WO2021077759A1 - 一种图像的匹配方法、装置、设备及存储介质 - Google Patents

一种图像的匹配方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2021077759A1
WO2021077759A1 PCT/CN2020/094875 CN2020094875W WO2021077759A1 WO 2021077759 A1 WO2021077759 A1 WO 2021077759A1 CN 2020094875 W CN2020094875 W CN 2020094875W WO 2021077759 A1 WO2021077759 A1 WO 2021077759A1
Authority
WO
WIPO (PCT)
Prior art keywords
image sequence
image
frame
sequence
lesion
Prior art date
Application number
PCT/CN2020/094875
Other languages
English (en)
French (fr)
Inventor
孙蒙蒙
王少康
陈宽
Original Assignee
推想医疗科技股份有限公司
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 推想医疗科技股份有限公司 filed Critical 推想医疗科技股份有限公司
Priority to JP2021548143A priority Critical patent/JP7190059B2/ja
Priority to EP20880082.1A priority patent/EP3910592A4/en
Publication of WO2021077759A1 publication Critical patent/WO2021077759A1/zh
Priority to US17/398,790 priority patent/US11954860B2/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G06T7/0014—Biomedical image inspection using an image reference approach
    • G06T7/0016—Biomedical image inspection using an image reference approach involving temporal comparison
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/38—Registration of image sequences
    • 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]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30061—Lung
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30096—Tumor; Lesion

