Image quantization method and storage medium

By segmenting and converting the image in the region of interest, an automated quantization process is realized, which solves the problems of inaccurate and time-consuming quantization results in the prior art, and improves the accuracy and efficiency of image quantization.

CN111951316BActive Publication Date: 2024-11-22SHANGHAI UNITED IMAGING INTELLIGENCE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010729389.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-07-27
Publication Date
2024-11-22
Estimated Expiration
2040-11-22

Smart Images

  • Figure CN111951316B_ABST
    Figure CN111951316B_ABST
Patent Text Reader

Abstract

The present application relates to an image quantization method and storage medium. The method comprises: performing region of interest segmentation processing on the acquired reference original image and floating original image, determining a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image; determining a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image; performing conversion processing on the floating segmented image using the conversion relationship to obtain a converted segmented image; performing quantization processing on the reference segmented image and the converted segmented image respectively to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image; comparing the first quantization result and the second quantization result, and displaying the comparison result. The use of this method can improve the accuracy of image quantization results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of image processing, and in particular to an image quantization method and storage medium. Background Art

[0002] At present, in order to better analyze the patient's current images, the data on the patient's images taken at multiple periods are generally quantified, and then the quantification results of the images at multiple periods are post-processed and analyzed together to finally obtain the image analysis results of the current period.

[0003] In related technologies, doctors generally manually mark the areas of interest on the patient's images at multiple periods, and quantify the marked image data. They then post-process and comprehensively analyze the quantification results of multiple periods to obtain the analysis results of the patient's current period images.

[0004] However, the above technology has the problem that the quantization results obtained are inaccurate. Summary of the invention

[0005] Based on this, it is necessary to provide an image quantization method, device, computer equipment and storage medium that can improve the accuracy of image quantization results in response to the above technical problems.

[0006] An image quantization method, the method comprising:

[0007] Performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0008] Determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0009] The floating segmented image is transformed by using the transformation relationship to obtain a transformed segmented image;

[0010] The reference segmented image and the converted segmented image are quantized to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0011] The first quantization result and the second quantization result are compared, and the comparison result is displayed.

[0012] In one embodiment, before determining the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image, the method further includes:

[0013] Acquire a reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image;

[0014] The above-mentioned determining the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image includes:

[0015] A conversion relationship between the reference original image and the floating original image is determined according to the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image.

[0016] In one embodiment, the above-mentioned determining the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image, comprises:

[0017] Registering the reference original image and the floating original image to obtain a first conversion relationship;

[0018] Registering the reference segmentation image and the floating segmentation image, or registering the reference segmentation mask image and the floating segmentation mask image to obtain a second conversion relationship;

[0019] The first conversion relationship and the second conversion relationship are fused to determine the conversion relationship between the reference original image and the floating original image.

[0020] In one embodiment, the above-mentioned fusion processing of the first conversion relationship and the second conversion relationship to determine the conversion relationship between the reference original image and the floating original image includes:

[0021] Obtaining a first weight corresponding to the first conversion relationship and a second weight corresponding to the second conversion relationship;

[0022] The first conversion relationship and the second conversion relationship are fused according to the first weight and the second weight to determine the conversion relationship between the reference original image and the floating original image.

[0023] In one embodiment, the reference segmented image includes a first type of reference segmented image and a corresponding second reference segmented image, the converted segmented image includes a first type of floating segmented image and a corresponding second floating segmented image, and the reference segmented image and the converted segmented image are quantized respectively to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image, including:

[0024] Quantizing the first type of reference segmented image and the second reference segmented image to determine a first quantization result corresponding to the reference segmented image; the first quantization result is used to characterize the distribution and quality of the region of interest in the second reference segmented image;

[0025] The first type of floating segmented image and the second floating segmented image are quantized to determine a second quantization result corresponding to the floating segmented image; the second quantization result is used to characterize the distribution and quality of the region of interest in the second floating segmented image.

