117A1 (en) 2021-12-06 2022-12-05 Method, system and computer-readable medium for motion detection and correction of a medical image
EP22903397.2A EP4446973A1 (en) 2021-12-06 2022-12-05 Medical image movement detection and correction method and system, and computer readable medium
IL313372A IL313372A (en) 2021-12-06 2022-12-05 A method and system for detecting and correcting movement of a medical image, and a computer-readable medium
KR1020247022351A KR20240116526A (ko) 2021-12-06 2022-12-05 의료 영상의 움직임 검출 및 교정을 위한 방법, 시스템 및 컴퓨터 판독 가능 매체

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202163286133P 2021-12-06 2021-12-06
US63/286,133 2021-12-06

Publications (1)

Publication Number Publication Date
WO2023103975A1 true WO2023103975A1 (zh) 2023-06-15

Family

ID=86729606

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/136636 WO2023103975A1 (zh) 2021-12-06 2022-12-05 医学影像的移动检测和校正方法、系统和计算机可读介质

Country Status (7)

Country Link
EP (1) EP4446973A1 (zh)
KR (1) KR20240116526A (zh)
CN (1) CN118339582A (zh)
AU (1) AU2022405390A1 (zh)
CA (1) CA3239117A1 (zh)
IL (1) IL313372A (zh)
WO (1) WO2023103975A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101454801A (zh) * 2006-02-28 2009-06-10 皇家飞利浦电子股份有限公司 基于列表模式数据的局部运动补偿
CN101702232A (zh) * 2009-10-16 2010-05-05 昆明理工大学 正电子发射成像中呼吸校正技术
CN102067176A (zh) * 2008-06-18 2011-05-18 皇家飞利浦电子股份有限公司 结合局部运动监测、校正和评估的辐射成像
CN107569251A (zh) * 2017-08-29 2018-01-12 上海联影医疗科技有限公司 医学成像方法和系统及非暂态计算机可读存储介质
CN110619639A (zh) * 2019-08-26 2019-12-27 苏州同调医学科技有限公司 一种结合深度神经网络和概率图模型分割放疗影像的方法
US20210104037A1 (en) * 2019-10-07 2021-04-08 Siemens Medical Solutions Usa, Inc. Motion correction for medical image data

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101454801A (zh) * 2006-02-28 2009-06-10 皇家飞利浦电子股份有限公司 基于列表模式数据的局部运动补偿
CN102067176A (zh) * 2008-06-18 2011-05-18 皇家飞利浦电子股份有限公司 结合局部运动监测、校正和评估的辐射成像
CN101702232A (zh) * 2009-10-16 2010-05-05 昆明理工大学 正电子发射成像中呼吸校正技术
CN107569251A (zh) * 2017-08-29 2018-01-12 上海联影医疗科技有限公司 医学成像方法和系统及非暂态计算机可读存储介质
CN110619639A (zh) * 2019-08-26 2019-12-27 苏州同调医学科技有限公司 一种结合深度神经网络和概率图模型分割放疗影像的方法
US20210104037A1 (en) * 2019-10-07 2021-04-08 Siemens Medical Solutions Usa, Inc. Motion correction for medical image data

Also Published As

Publication number Publication date
KR20240116526A (ko) 2024-07-29
EP4446973A1 (en) 2024-10-16
AU2022405390A1 (en) 2024-06-06
IL313372A (en) 2024-08-01
CA3239117A1 (en) 2023-06-15
CN118339582A (zh) 2024-07-12

Similar Documents

Publication Publication Date Title
US12026832B2 (en) System and method for gating radiation exposure
US10699410B2 (en) Automatic change detection in medical images
US10937209B2 (en) Tomography imaging apparatus and method of reconstructing tomography image
US20230363832A1 (en) Mixed reality image guidance for medical interventions
US11950940B2 (en) System and method for determining radiation parameters
US8655040B2 (en) Integrated image registration and motion estimation for medical imaging applications
CN114943714A (zh) 医学图像处理系统、装置、电子设备及存储介质
CN110458779B (zh) 对呼吸或心脏的pet图像进行衰减校正的校正信息获取方法
US11495346B2 (en) External device-enabled imaging support
US11317875B2 (en) Reconstruction of flow data
WO2023103975A1 (zh) 医学影像的移动检测和校正方法、系统和计算机可读介质
TWI790023B (zh) 醫學影像的移動偵測及校正的方法、系統及電腦可讀媒介
CN114469153B (zh) 一种基于ct影像的血管造影装置、设备及计算机可读介质
CN110858412A (zh) 基于图像配准的心脏冠脉cta模型建立方法
Buliev et al. Estimation of the heart respiratory motion with applications for cone beam computed tomography imaging: a simulation study
Chambon et al. CT-PET landmark-based lung registration using a dynamic breathing model
James Visual Dissection and Motion Field Mosaics for Breathing Motion Correction in Nuclear Medicine Imaging

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: 22903397

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2022405390

Country of ref document: AU

Ref document number: AU2022405390

Country of ref document: AU

ENP Entry into the national phase

Ref document number: 3239117

Country of ref document: CA

ENP Entry into the national phase

Ref document number: 2024533246

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 202280080685.8

Country of ref document: CN

Ref document number: 313372

Country of ref document: IL

ENP Entry into the national phase

Ref document number: 2022405390

Country of ref document: AU

Date of ref document: 20221205

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20247022351

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2022903397

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2022903397

Country of ref document: EP

Effective date: 20240708

WWE Wipo information: entry into national phase

Ref document number: 11202403866W

Country of ref document: SG