Definitions

  • This application relates to the field of information processing technology, and more specifically, to an image matching method, device, device, and storage medium.
  • the images taken each time are a set of image sequences (such as CT image sequences) that include multiple image frames
  • image frames that include the same area (for CT image sequence, the area can be a scan layer) in the photographed tissue are called corresponding image frames
  • the corresponding relationship is the corresponding relationship between the image frames. For example, if the scan layer included in the 3rd frame of the last shot of the image sequence appears in the 5th frame of the last shot of the image sequence, then the 3rd frame of the last shot of the image sequence and this shot Corresponds to the 5th frame of the image sequence.
  • the method for determining the correspondence between image frames is to artificially set the difference of the image frame numbers according to experience, so as to correspondingly view the two image frames whose numbers match the difference.
  • human experience is based on the observation of tissues (such as lesions) to determine the third frame in the last image sequence shot, and the difference is set to 2 for the fifth frame of the image sequence shot next time.
  • Difference, the 8th frame of the image sequence taken this time corresponds to the 6th frame of the image sequence taken last time.
  • the image sequence taken each time is different. For example, changes in the internal organs of the patient (such as breathing movement) may cause the image frame distribution of the organs in different image sequences to be quite different, so it is suitable for between two image frames.
  • the difference between the two images may not necessarily apply to all image frames in the two affected sequences.
  • the 8th frame of the image sequence taken this time is likely to be the same as the scan included in the 6th frame of the image sequence taken last time.
  • the layers are different. Therefore, the accuracy of the corresponding relationship between the image frames obtained by artificially setting the difference value is low.
  • the present application provides an image matching method, device, device, and storage medium to obtain the corresponding relationship between image frames with high accuracy, as follows:
  • An image matching method including:
  • the registration result includes any pixel on the first object and one pixel on the second object.
  • mapping relationship is used to indicate the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence;
  • a control image frame is correspondingly displayed, and the control image frame is in the second image sequence and corresponds to the target image frame
  • the target image frame is any image frame in the first image sequence.
  • acquiring the first image sequence and the second image sequence includes:
  • receiving the first image sequence includes:
  • the first image sequence is acquired from the image sequence imaging device.
  • registering the first object and the second object to obtain a registration result includes:
  • the vertices of the first circumscribed polygon and the second circumscribed polygon are paired to obtain a matching point.
  • the first circumscribed polygon is the circumscribed polygon obtained by dividing the first object.
  • the polygon is each circumscribed polygon obtained by dividing the second object;
  • the registration result is obtained by using the least square method to solve the registration matrix equation.
  • the method further includes:
  • the first lesion information represents the diagnosis information of the lesion obtained based on the first image sequence
  • the second lesion information represents the diagnosis information of the lesion obtained based on the second image sequence Diagnostic information
  • the content of the same item in the first lesion information and the second lesion information is displayed.
  • the method further includes:
  • the second frame identifier is an identifier of an image frame where a lesion is located in the second image sequence
  • a first frame identifier is determined, the first frame identifier is the identifier of the image frame where the lesion is located in the first image sequence, and the first image The acquisition time of the sequence is later than the acquisition time of the second image sequence.
  • An image matching device including:
  • An image sequence obtaining unit configured to obtain a first image sequence and a second image sequence, the first image sequence and the second image sequence being a sequence of image frames collected from the same object;
  • An object acquisition unit for acquiring a first object and a second object, the first object is the object reconstructed using the first image sequence, and the second object is reconstructed using the second image sequence Said object;
  • the registration unit is used to register the first object and the second object to obtain a registration result.
  • the registration result includes any pixel on the first object, and the second object One-to-one correspondence of pixels;
  • a mapping relationship obtaining unit configured to obtain a mapping relationship according to the registration result, where the mapping relationship is used to indicate a corresponding relationship between image frames in the first image sequence and image frames in the second image sequence;
  • the image display unit is configured to display a control image frame corresponding to the mapping relationship in the case of displaying the target image frame in the first image sequence, where the control image frame is in the second image sequence, and The image frame corresponding to the target image frame, and the target image frame is any image frame in the first image sequence.
  • the image sequence obtaining unit is configured to obtain the first image sequence and the second image sequence, including: the image sequence obtaining unit is specifically configured to:
  • the image sequence acquiring unit configured to receive the first image sequence includes:
  • the image sequence obtaining unit is specifically configured to obtain the first image sequence from the image sequence imaging device when the image sequence imaging device generates the first image sequence.
  • the registration unit is configured to register the first object and the second object to obtain a registration result, including: the registration unit is specifically configured to:
  • the vertices of the first circumscribed polygon and the second circumscribed polygon are paired to obtain a matching point.
  • the first circumscribed polygon is the circumscribed polygon obtained by dividing the first object.
  • the polygon is each circumscribed polygon obtained by dividing the second object;
  • the registration result is obtained by using the least square method to solve the registration matrix equation.
  • the device further includes:
  • the lesion information acquiring unit is configured to acquire first lesion information and second lesion information; the first lesion information represents diagnosis information of the lesion obtained based on the first image sequence; the second lesion information represents the diagnosis information based on the first image sequence; 2. The diagnosis information of the lesion obtained from the image sequence;
  • the lesion information display unit is configured to correspondingly display the content of the same item in the first lesion information and the second lesion information.
  • the device further includes:
  • a second frame identification acquiring unit configured to acquire a second frame identification, where the second frame identification is an identification of an image frame where a lesion is located in the second image sequence;
  • a first frame identification acquisition unit configured to determine a first frame identification according to the second frame identification and the mapping relationship, where the first frame identification is the image where the lesion is located in the first image sequence
  • the identification of the frame, the acquisition time of the first image sequence is later than the acquisition time of the second image sequence.
  • An image matching device including: a memory and a processor
  • the memory is used to store programs
  • the processor is configured to execute the program to implement the steps of the image matching method described above.
  • the image matching method, device, equipment, and storage medium acquire the first image sequence and the second image sequence, and thus obtain the image sequence based on the first image sequence and the second image sequence, respectively.
  • Rebuild the generated first object and second object Since the first image sequence and the second image sequence are image frame sequences collected on the same object, both the first object and the second object are the results of collecting the object, that is, the first object and the second object have heights. Similar form. Therefore, the first object and the second object are further registered to obtain the registration result, and according to the registration result, the obtained mapping relationship can indicate the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence .
  • the method can correspondingly display the target image frame and the control image frame according to the mapping relationship.
  • the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence obtained by this image matching method greatly improves the accuracy of matching compared to artificially setting the difference.
  • FIG. 1 is a schematic flowchart of an image matching method provided by an embodiment of the application
  • FIG. 2 is a schematic flowchart of a method for registering a first object and a second object according to an embodiment of the application
  • FIG. 3 is a schematic structural diagram of an image matching device provided by an embodiment of the application.
  • FIG. 4 is a schematic structural diagram of an image matching device provided by an embodiment of the application.
  • Fig. 1 is a flowchart of an image matching method disclosed in an embodiment of the application, which specifically includes:
  • each image sequence includes a plurality of consecutive image frames.
  • the first image sequence and the second image sequence acquired in this step are image frame sequences acquired from the same object.
  • An optional acquisition method is to directly select the first image sequence and the second image sequence.
  • the first image sequence may be a follow-up CT image sequence of the patient
  • the second image sequence may be a historical CT image sequence of the patient
  • the historical CT image sequence may include multiple CT image sequence. It is understandable that each image sequence is a CT image sequence taken for the same part of the patient.
  • multiple CT image sequences of the patient can be retrieved from the PACS (picture archiving and communication system) directly according to the patient's information to perform subsequent image matching methods.
  • PACS picture archiving and communication system
  • an embodiment of the present application provides a method for automatically acquiring a first image sequence and a second image sequence, as follows:
  • A1. Receive the first image sequence.
  • the method of receiving the first image sequence may be, in the case that the image sequence imaging device generates the first image sequence, acquiring the first image sequence from the image sequence imaging device.
  • this method can monitor the dynamics of the imaging device by directly connecting with a front-end image sequence imaging device (such as a CT imaging device).
  • a front-end image sequence imaging device such as a CT imaging device.
  • image sequence imaging the image sequence sent by the imaging device is automatically received and used as the first image. sequence.
  • the method of receiving the first image sequence may be that this method monitors the storage dynamics of the image sequence by directly connecting with the PACS, and when a new image sequence is stored, it automatically receives the image sequence sent by the PACS device and uses it as The first image sequence.
  • A2. Collect an image sequence with the same identifier as the first image sequence from the historical image sequence as the second image sequence.
  • each image sequence has a unique identifier, which represents the user to which the image sequence belongs.
  • the identification may include the personal information (name, gender, age, etc.) of the patient to which the first image sequence belongs and the image type (abdominal CT, chest CT, head CT, etc.).
  • the historical image sequence is stored in the PACS, and each historical image sequence corresponds to a unique target, and the same type of image sequence shot at different time points.
  • other acquisition conditions for example, shooting time
  • the patient takes a follow-up CT image sequence under the guidance of a doctor.
  • This method directly collects the follow-up CT image sequence from the CT device as the first image sequence.
  • the personal information of the follow-up CT image sequence name Zhang San, gender male, age 30, and image type: abdominal CT.
  • the historical CT image sequence in the PACS that conforms to the above-mentioned personal information and the image type is collected as the second image sequence.
  • the selection condition as the acquisition time of two years, then the historical CT image sequences in the PACS that meet the above-mentioned personal information and image types are collected, and the historical CT image sequences within two years are selected from the above-mentioned historical CT image sequences. , As the second image sequence.
  • the first image sequence and the second image sequence are image frame sequences collected on the same object, so the first object is the object reconstructed using the first image sequence, and the second object is the second image sequence reconstructed The object.
  • the reconstructed first object and the second object are both the three-dimensional reconstruction results of the patient Zhang San's lungs.
  • this step may use traditional threshold method, watershed method, or deep learning-based segmentation algorithm to obtain the object area, and then reconstruct the object area to obtain the first object and the second object.
  • first object and the second object are the same object, so the shapes of the first object and the second object are highly similar.
  • the point of the first object can be represented by coordinates in the space established based on the first image sequence.
  • the points of the second object can be represented by coordinates in the space established based on the second image sequence.
  • any point in the first object can be registered with a point in the second object that is located at the same position of the object.
  • the registration result of the first object and the second object is obtained.
  • the registration result represents the registration relationship between the first object and the second object
  • the registration relationship is the spatial mapping relationship between the points in the first object and the second object (for any real point in the object, the registration The relationship is the corresponding relationship between the coordinate point of the point in the first object and the coordinate point in the second object). Since the image frame is a frame in the image sequence, the first object corresponds to a plurality of image frames in the first image sequence, and the second object corresponds to a plurality of image frames in the second image sequence. Therefore, in this step, the correspondence between each image frame in the first image sequence and each image frame in the second image sequence can be obtained according to the registration result by transforming the spatial mapping relationship and the physical mapping relationship.
  • transformation of the spatial mapping relationship and the physical mapping relationship can refer to an existing transformation method, which is not limited in the embodiment of the present application.
  • each image frame in the image sequence corresponds to a frame identifier
  • the frame identifier of the image frame indicates the position of the image frame in the image sequence. Therefore, the correspondence can be stored as a correspondence table of the frame identifiers of the image frames.