[0026] In one embodiment, the first type of reference segmented image includes a plurality of first reference segmented images, each of which includes a first region of interest, and the second reference segmented image includes a second region of interest, and the quantization processing of the first type of reference segmented image and the second reference segmented image to determine a first quantization result corresponding to the reference segmented image includes:

[0027] calculating the volume occupied by the second region of interest on each first region of interest;

[0028] Determine, according to the volume occupied by the second region of interest on each first region of interest, the volume percentage occupied by the second region of interest on each first region of interest and the mass of the region of interest corresponding to the second region of interest on each first region of interest;

[0029] Calculate the distribution histogram corresponding to the second region of interest according to the image value on the second region of interest;

[0030] The volume percentage, the quality of the region of interest and the distribution histogram are determined as the first quantitative result.

[0031] In one embodiment, the calculating the volume occupied by the second region of interest on each first region of interest includes:

[0032] Obtaining a reconstruction interval corresponding to the reference segmentation image, and determining a unit volume corresponding to each voxel in the reference segmentation image according to the reconstruction interval;

[0033] The intersection area between the second region of interest and each first region of interest is calculated, and the volume occupied by the second region of interest on each first region of interest is determined according to each intersection area and the unit volume.

[0034] In one embodiment, the determining, based on the volume occupied by the second region of interest on each first region of interest, the volume percentage occupied by the second region of interest on each first region of interest and the region of interest mass corresponding to the second region of interest on each first region of interest includes:

[0035] According to the unit volume, the volume of each first region of interest is calculated;

[0036] According to the volume occupied by the second region of interest on each first region of interest and the volume of each first region of interest, obtaining the volume percentage occupied by the second region of interest on each first region of interest;

[0037] The quality of the region of interest corresponding to the second region of interest on each first region of interest is determined according to the volume occupied by the second region of interest on each first region of interest and the image value on the second region of interest.

[0038] In one embodiment, the above-mentioned conversion process of the floating segmented image using the conversion relationship to obtain the converted segmented image includes:

[0039] Using the conversion relationship, each first reference segmented image is converted to obtain a floating segmented image corresponding to each first reference segmented image, and a first type of floating segmented image is obtained according to the floating segmented image corresponding to each first reference segmented image;

[0040] The second reference segmented image is transformed using the transformation relationship to obtain a second floating segmented image.

[0041] An image quantization device, comprising:

[0042] A segmentation module is used to perform region of interest segmentation processing on the acquired reference original image and floating original image, and determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0043] A conversion relationship determination module, used to determine the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0044] A conversion module, used for converting the floating segmented image using the conversion relationship to obtain a converted segmented image;

[0045] A quantization module, used to perform quantization processing on the reference segmented image and the converted segmented image respectively, to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0046] The comparison and display module is used to compare the first quantization result and the second quantization result and display the comparison result.

[0047] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0048] Performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0049] Determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0050] The floating segmented image is transformed by using the transformation relationship to obtain a transformed segmented image;

[0051] The reference segmented image and the converted segmented image are quantized to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0052] The first quantization result and the second quantization result are compared, and the comparison result is displayed.

[0053] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0054] Performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0055] Determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0056] The floating segmented image is transformed by using the transformation relationship to obtain a transformed segmented image;

[0057] The reference segmented image and the converted segmented image are quantized to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0058] The first quantization result and the second quantization result are compared, and the comparison result is displayed.

[0059] The above-mentioned image quantization method, device, computer equipment and storage medium perform region of interest segmentation processing on the acquired reference original image and floating original image to obtain corresponding reference segmented image and floating segmented image, determine the conversion relationship between the reference original image and the floating original image based on the reference original image and the floating original image, and use the conversion relationship to convert the floating segmented image to obtain a converted segmented image, perform quantization processing on the reference segmented image and the converted segmented image respectively to obtain a first quantization result and a second quantization result, compare the first quantization result and the second quantization result, and display the comparison result. In this method, since the process of segmenting, aligning and quantizing the reference original image and the floating original image is a completely automated process, it does not require manual labeling of the region of interest and manual calculation for quantification. Therefore, this method can save labor costs and time and reduce the workload of doctors. In addition, precisely because it does not require manual labeling and manual calculation for quantification, it can avoid some quantification errors caused by differences in experience of different doctors, thereby improving the objectivity and accuracy of the quantification results to a certain extent. Furthermore, since the two quantized images are also aligned before the quantification process, it can avoid quantification errors caused by misaligned positions, thereby making the final quantification results more accurate, so that the comparison of the quantification results will also be more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is an internal structure diagram of a computer device in one embodiment;