  • the target image frame is any image frame in the first image sequence
  • the control image frame is an image frame corresponding to the target image frame in the second image sequence.
  • the display of the target image frame in the first image sequence may be triggered to display any image frame in the first image sequence based on a user's operation, and the image frame is the target image frame.
  • the image frame corresponding to the target image frame can be obtained from the second image sequence according to the above-mentioned mapping relationship according to the method, that is, the control image frame. Furthermore, the target image frame and the control image frame are displayed.
  • the doctor can choose to display any image frame in the follow-up image sequence, and this method can determine the control image frame corresponding to the image frame in the previous image sequence based on the mapping relationship. Based on this, the image frame and the control image frame of the image frame are displayed at the same time.
  • the specific implementation of the display can be through connection with a preset display interface, which can provide the user to view each image frame in linkage, and has the function of operating (zooming in, zooming out, panning, etc.) on the image frame.
  • the image matching method, device, equipment, and storage medium acquire the first image sequence and the second image sequence, and thus obtain the image sequence based on the first image sequence and the second image sequence respectively.
  • Rebuild the generated first object and second object Since the first image sequence and the second image sequence are image frame sequences collected on the same object, both the first object and the second object are the results of collecting the object, that is, the first object and the second object have heights. Similar form. Therefore, the first object and the second object are further registered to obtain the registration result, and according to the registration result, the obtained mapping relationship can indicate the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence .
  • the method can correspondingly display the target image frame and the control image frame according to the mapping relationship.
  • the corresponding relationship between the image frame in the first image sequence and the image frame in the second image sequence can be obtained by the image matching method. Compared with the artificial setting of the difference, the accuracy of the matching is greatly improved.
  • FIG. 2 is a schematic flowchart of a method for registering a first object and a second object disclosed in an embodiment of the application.
  • the first image sequence and the second image sequence are abdominal CT image sequences, and the object is the lung as an example, to describe the registration process of the first object and the second object. It can include:
  • the first image sequence is the follow-up CT image sequence as DICOM1
  • the second image sequence is the previous CT image sequence as DICOM2.
  • the first object (lung) is denoted as LUNG1
  • the second object (lung) is denoted as LUNG2.
  • LUNG1 and LUNG2 can be equally divided into k parts along the long axis of the human body. It is understandable that since LUNG1 and LUNG2 are the same object, the i-th (1 ⁇ i ⁇ k) part of LUNG1 is highly similar to the i-th part of LUNG2.
  • the circumscribed polygons of the multiple objects obtained by dividing the first object and the second object are calculated.
  • the first circumscribed polygon may be the smallest enclosing cube of each part obtained by dividing the first object LUNG1
  • the second circumscribed polygon may be the smallest enclosing cube of each part obtained by dividing the second object LUNG2. It is known from the above that the i-th first circumscribed polygon and the i-th second circumscribed polygon are the same part of the same object, so that the vertices of each first circumscribed polygon and its corresponding second circumscribed polygon The vertices are paired. Get multiple matching points.
  • the first circumscribed polygon includes k pieces, denoted as LUNG1 1 , LUNG1 2 ,..., LUNG1 k
  • the first circumscribed polygon includes k pieces, denoted as LUNG2 1 , LUNG2 2 ,..., LUNG2 k .
  • LUNG1 1 and LUNG2 1 , LUNG1 2 and LUNG2 2 , ..., LUNG1 k and LUNG2 k are paired in pairs to obtain 8k pairs of matching points.
  • LUNG1 1 and LUNG2 1 LUNG1 1 includes eight vertices, the eight vertices corresponding to the vertex position 8 in LUNG2 1 paired off, to thereby obtain 8 pairs of matching points LUNG1 1 and LUNG2 1 .
  • the coordinates of the 8k vertices in each object can be arranged into a vertex matrix V with a size of 8k ⁇ 3, as follows:
  • x ij represents the coordinates of the j (1 ⁇ j ⁇ 8) vertex of the i (1 ⁇ i ⁇ k) circumscribed polygon in the object.
  • V tar is used to represent the vertex matrix of the first object, Represents the element in the rth row and the first column of the vertex matrix V tar (1 ⁇ r ⁇ 8k), that is, the x coordinate of the rth vertex in V tar, Represents the element in the r-th row and the second column of the vertex matrix V tar , that is, the y coordinate of the r-th vertex in V tar, Represents the element in the r-th row and the third column of the vertex matrix V tar , that is, the z coordinate of the r-th vertex in V tar.
  • the vertex matrix W of the first object is obtained by transforming the V tar matrix form as follows:
  • V org denote the vertex matrix of the second object, Represents the rth row of the vertex matrix V org.
  • the mapping relational expression is obtained by using 8k registration points of the first object and the second object through coordinate transformation, so the mapping relational expression can also represent the registration of two points located in the same position of the object in the first object and the second object relationship. That is, for any point in the first object, the coordinates of the registration point of the point in the second object can be calculated through the coordinates of the point in the first object and the mapping relationship.
  • this method can also display the lesion information of the first image sequence and the second image sequence, and the specific display method is:
  • the first lesion information indicates the diagnosis information of the lesion obtained based on the first image sequence
  • the second lesion information indicates the diagnosis information of the lesion obtained based on the second image sequence
  • Each lesion information includes multiple information items.
  • the information items of the diagnosis information may be physical diagnosis information of the lesion (size, shape, range, etc.), manual diagnosis information (the nature, degree, etc. of the lesion).
  • the information items in the diagnosis information can be obtained from the PACS, or can be obtained from other information systems such as the diagnosis system according to the identification of the image sequence.
  • first lesion information and the second lesion information will include the same information items.
  • first lesion information can be further correspondingly displayed The content of the same item as in the second lesion information.
  • a correspondence table of information items may be generated, and the same items in the table are arranged correspondingly, and different items may be arranged separately.
  • the corresponding relationship table is displayed.
  • this technical solution further provides a display method, which can simultaneously display two image frames with a mapping relationship in the first image sequence and the second image sequence, which is convenient for users to compare and observe the images in the image frames, and can correspond to By displaying the same items in the first lesion information and the second lesion information, multiple diagnosis results can be simply and clearly compared to obtain the change information of the lesion, thereby providing the user with a basis for further diagnosis, and further improving the accuracy of the diagnosis.
  • the method further includes:
  • the second frame identifier is acquired, and the first frame identifier is determined according to the second frame identifier and the mapping relationship.
  • the second frame identifier is the identifier of the image frame where the lesion is located in the second image sequence.
  • the first frame identifier is the identifier of the image frame where the lesion is located in the first image sequence. It should be noted that the acquisition time of the first image sequence is later than the acquisition time of the second image sequence.
  • the second image sequence is a historical image sequence
  • the specific location of the lesion in the second image sequence has been obtained through user diagnosis, that is, the identification of the image frame where the lesion is located in the second image sequence.
  • the first image sequence can be searched according to the second frame identifier, An image frame corresponding to the image frame indicated by the second frame identifier, and the frame identifier of the image frame is obtained as the frame identifier.
  • the image frame indicated by the first frame identifier can be directly selected to view the lesion. There is no need to manually view all image frames one by one to get the image frame where the lesion is located. This improves work efficiency.
  • this method can be applied to smart electronic devices, such as mobile phones, IPADs, computers, etc., and can be run in the form of independent software. At this time, it needs to be connected to a PACS system, other information systems or display systems. Or, it can be embedded in an existing system, for example, embedded in a PACS system.
  • the method can store the results of the mapping relationship, the first lesion information, the second lesion information, or the corresponding relationship table obtained in each of the foregoing embodiments.
  • the embodiment of the present application also provides an image matching device.
  • the image matching device provided in the embodiment of the present application will be described below.
  • the image matching device described below and the image matching method described above can be referenced correspondingly.
  • FIG. 3 shows a schematic structural diagram of an image matching device provided by an embodiment of the present application.
  • the device may include:
  • the image sequence obtaining unit 301 is configured to obtain a first image sequence and a second image sequence, where the first image sequence and the second image sequence are image frame sequences collected from the same object;
  • the object acquiring unit 302 is configured to acquire a first object and a second object, where the first object is the object reconstructed using the first image sequence, and the second object is reconstructed using the second image sequence The object obtained;
  • the registration unit 303 is configured to register the first object and the second object to obtain a registration result, where the registration result includes any pixel on the first object, and the second object One-to-one correspondence of pixels on the above;
  • the mapping relationship acquiring unit 304 is configured to obtain a mapping relationship according to the registration result, and the mapping relationship is used to indicate the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence ;
  • the image display unit 305 is configured to display a control image frame corresponding to the mapping relationship in the case of displaying the target image frame in the first image sequence, and the control image frame is in the second image sequence, An image frame corresponding to the target image frame, where the target image frame is any image frame in the first image sequence.
  • the image sequence obtaining unit is configured to obtain the first image sequence and the second image sequence, including: the image sequence obtaining unit is specifically configured to:
  • the image sequence acquiring unit configured to receive the first image sequence includes:
  • the image sequence obtaining unit is specifically configured to obtain the first image sequence from the image sequence imaging device when the image sequence imaging device generates the first image sequence.
  • the registration unit is configured to register the first object and the second object to obtain a registration result, including: the registration unit is specifically configured to:
  • the vertices of the first circumscribed polygon and the second circumscribed polygon are paired to obtain a matching point.
  • the first circumscribed polygon is the circumscribed polygon obtained by dividing the first object.
  • the polygon is each circumscribed polygon obtained by dividing the second object;
  • the registration result is obtained by using the least square method to solve the registration matrix equation.
  • the device further includes:
  • the lesion information acquiring unit is configured to acquire first lesion information and second lesion information; the first lesion information represents diagnosis information of the lesion obtained based on the first image sequence; the second lesion information represents the diagnosis information based on the first image sequence; 2. The diagnosis information of the lesion obtained from the image sequence;
  • the lesion information display unit is configured to correspondingly display the content of the same item in the first lesion information and the second lesion information.
  • the device further includes:
  • a second frame identification acquiring unit configured to acquire a second frame identification, where the second frame identification is an identification of an image frame where a lesion is located in the second image sequence;
  • a first frame identification acquisition unit configured to determine a first frame identification according to the second frame identification and the mapping relationship, where the first frame identification is the image where the lesion is located in the first image sequence
  • the identification of the frame, the acquisition time of the first image sequence is later than the acquisition time of the second image sequence.
  • FIG. 4 shows a schematic structural diagram of the image matching device.
  • the device may include: at least one processor 401, at least one communication interface 402, and at least one Memory 403 and at least one communication bus 404;
  • the number of the processor 401, the communication interface 402, the memory 403, and the communication bus 404 is at least one, and the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404;
  • the processor 401 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention, etc.;
  • CPU central processing unit
  • ASIC Application Specific Integrated Circuit
  • the memory 403 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), for example, at least one disk memory;
  • the memory stores a program
  • the processor can call the program stored in the memory, and the program is used for:
  • the registration result includes any pixel on the first object and one pixel on the second object.
  • mapping relationship is used to indicate the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence;
  • a control image frame is correspondingly displayed, and the control image frame is in the second image sequence and corresponds to the target image frame
  • the target image frame is any image frame in the first image sequence.
  • the embodiments of the present application also provide a storage medium, which can store a program suitable for execution by a processor, and the program is used for:
  • the registration result includes any pixel on the first object and one pixel on the second object.
  • mapping relationship is used to indicate the corresponding relationship between the image frames in the first image sequence and the image frames in the second image sequence;
  • a control image frame is correspondingly displayed, and the control image frame is in the second image sequence and corresponds to the target image frame
  • the target image frame is any image frame in the first image sequence.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mathematical Physics (AREA)
  • Computational Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Medical Informatics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Operations Research (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Analysis (AREA)

Abstract

一种图像的匹配方法、装置、设备及存储介质，获取第一影像序列以及第二影像序列，由此得到分别基于第一影像序列和第二影像序列重建生成的第一对象和第二对象；由于第一影像序列与第二影像序列为对同一对象采集的图像帧序列，因此第一对象和第二对象具有高度相似的形态；进一步配准第一对象和第二对象，依据配准结果，得到的映射关系可以指示第一影像序列中的图像帧与第二影像序列中的图像帧的对应关系。由所述图像的匹配方法得到第一影像序列中的图像帧与第二影像序列中的图像帧的对应关系，相比于人为设置差值，提高了匹配的准确度。

Description

一种图像的匹配方法、装置、设备及存储介质 技术领域
本申请涉及信息处理技术领域，更具体地说，涉及一种图像的匹配方法、装置、设备及存储介质。
发明背景
在医学影像领域，需要对照多次拍摄的医学影像序列中的图像帧，以做出更加准确的诊断。例如，病人在第一次被影像诊断出病灶之后，经临床医生诊断不需要立即做处理，要求病人一定时间内继续复查，病人在复查时，会拍摄随访影像，医生会基于两次影像中的病灶对比，再次做出诊断，以判断病人的患病情况。因为每次拍摄的影像均为一组包括多张图像帧的影像序列(例如CT影像序列)，所以在对比查看时，需要确定两组影像序列中图像帧的对应关系，通常，不同的影像序列中，包括被拍摄组织中的同一个区域(对于CT影像序列而言，区域可以为一个扫描层)的图像帧，称为对应的图像帧，对应关系即为图像帧间的对应关系。例如，上一次拍摄的影像序列中的第3帧中包括的扫描层，对应在该次拍摄的影像序列的第5帧中出现，则上一次拍摄的影像序列中的第3帧和该次拍摄的影像序列的第5帧对应。
目前，确定图像帧对应关系的方法为，根据经验人为设置图像帧编号差值，以对应查看编号符合差值的两张图像帧。例如，人为依据经验基于对组织(如病灶)的观察，确定上一次拍摄的影像序列中的第3帧，对应该次拍摄的影像序列的第5帧，则将差值设定为2，依据差值，该次拍摄的影像序列的第8帧，对应上一次拍摄的影像序列中的第6帧。但是，每次拍摄的影像序列具有差异性，例如，病人内部脏器的变化(例如呼吸运动)可能造成脏器在不同影像序列中的图像帧分布差异较大，所以适用于两个图像帧之间的差值，不一定适用于两个影响序列中所有图像帧，例如，该次拍摄的影像序列的第8帧，很有可能与上一次拍摄的影像序列中的第6帧中包括的扫描层不同。因此，由人为设置差值得到的图像帧的对应关系的准确度低。
发明内容
有鉴于此，本申请提供了一种图像的匹配方法、装置、设备及存储介质，以得到具有高准确度的图像帧的对应关系，如下：
一种图像的匹配方法，包括：
获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
可选地，获取第一影像序列以及第二影像序列，包括：
接收所述第一影像序列；
从历史影像序列中采集与所述第一影像序列具有相同标识的影像序列，作为所述第二影像序列。
可选地，接收所述第一影像序列包括：
在影像序列成像设备生成所述第一影像序列的情况下，从所述影像序列成像设备获取所述第一影像序列。
可选地，配准所述第一对象和所述第二对象，得到配准结果包括：
将所述第一对象和所述第二对象分别划分为多份；
将第一外接多边体和第二外接多边体的顶点进行配对，得到匹配点，所述第一外接多边体为所述第一对象划分得到的每一份的外接多边体，所述第二外接多边体为所述第二对象划分得到的每一份的外接多边体；
依据所述匹配点确定配准矩阵方程；
通过使用最小二乘法求解所述配准矩阵方程，得到所述配准结果。
可选地，本方法还包括：
获取第一病灶信息和第二病灶信息；所述第一病灶信息表示基于所述第一影像序列得到的病灶的诊断信息；所述第二病灶信息表示基于所述第二影像序列得到的病灶的诊断信息；
对应展示所述第一病灶信息和所述第二病灶信息中的相同项的内容。
可选地，本方法还包括：
获取第二帧标识，所述第二帧标识为病灶在所述第二影像序列中所处的图像帧的标识；
依据所述第二帧标识和所述映射关系，确定第一帧标识，所述第一帧标识为所述病灶在所述第一影像序列中所处的图像帧的标识，所述第一影像序列的获取时间，晚于所述第二影像序列的获取时间。
一种图像的匹配装置，包括：
影像序列获取单元，用于获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
对象获取单元，用于获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
配准单元，用于配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
映射关系获取单元，用于依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
图像展示单元，用于在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
可选地，影像序列获取单元用于获取第一影像序列以及第二影像序列，包括：所述影像序列获取单元具体用于：
接收所述第一影像序列；
从历史影像序列中采集与所述第一影像序列具有相同标识的影像序列，作为所述第二影像序列。
可选地，影像序列获取单元用于接收所述第一影像序列，包括：
所述影像序列获取单元具体用于，在影像序列成像设备生成所述第一影像序列的情况下，从所述影像序列成像设备获取所述第一影像序列。
可选地，配准单元用于配准所述第一对象和所述第二对象，得到配准结果，包括：所述配准单元具体用于：
将所述第一对象和所述第二对象分别划分为多份；
将第一外接多边体和第二外接多边体的顶点进行配对，得到匹配点，所述第一外接多边体为所述第一对象划分得到的每一份的外接多边体，所述第二外接多边体为所述第二对象划分得到的每一份的外接多边体；
依据所述匹配点确定配准矩阵方程；
通过使用最小二乘法求解所述配准矩阵方程，得到所述配准结果。
可选地，本装置还包括：
病灶信息获取单元，用于获取第一病灶信息和第二病灶信息；所述第一病灶信息表示基于所述第一影像序列得到的病灶的诊断信息；所述第二病灶信息表示基于所述第二影像序列得到的病灶的诊断信息；
病灶信息展示单元，用于对应展示所述第一病灶信息和所述第二病灶信息中的相同项的内容。
可选地，本装置还包括：
第二帧标识获取单元，用于获取第二帧标识，所述第二帧标识为病灶在所述第二影像序列中所处的图像帧的标识；
第一帧标识获取单元，用于依据所述第二帧标识和所述映射关系，确定第一帧标识，所述第一帧标识为所述病灶在所述第一影像序列中所处的图像帧的标识，所述第一影像序列的获取时间，晚于所述第二影像序列的获取时间。
一种图像的匹配设备，包括：存储器和处理器；
所述存储器，用于存储程序；
所述处理器，用于执行所述程序，实现如上所述的图像的匹配方法的各个步骤。
一种存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时，实现如上所述的图像的匹配方法的各个步骤。
从上述的技术方案可以看出，本申请提供的图像的匹配方法、装置、设备及存储介质，获取第一影像序列以及第二影像序列，由此得到分别基于第一影像序列和第二影像序列重建生成的第一对象和第二对象。由于，第一影像序列与第二影像序列为对同一对象采集的图像帧序列，因此，第一对象和第二对象均为采集该对象得到的结果，即，第一对象和第二对象具有高度相似的形态。所以，进一步配准第一对象和第二对象，得到配准结果，并依据配准结果，得到的映射关系可以指示第一影像序列中的图像帧与第二影像序列中的图像帧的对应关系。进一步，本方法可以依据该映射关系，对应展示目标图像帧以及对照图像帧。综上，由本图像的匹配方法得到第一影像序列中的图像帧与第二影像序列中的图像帧的对应关系，相比于人为设置差值，大大提高了匹配的准确度。
附图简要说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的实施例，对于本领域普通技术人员来讲，在不付出创 造性劳动的前提下，还可以根据提供的附图获得其他的附图。
图1为本申请实施例提供的图像的匹配方法的流程示意图；
图2为本申请实施例提供的一种配准第一对象和第二对象的方法流程示意图；
图3为本申请实施例提供的一种图像的匹配装置的结构示意图；
图4为本申请实施例提供的一种图像的匹配设备的结构示意图。
实施本发明的方式
本申请公开的技术方案，适用但不限于医学图像。
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
图1为本申请实施例公开的一种图像的匹配方法的流程图，具体包括：
S101、获取第一影像序列以及第二影像序列。
具体地，每一影像序列包括多个连续的图像帧。本步骤获取的第一影像序列与第二影像序列为对同一对象采集的图像帧序列。
获取第一影像序列和第二影像序列的方法可以包括多种。