[0061] Figure 2 is a schematic diagram of a flow chart of an image quantization method in an embodiment;

[0062] Figure 3 is a schematic flow chart of an image quantization method in another embodiment;

[0063] Figure 4 is a schematic flow chart of an image quantization method in another embodiment;

[0064] Figure 5 FIG. 4 is a structural block diagram of an image quantization device in an embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] The image quantization method provided in the present application can be applied to a computer device, which can be a terminal or a server. Taking the computer device as a terminal as an example, its internal structure diagram can be as follows: Figure 1 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an image quantization method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0067] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0068] It should be noted that the above-mentioned computer device can be connected to any type of scanning device, or integrated inside any scanning device, to process the data collected by the scanning device.

[0069] In addition, the execution subject of the embodiment of the present application can be a computer device or an image quantization device. The following description will be made with the execution subject being a computer device.

[0070] In one embodiment, an image quantization method is provided. This embodiment involves the specific process of segmenting, registering, quantizing and comparing the reference original image and the floating original image. Figure 2 As shown, the method may include the following steps:

[0071] S202, performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image.

[0072] The reference original image and the floating original image may be images of the same part of the same object taken at different times, and may be plant images, animal images, human images, etc. The region of interest may be a region of interest or a point of interest, for example, the root or leaf of a plant; for example, an organ or a lesion on an organ, etc.

[0073] When performing segmentation of the region of interest, a pre-trained segmentation model can be used for segmentation. Taking a human organ as the region of interest as an example, the pre-trained segmentation model can be obtained by training based on a large number of sample images of the organ, the large number of sample images are used as the input of the initial segmentation model, and the organ position marked on each sample image is used as the reference output of the initial segmentation model. The trained segmentation model can be obtained by training the initial segmentation model.

[0074] After the segmentation model is trained, the reference original image and the floating original image can be input into the trained segmentation model, and the reference original image can be segmented into the region of interest, so that the reference segmentation image corresponding to the reference original image and the floating segmentation image corresponding to the floating original image can be obtained. Here, both the reference segmentation image and the floating segmentation image include the region of interest.

[0075] S204: Determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image.

[0076] In this step, key points can be selected on the reference original image and the floating original image respectively, and then similarity measurement is performed on each key point on the reference original image and the floating original image to obtain matching feature point pairs on the reference original image and the floating original image. Then, the spatial coordinate transformation parameters between the reference original image and the floating original image are obtained through the relative position relationship between the matching feature point pairs, which can also be called the spatial transformation relationship, that is, the conversion relationship here.

[0077] S206, performing conversion processing on the floating segmented image using the conversion relationship to obtain a converted segmented image.

[0078] In this step, after the conversion relationship is obtained as mentioned above, the conversion relationship can be used to convert each point on the floating segmented image, and the spatial position of each point on the floating segmented image is converted to the spatial position of each point on the reference segmented image to obtain the converted floating segmented image, which is recorded as the converted segmented image.

[0079] S208 , performing quantization processing on the reference segmented image and the converted segmented image respectively, to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image.

[0080] In this step, after obtaining the reference segmentation image and the converted segmentation image, the relevant information of the region of interest included in the reference segmentation image and the converted segmentation image can be obtained, for example, the number of voxels in the region of interest, the position of the region of interest, the number of regions of interest, the identification of the region of interest, the image value of the region of interest (such as HU value, Hounsfield Unit, Hounsfield unit), etc.

[0081] Afterwards, the volume, area, HU value histogram distribution and other related quantization parameters of the region of interest in the reference segmented image can be calculated based on the relevant information of the region of interest in the reference segmented image to achieve quantization of the region of interest in the reference segmented image, and the calculated related quantization parameters are used as the first quantization result.

[0082] Correspondingly, the volume, area, HU value histogram distribution and other related quantization parameters of the region of interest in the floating segmented image can be calculated based on the relevant information of the region of interest in the floating segmented image to achieve quantization of the region of interest in the floating segmented image, and the related quantization parameters calculated here are used as the second quantization result.

[0083] S210, comparing the first quantization result and the second quantization result, and displaying the comparison result.

[0084] Specifically, after obtaining the first quantization result corresponding to the region of interest in the reference segmented image and the second quantization result corresponding to the region of interest in the floating segmented image, the same type of quantization parameters in the first quantization result and the second quantization result can be compared to obtain the change trend of the region of interest.