可选的一种获取方法为直接选取第一影像序列和第二影像序列。
具体地，以医学影像中的CT影像序列为例，第一影像序列可以为病人的随访CT影像序列，第二影像序列可以为病人的历史CT影像序列，其中历史CT影像序列可以包括多次的CT影像序列。可以理解的是，每一影像序列均为针对病人的同一部位拍摄的CT影像序列。此时，可以直接根据病人的信息，从PACS(picture archiving and communication system，医学影像存档与通讯系统)中调取该病人的多次CT影像序列，以进行后续的图像的匹配方法。
可选地，本申请实施例提供了一种自动获取第一影像序列和第二影像序列的方法，如下：
A1、接收第一影像序列。
可选地，接收第一影像序列的方法可以为，在影像序列成像设备生成第一影像序列的情况下，从影像序列成像设备获取第一影像序列。
具体地，本方法可以通过与前端影像序列成像设备(如CT成像设备)直连监测成像设备的动态，当有影像序列成像时，自动接收成像设备发送的该影像序列，将其作为第一影像序列。
或者，接收第一影像序列的方法可以为，本方法通过与PACS直连，监测影像序列的存储动态，当有新的影像序列存入时，自动接收PACS设备发送的该影像序列，将其作为第一影像序列。
A2、从历史影像序列中采集与第一影像序列具有相同标识的影像序列，作为第二影像序列。
可以理解的是，每一影像序列都具有唯一的标识，该标识表征影像序列所属的用户。可选地，该标识可以包括第一影像序列所属的病人的个人信息(姓名、性别、年龄等)以及影像类型(腹部CT、胸部CT、头部CT等)。
其中，历史影像序列存储于PACS中，每一历史影像序列都对应于唯一的标，在不同时间点拍摄的同一类型的影像序列。
可选地，还可以预设其他采集条件(例如拍摄时间)，以进一步从多组历史影像序列中，筛选符合预期的第二影像序列。
以病人的随访影片序列为例，病人在医生的指导下拍摄随访CT影像序列，本方法直接从CT设备采集该随访CT影像序列，作为第一影像序列。同时得到该随访CT影像序列的个人信息：姓名张三、性别男、年龄30，以及影像类型：腹部CT。进一步，采集PACS中的符合上述个人信息、以及影像类型的历史CT影像序列，作为第二影像序列。进一步，可以预先设置筛选条件为采集时间两年，则，采集PACS中的符合上述个人信息、以及影像类型的历史CT影像序列，并在上述历史CT影像序列中筛选出两年内的历史CT影像序列，作为第二影像序列。
S102、获取第一对象和第二对象。
上述可知，第一影像序列与第二影像序列为对同一对象采集的图像帧序列，所以，第一对象为使用第一影像序列重建得到的该对象，第二对象为使用第二影像序列重建得到该对象。例如，基于病人张三的两次腹部CT影像序列，针对同一对象(例如肺部)，重建得到的第一对象和第二对象均为病人张三肺部的三维重建结果。
可选地，本步骤可以使用传统阈值法，分水岭法，或者基于深度学习的分割算法得到对象区域，再由对象区域重建，得到第一对象和第二对象。
S103、配准第一对象和第二对象，得到配准结果。
可以理解的是，第一对象和第二对象为同一对象，所以第一对象和第二对象的形态具有高度相似性。又因为，第一对象的点可以用基于第一影像序列建立的空间中的坐标表示。第二对象的点可以用基于第二影像序列建立的空间中的坐标表示。由此，可以基于空间坐标的变换，配准第一对象和第二对象中的各个点。
可选地，可以依据空间坐标变换方法，将第一对象中的任一点，与第二对象中的、位于对象相同位置的点进行配准。通过配准多个点，得到第一对象和第二对象的配准结果。
S104、依据配准结果，得到映射关系。
具体地，配准结果表征第一对象和第二对象的配准关系，该配准关系为第一对象和第二对象中的点的空间映射关系(对于对象中的任意一个真实点，配准关系为，该点在第一对象中的坐标点与在第二对象中的坐标点之间的对应关系)。由于，图像帧为影像序列中的一帧画面，第一对象对应于第一影像序列中的多个图像帧，第二对象对应于第二影像序列中的多个图像帧。所以，本步骤可以通过空间映射关系和物理映射关系的变换，依据配准结果，得到第一影像序列中的每一帧图像帧以及第二影像序列中的每一帧图像帧的对应关系。
需要说明的是，空间映射关系和物理映射关系的变换可参照现有的变换方法，本申请实施例对此不做限定。
可选地，影像序列中的每一图像帧都对应于一个帧标识，图像帧的帧标识表示该图像帧在影像序列中的位置。所以，该对应关系可以存储为图像帧的帧标识的对应关系表。
S105、在展示第一影像序列中的目标图像帧的情况下，依据映射关系，对应展示对照图像帧。
其中，目标图像帧为第一影像序列中的任一图像帧，对照图像帧为第二影像序列中，与目标图像帧对应的图像帧。
可选地，展示第一影像序列中的目标图像帧的情况可以为，基于用户的操作触发展示第一影像序列中的任一图像帧，该帧图像帧即为目标图像帧。此时，可以依据本方法得到上述映射关系从第二影像序列中获取与目标图像帧对应的图像帧，即为对照图像帧。进一步，展示目标图像帧以及对照图像帧。
例如，医生在获取病人的随访影像序列后，需要对照查看随访影像序列以及前次影像序列，以得到更准确地诊断。此时，医生可以选择展示随访影像序列中的任一图像帧，本方法可以基于映射关系，确定前次影像序列中与该图像帧对应的对照图像帧。基于此，同时展示该图像帧以及该图像帧的对照图像帧。
需要说明的是，展示的具体实现方式可以为通过与预设的显示界面连接，该显示界面可以提供用户联动查看各个图像帧，并具有对图像帧的操作(放大、缩小、平移等)功能。
从上述的技术方案可以看出，本申请提供的图像的匹配方法、装置、设备及存储介质，获取第一影像序列以及第二影像序列，由此得到分别基于第一影像序列和第二影像序列重建生成的第一对象和第二对象。由于，第一影像序列与第二 影像序列为对同一对象采集的图像帧序列，因此，第一对象和第二对象均为采集该对象得到的结果，即，第一对象和第二对象具有高度相似的形态。所以，进一步配准第一对象和第二对象，得到配准结果，并依据配准结果，得到的映射关系可以指示第一影像序列中的图像帧与第二影像序列中的图像帧的对应关系。进一步，本方法可以依据该映射关系，对应展示目标图像帧以及对照图像帧，综上，由本图像的匹配方法得到第一影像序列中的图像帧与第二影像序列中的图像帧的对应关系，相比于人为设置差值，大大提高了匹配的准确度。
图2为本申请实施例公开的一种配准第一对象和第二对象的方法流程示意图。本申请实施例以第一影像序列和第二影像序列为腹部CT影像序列、对象为肺部为例，对第一对象和第二对象的配准过程进行说明。具体可以包括：
S201、将第一对象和第二对象分别划分为多份。
其中，第一影像序列为随访CT影像序列记为DICOM1，第二影像序列为前次CT影像序列记为DICOM2。将第一对象(肺部)表示为LUNG1，第二对象(肺部)表示为LUNG2。
本步骤可以将LUNG1和LUNG2分别沿人体长轴方向等分为k份。可以理解的是，由于LUNG1和LUNG2为同一对象，所以，LUNG1的第i(1≤i≤k)份与LUNG2的第i份形态高度相似。
S202、将第一外接多边体和第二外接多边体的顶点进行配对，得到匹配点。
具体地，计算上述第一对象与第二对象划分得到的多份对象的外接多边体。其中，第一外接多边体可以为第一对象LUNG1划分得到的每一份的最小包围立方体，第二外接多边体可以为第二对象LUNG2划分得到的每一份的最小包围立方体。上述已知，第i份第一外接多边体与第i份第二外接多边体为同一对象的相同部分，由此将每一份第一外接多边体的顶点和其对应的第二外接多边体的顶点进行配对。得到多个匹配点。
例如，第一外接多边体包括k个，记为LUNG1 1、LUNG1 2、…、LUNG1 k，以及第一外接多边体包括k个，记为LUNG2 1、LUNG2 2、…、LUNG2 k。
进一步，分别将LUNG1 1与LUNG2 1、LUNG1 2与LUNG2 2、...、LUNG1 k与LUNG2 k的中各个顶点两两配对，得到8k对匹配点。以LUNG1 1与LUNG2 1为例，LUNG1 1中包括8个顶点，将该8个顶点与LUNG2 1中的相应位置的8个顶点两两配对，由此得到LUNG1 1与LUNG2 1的8对匹配点。
S203、依据匹配点确定配准矩阵方程。
具体地，可将每一对象中的8k个顶点的坐标排列为8k×3大小的顶点矩阵V，如下：
Figure PCTCN2020094875-appb-000001
其中x ij表示对象中第i(1≤i≤k)份外接多边体的第j(1≤j≤8)个顶点坐标。
本步骤用V tar表示第一对象的顶点矩阵，
Figure PCTCN2020094875-appb-000002
表示顶点矩阵V tar的第r行第1列元素(1≤r≤8k)，即，V tar中的第r个顶点的x坐标，
Figure PCTCN2020094875-appb-000003
表示顶点矩阵V tar的第r行第2列元素，即，V tar中的第r个顶点的y坐标，
Figure PCTCN2020094875-appb-000004
表示顶点矩阵V tar的第r行第3列元素，即，V tar中的第r个顶点的z坐标。基于上述表示方法，将V tar变换矩阵形式得到第一对象的顶点矩阵W，如下：
Figure PCTCN2020094875-appb-000005
用V org表示第二对象的顶点矩阵，
Figure PCTCN2020094875-appb-000006
表示顶点矩阵V org的第r行。
可以假设映射关系式为12维向量T，T＝[A 1 b 1 A 2 b 2 A 3 b 3] T，其中，A 1＝[a 11 a 12 a 13]、A 2＝[a 21 a 22 a 23]、A 3＝[a 31 a 32 a 33]。基于坐标变换，对第一对象和第二对象进行配准，配准矩阵方程如下：
Figure PCTCN2020094875-appb-000007
S204、通过使用最小二乘法求解配准矩阵方程，得到配准结果。
具体地，使用最小二乘法求解上述配准矩阵方程，即可计算得到映射关系式T＝[A 1 b 1 A 2 b 2 A 3 b 3] T。
该映射关系式为利用第一对象和第二对象的8k个配准点经过坐标变换得到，所以该映射关系式也可表示第一对象和第二对象中位于对象相同位置的两个点的配准关系。也即，第一对象中的任一点，均可通过该点在第一对象中的坐标，以及映射关系式，计算得到该点的配准点的在第二对象中的坐标。
可选地，本方法还可以展示第一影像序列以及第二影像序列的病灶信息，具体的展示方法为：
首先，获取第一病灶信息和第二病灶信息。
其中，第一病灶信息表示基于第一影像序列得到的病灶的诊断信息；第二病灶信息表示基于第二影像序列得到的病灶的诊断信息；
每一病灶信息包括多个信息项。例如，诊断信息的信息项可以为病灶的物理诊断信息(大小、形态、范围等)、人工诊断信息(病灶性质、程度等)。诊断信息中的各信息项可以从PACS中获取，或者可以依据影像序列的标识，从其他的信息系统例如诊断系统中获取。
可以理解的是，第一病灶信息和第二病灶信息中会包括相同的信息项，为了便于用户直观的对比第一影像序列以及第二影像序列的各项信息，可以进一步对应展示第一病灶信息和第二病灶信息中的相同项的内容。
可选地，可以生成信息项的对应关系表，表中相同项对应排列，不同项单独排列即可。在基于用户的操作展示第一影像序列的第一病灶信息的情况下，展示该对应关系表。
综上，本技术方案进一步提供展示方法，可以同时展示第一影像序列以及第二影像序列中的具有映射关系的两个图像帧，便于用户对图像帧中的影像进行对比观察，并且，可以对应展示第一病灶信息和第二病灶信息中的相同项，可以简单清晰的对比多次的诊断结果，得出病灶的变化信息，由此提供用户进一步诊断的依据，进一步可以提高诊断的准确度。
当接收第一影像序列后，为了避免用户在多张图像帧中查找病灶导致的低效率，本方法还包括：
获取第二帧标识，并依据第二帧标识和映射关系，确定第一帧标识。
其中，第二帧标识为病灶在第二影像序列中所处的图像帧的标识。第一帧标识为病灶在第一影像序列中所处的图像帧的标识，需要说明的是，第一影像序列的获取时间，晚于第二影像序列的获取时间。
可以理解的是，第二影像序列为历史影像序列，该第二影像序列已经通过用户诊断得到其中病灶的具体位置，即，病灶在第二影像序列中所处的图像帧的标识。又因为，上述实施例获得了第一影像序列中的每一帧图像帧以及第二影像序列中的每一帧图像帧的对应关系，所以，可以依据第二帧标识查找第一影像序列中，与该第二帧标识指示的图像帧对应的图像帧，并得到该图像帧的帧标识，作为帧标识。
可以理解的是，可以直接选择第一帧标识指示的图像帧，即可查看病灶。不需要人工逐张查看所有图像帧以得到病灶所在的图像帧。由此提高工作效率。
需要说明的是，本方法可以应用于智能电子设备，例如手机、IPAD、计算机等，可以以独立软件的形式运行，此时需要连接PACS系统、其他信息系统或显示系统等。或者，可以嵌入现有的系统，例如，嵌入PACS系统。并且，本方法可以存储上述各个实施例中得到的映射关系、第一病灶信息、第二病灶信息、或对应关系表等结果。
本申请实施例还提供了一种图像的匹配装置，下面对本申请实施例提供的图像的匹配装置进行描述，下文描述的图像的匹配装置与上文描述的图像的匹配方法可相互对应参照。
请参阅图3，示出了本申请实施例提供的一种图像的匹配装置的结构示意图，如图3所示，该装置可以包括：
影像序列获取单元301，用于获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
对象获取单元302，用于获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
配准单元303，用于配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
映射关系获取单元304，用于依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
图像展示单元305，用于在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
可选地，影像序列获取单元用于获取第一影像序列以及第二影像序列，包括：所述影像序列获取单元具体用于：
接收所述第一影像序列；
从历史影像序列中采集与所述第一影像序列具有相同标识的影像序列，作为所述第二影像序列。
可选地，影像序列获取单元用于接收所述第一影像序列，包括：
所述影像序列获取单元具体用于，在影像序列成像设备生成所述第一影像序列的情况下，从所述影像序列成像设备获取所述第一影像序列。
可选地，配准单元用于配准所述第一对象和所述第二对象，得到配准结果，包括：所述配准单元具体用于：
将所述第一对象和所述第二对象分别划分为多份；
将第一外接多边体和第二外接多边体的顶点进行配对，得到匹配点，所述第一外接多边体为所述第一对象划分得到的每一份的外接多边体，所述第二外接多边体为所述第二对象划分得到的每一份的外接多边体；
依据所述匹配点确定配准矩阵方程；
通过使用最小二乘法求解所述配准矩阵方程，得到所述配准结果。
可选地，本装置还包括：
病灶信息获取单元，用于获取第一病灶信息和第二病灶信息；所述第一病灶信息表示基于所述第一影像序列得到的病灶的诊断信息；所述第二病灶信息表示基于所述第二影像序列得到的病灶的诊断信息；
病灶信息展示单元，用于对应展示所述第一病灶信息和所述第二病灶信息中的相同项的内容。
可选地，本装置还包括：
第二帧标识获取单元，用于获取第二帧标识，所述第二帧标识为病灶在所述第二影像序列中所处的图像帧的标识；
第一帧标识获取单元，用于依据所述第二帧标识和所述映射关系，确定第一帧标识，所述第一帧标识为所述病灶在所述第一影像序列中所处的图像帧的标识，所述第一影像序列的获取时间，晚于所述第二影像序列的获取时间。
本申请实施例还提供了一种图像的匹配设备，请参阅图4，示出了该图像的 匹配设备的结构示意图，该设备可以包括：至少一个处理器401，至少一个通信接口402，至少一个存储器403和至少一个通信总线404；
在本申请实施例中，处理器401、通信接口402、存储器403、通信总线404的数量为至少一个，且处理器401、通信接口402、存储器403通过通信总线404完成相互间的通信；
处理器401可能是一个中央处理器CPU，或者是特定集成电路ASIC(Application Specific Integrated Circuit)，或者是被配置成实施本发明实施例的一个或多个集成电路等；
存储器403可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)等，例如至少一个磁盘存储器；
其中，存储器存储有程序，处理器可调用存储器存储的程序，所述程序用于：
获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
可选的，所述程序的细化功能和扩展功能可参照上文描述。
本申请实施例还提供一种存储介质，该存储介质可存储有适于处理器执行的程序，所述程序用于：
获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
可选的，所述程序的细化功能和扩展功能可参照上文描述。
最后，还需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本申请。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本申请的精神或范围的情况下，在其它实施例中实现。因此，本申请将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (14)

  1. 一种图像的匹配方法，其特征在于，包括：
    获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
    获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
    配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
    依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
    在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
  2. 根据权利要求1所述的方法，其特征在于，所述获取第一影像序列以及第二影像序列，包括：
    接收所述第一影像序列；
    从历史影像序列中采集与所述第一影像序列具有相同标识的影像序列，作为所述第二影像序列。
  3. 根据权利要求1所述的方法，其特征在于，所述接收所述第一影像序列包括：
    在影像序列成像设备生成所述第一影像序列的情况下，从所述影像序列成像设备获取所述第一影像序列。
  4. 根据权利要求3所述的方法，其特征在于，所述配准所述第一对象和所述第二对象，得到配准结果包括：
    将所述第一对象和所述第二对象分别划分为多份；
    将第一外接多边体和第二外接多边体的顶点进行配对，得到匹配点，所述第一外接多边体为所述第一对象划分得到的每一份的外接多边体，所述第二外接多边体为所述第二对象划分得到的每一份的外接多边体；
    依据所述匹配点确定配准矩阵方程；
    通过使用最小二乘法求解所述配准矩阵方程，得到所述配准结果。
  5. 根据权利要求1至4中的任一项所述的方法，其特征在于，还包括：
    获取第一病灶信息和第二病灶信息；所述第一病灶信息表示基于所述第一影像序列得到的病灶的诊断信息；所述第二病灶信息表示基于所述第二影像序列得到的病灶的诊断信息；
    对应展示所述第一病灶信息和所述第二病灶信息中的相同项的内容。
  6. 根据权利要求1至4中的任一项所述的方法，其特征在于，还包括：
    获取第二帧标识，所述第二帧标识为病灶在所述第二影像序列中所处的图像帧的标识；
    依据所述第二帧标识和所述映射关系，确定第一帧标识，所述第一帧标识为所述病灶在所述第一影像序列中所处的图像帧的标识，所述第一影像序列的获取时间，晚于所述第二影像序列的获取时间。
  7. 一种图像的匹配装置，其特征在于，包括：
    影像序列获取单元，用于获取第一影像序列以及第二影像序列，所述第一影像序列与所述第二影像序列为对同一对象采集的图像帧序列；
    对象获取单元，用于获取第一对象和第二对象，所述第一对象为使用所述第一影像序列重建得到的所述对象，所述第二对象为使用所述第二影像序列重建得到的所述对象；
    配准单元，用于配准所述第一对象和所述第二对象，得到配准结果，所述配准结果包括所述第一对象上的任意一个像素点，与所述第二对象上的像素点的一一对应关系；
    映射关系获取单元，用于依据所述配准结果，得到映射关系，所述映射关系用于指示所述第一影像序列中的图像帧与所述第二影像序列中的图像帧的对应关系；
    图像展示单元，用于在展示所述第一影像序列中的目标图像帧的情况下，依据所述映射关系，对应展示对照图像帧，所述对照图像帧为所述第二影像序列中，与所述目标图像帧对应的图像帧，所述目标图像帧为所述第一影像序列中的任一图像帧。
  8. 根据权利要求7所述的装置，其特征在于，所述影像序列获取单元用于获取第一影像序列以及第二影像序列，包括：所述影像序列获取单元具体用于：
    接收所述第一影像序列；
    从历史影像序列中采集与所述第一影像序列具有相同标识的影像序列，作为所述第二影像序列。
  9. 根据权利要求7所述的装置，其特征在于，所述影像序列获取单元用于接收所述第一影像序列，包括：
    所述影像序列获取单元具体用于，在影像序列成像设备生成所述第一影像序列的情况下，从所述影像序列成像设备获取所述第一影像序列。
  10. 根据权利要求9所述的装置，其特征在于，所述配准单元用于配准所述第一对象和所述第二对象，得到配准结果，包括：所述配准单元具体用于：
    将所述第一对象和所述第二对象分别划分为多份；
    将第一外接多边体和第二外接多边体的顶点进行配对，得到匹配点，所述第一外接多边体为所述第一对象划分得到的每一份的外接多边体，所述第二外接多边体为所述第二对象划分得到的每一份的外接多边体；
    依据所述匹配点确定配准矩阵方程；
    通过使用最小二乘法求解所述配准矩阵方程，得到所述配准结果。
  11. 根据权利要求7至10中的任一项所述的装置，其特征在于，还包括：
    病灶信息获取单元，用于获取第一病灶信息和第二病灶信息；所述第一病灶信息表示基于所述第一影像序列得到的病灶的诊断信息；所述第二病灶信息表示基于所述第二影像序列得到的病灶的诊断信息；
    病灶信息展示单元，用于对应展示所述第一病灶信息和所述第二病灶信息中的相同项的内容。
  12. 根据权利要求7至10中的任一项所述的装置，其特征在于，还包括：
    第二帧标识获取单元，用于获取第二帧标识，所述第二帧标识为病灶在所述第二影像序列中所处的图像帧的标识；
    第一帧标识获取单元，用于依据所述第二帧标识和所述映射关系，确定第一帧标识，所述第一帧标识为所述病灶在所述第一影像序列中所处的图像帧的标识，所述第一影像序列的获取时间，晚于所述第二影像序列的获取时间。
  13. 一种图像的匹配设备，其特征在于，包括：存储器和处理器；
    所述存储器，用于存储程序；
    所述处理器，用于执行所述程序，实现如权利要求1～6中任一项所述的图像的匹配方法的各个步骤。
  14. 一种存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时，实现如权利要求1～6中任一项所述的图像的匹配方法的各个步骤。
PCT/CN2020/094875 2019-10-21 2020-06-08 一种图像的匹配方法、装置、设备及存储介质 WO2021077759A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2021548143A JP7190059B2 (ja) 2019-10-21 2020-06-08 画像のマッチング方法、装置、デバイス及び記憶媒体
EP20880082.1A EP3910592A4 (en) 2019-10-21 2020-06-08 IMAGE MATCHING METHOD, APPARATUS AND DEVICE, AND STORAGE MEDIA
US17/398,790 US11954860B2 (en) 2019-10-21 2021-08-10 Image matching method and device, and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911001697.3 2019-10-21
CN201911001697.3A CN110766735B (zh) 2019-10-21 2019-10-21 一种图像的匹配方法、装置、设备及存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/398,790 Continuation US11954860B2 (en) 2019-10-21 2021-08-10 Image matching method and device, and storage medium

Publications (1)

Publication Number Publication Date
WO2021077759A1 true WO2021077759A1 (zh) 2021-04-29

Family

ID=69332872

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/094875 WO2021077759A1 (zh) 2019-10-21 2020-06-08 一种图像的匹配方法、装置、设备及存储介质

Country Status (5)

Country Link
US (1) US11954860B2 (zh)
EP (1) EP3910592A4 (zh)
JP (1) JP7190059B2 (zh)
CN (1) CN110766735B (zh)
WO (1) WO2021077759A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110766735B (zh) 2019-10-21 2020-06-26 北京推想科技有限公司 一种图像的匹配方法、装置、设备及存储介质
CN111462083B (zh) * 2020-03-31 2023-05-02 北京东软医疗设备有限公司 图像处理方法、装置、电子设备及存储介质
CN111462203B (zh) * 2020-04-07 2021-08-10 广州柏视医疗科技有限公司 Dr病灶演化分析装置和方法
CN111696649A (zh) * 2020-06-10 2020-09-22 杭州联众医疗科技股份有限公司 一种实现病种控费的手术前后影像数据比对的系统
CN112686866A (zh) * 2020-12-31 2021-04-20 安徽科大讯飞医疗信息技术有限公司 基于医学影像的随访方法、装置及计算机可读存储介质
CN115546174B (zh) * 2022-10-20 2023-09-08 数坤(北京)网络科技股份有限公司 图像处理方法、装置、计算设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106157282A (zh) * 2015-03-31 2016-11-23 深圳迈瑞生物医疗电子股份有限公司 图像处理系统及方法
CN106815832A (zh) * 2016-12-20 2017-06-09 华中科技大学 一种表面贴装技术的钢网图像自动配准方法和系统
CN107465940A (zh) * 2017-08-30 2017-12-12 苏州科达科技股份有限公司 视频对准方法、电子设备及存储介质
US20190197689A1 (en) * 2017-12-27 2019-06-27 Industrial Technology Research Institute Medical image comparison method and system thereof
CN110766735A (zh) * 2019-10-21 2020-02-07 北京推想科技有限公司 一种图像的匹配方法、装置、设备及存储介质

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050135707A1 (en) * 2003-12-18 2005-06-23 Turek Matthew W. Method and apparatus for registration of lung image data
JP4797173B2 (ja) * 2005-06-21 2011-10-19 国立大学法人金沢大学 Ｘ線診断支援装置、プログラム及び記録媒体
CN101616310B (zh) * 2009-07-17 2011-05-11 清华大学 可变视角及分辨率的双目视觉系统目标图像稳定化方法
US8848990B2 (en) * 2010-09-28 2014-09-30 Siemens Aktiengesellschaft Automatic registration of image series with varying contrast based on synthetic images derived from intensity behavior model
KR101932721B1 (ko) * 2012-09-07 2018-12-26 삼성전자주식회사 의료 영상들의 정합 방법 및 장치
CN104750951A (zh) * 2013-11-21 2015-07-01 上海联影医疗科技有限公司 一种医疗图像数据的分析处理方法及装置
CN103927746B (zh) * 2014-04-03 2017-02-15 北京工业大学 一种三维网格序列的配准及压缩方法
KR20160028273A (ko) * 2014-09-03 2016-03-11 삼성전자주식회사 병변 검출 보간 장치 및 방법
CN104881568A (zh) * 2015-04-27 2015-09-02 苏州敏宇医疗科技有限公司 一种基于云计算的肿瘤治疗疗效早期评价的系统和方法
CN106934807B (zh) * 2015-12-31 2022-03-01 深圳迈瑞生物医疗电子股份有限公司 一种医学影像分析方法、系统及医疗设备
US20190056498A1 (en) * 2016-03-01 2019-02-21 Brightway Vision Ltd. Gated imaging apparatus, system and method
JP2018011807A (ja) * 2016-07-22 2018-01-25 株式会社日立製作所 画像表示装置、ｘ線画像診断装置、及び画像表示方法
US10380763B2 (en) * 2016-11-16 2019-08-13 Seiko Epson Corporation Hybrid corner and edge-based tracking
CN107292872A (zh) * 2017-06-16 2017-10-24 艾松涛 图像处理方法/系统、计算机可读存储介质及电子设备
WO2019023382A1 (en) * 2017-07-26 2019-01-31 Canon U.S.A., Inc. METHOD FOR EVALUATING CARDIAC MOVEMENT USING ANGIOGRAPHIC IMAGE
CN107454389B (zh) * 2017-08-30 2019-04-23 苏州科达科技股份有限公司 待测系统的视频质量评价方法及系统
CN107595390B (zh) * 2017-10-19 2020-12-08 青岛大学附属医院 一种超声影像与ct影像的实时匹配融合方法
US11017548B2 (en) * 2018-06-21 2021-05-25 Hand Held Products, Inc. Methods, systems, and apparatuses for computing dimensions of an object using range images
CN109035234B (zh) * 2018-07-25 2020-12-01 腾讯科技（深圳）有限公司 一种结节检测方法、装置和存储介质
CN110021025B (zh) * 2019-03-29 2021-07-06 上海联影智能医疗科技有限公司 感兴趣区域的匹配和显示方法、装置、设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106157282A (zh) * 2015-03-31 2016-11-23 深圳迈瑞生物医疗电子股份有限公司 图像处理系统及方法
CN106815832A (zh) * 2016-12-20 2017-06-09 华中科技大学 一种表面贴装技术的钢网图像自动配准方法和系统
CN107465940A (zh) * 2017-08-30 2017-12-12 苏州科达科技股份有限公司 视频对准方法、电子设备及存储介质
US20190197689A1 (en) * 2017-12-27 2019-06-27 Industrial Technology Research Institute Medical image comparison method and system thereof
CN110766735A (zh) * 2019-10-21 2020-02-07 北京推想科技有限公司 一种图像的匹配方法、装置、设备及存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3910592A4 *

Also Published As

Publication number Publication date
EP3910592A1 (en) 2021-11-17
US20210366121A1 (en) 2021-11-25
JP2022520480A (ja) 2022-03-30
JP7190059B2 (ja) 2022-12-14
US11954860B2 (en) 2024-04-09
CN110766735B (zh) 2020-06-26
EP3910592A4 (en) 2022-05-11
CN110766735A (zh) 2020-02-07

Similar Documents

Publication Publication Date Title
WO2021077759A1 (zh) 一种图像的匹配方法、装置、设备及存储介质
JP7225295B2 (ja) 医用画像表示装置、方法およびプログラム
CN107133946B (zh) 医学图像处理方法、装置及设备
CN107123137B (zh) 医学图像处理方法及设备
US20220058821A1 (en) Medical image processing method, apparatus, and device, medium, and endoscope
JP5812841B2 (ja) 画像診断支援装置、画像診断支援方法、画像診断支援プログラム
Hernandez-Matas et al. Retinal image registration under the assumption of a spherical eye
JP6541334B2 (ja) 画像処理装置、画像処理方法、およびプログラム
JP6824845B2 (ja) 画像処理システム、装置、方法およびプログラム
Hernandez-Matas et al. Retinal image registration through simultaneous camera pose and eye shape estimation
KR20170069587A (ko) 영상처리장치 및 그의 영상처리방법
US7295951B2 (en) System and method for quantifying the quality of motion correction in image registration
JP6995535B2 (ja) 画像処理装置、画像処理方法およびプログラム
JP5599683B2 (ja) 医用画像処理装置、医用画像処理方法
JP7409624B2 (ja) 情報処理装置、情報処理方法、及びプログラム
US20220108525A1 (en) Patient-specific cortical surface tessellation into dipole patches
US10299864B1 (en) Co-localization of multiple internal organs based on images obtained during surgery
CN112819051A (zh) 胶囊内镜图像相似度评估方法、系统、终端及存储介质
CN110415239B (zh) 图像处理方法、装置、设备、医疗电子设备以及介质
CN114998239A (zh) 呼吸校正方法、计算机设备和存储介质
CN116612241A (zh) 三维重建方法、装置、设备及存储介质
CN116013465A (zh) 表面配准误差确定方法、装置、电子设备及可读存储介质
Tae-Eun A study of medical image analysis system
JP2022168985A (ja) 医用画像処理装置及び医用画像処理方法
JP6131149B2 (ja) 画像処理装置、方法およびプログラム

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: 20880082

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021548143

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2020880082

Country of ref document: EP

Effective date: 20210812

NENP Non-entry into the national phase

Ref country code: DE