[0085] Finally, each quantization parameter in the first quantization result, each quantization parameter in the second quantization result, and the change trend of each quantization parameter can be displayed on the interface.

[0086] In the above-mentioned image quantization method, by performing region of interest segmentation processing on the acquired reference original image and floating original image, a corresponding reference segmented image and floating segmented image are obtained, a conversion relationship between the reference original image and the floating original image is determined according to the reference original image and the floating original image, and the floating segmented image is converted using the conversion relationship to obtain a converted segmented image, the reference segmented image and the converted segmented image are quantized respectively to obtain a first quantization result and a second quantization result, the first quantization result and the second quantization result are compared, and the comparison result is displayed. In this method, since the process of segmenting, aligning and quantizing the reference original image and the floating original image is a completely automated process, it does not require manual labeling of the region of interest and manual calculation for quantification. Therefore, this method can save labor costs and time and reduce the workload of doctors. In addition, precisely because it does not require manual labeling and manual calculation for quantification, it can avoid some quantification errors caused by differences in experience of different doctors, thereby improving the objectivity and accuracy of the quantification results to a certain extent. Furthermore, since the two quantized images are also aligned before the quantification process, it can avoid quantification errors caused by misaligned positions, thereby making the final quantification results more accurate, so that the comparison of the quantification results will also be more accurate.

[0087] In another embodiment, another image quantization method is provided. This embodiment relates to a specific process of obtaining a segmentation mask image from an original image and obtaining a conversion relationship based on the segmentation mask image, the original image, and the segmented image. Figure 3 As shown, before the above S204, the above method may further include the following steps:

[0088] S302 : Acquire a reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image.

[0089] In this step, when the reference original image is segmented by the segmentation model to obtain the reference segmented image, the reference original image is usually segmented first to obtain the reference segmented mask image, and then the reference segmented mask image is further segmented to obtain the reference segmented image. When the reference segmented mask image is obtained during the segmentation process, the reference segmented mask image can be output and saved, so that the reference segmented mask image corresponding to the reference segmented image can be obtained. This operation is also performed on the floating original image to obtain the floating segmented mask image corresponding to the floating segmented image.

[0090] In addition, the reference segmentation mask image corresponding to the reference segmentation image may be one or more. Similarly, the floating segmentation mask image corresponding to the floating segmentation image may be one or more. Usually, the number of reference segmentation mask images and the number of floating segmentation mask images are equal and correspond one to one.

[0091] For example, assuming that the reference original image and the floating original image are original images of the lungs, then by scanning the lungs of a human or animal body at different times, a reference original image of the lungs and a floating original image of the lungs can be obtained, and then the lung regions (which may include lung fields, lung segments, and lung lobes) in the two lung images can be segmented to obtain corresponding lung segmentation mask images (such as lung field mask images, lung segment mask images, etc.), and then the lung segmentation mask images can be segmented for lung lesions to obtain segmented images of lung lesions, i.e., the reference segmentation image and the floating segmentation image here. The region of interest included in the reference segmentation image and the floating segmentation image here can be a lung lesion region.

[0092] After obtaining the reference segmentation mask image and the floating segmentation mask image, correspondingly, the above S204 may include the following steps S304:

[0093] S304, determining a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image.

[0094] In this step, that is, the conversion relationship can be obtained based on the combination of the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image; the conversion relationship can also be obtained based on the combination of the reference original image and the floating original image and the reference segmentation image and the floating segmentation image.

[0095] Optionally, the conversion relationship may be specifically obtained by adopting the following steps A1-A3:

[0096] A1, registering the reference original image and the floating original image to obtain a first conversion relationship.

[0097] Here, key points can be selected on the reference original image and the floating original image, and then similarity measurement can be performed on each key point on the reference original image and the floating original image to obtain matching feature point pairs on the reference original image and the floating original image. Next, the spatial coordinate transformation parameters between the reference original image and the floating original image are obtained through the relative position relationship between the matching feature point pairs, which is recorded as the first conversion relationship here.

[0098] A2: registering the reference segmentation image and the floating segmentation image, or registering the reference segmentation mask image and the floating segmentation mask image to obtain a second conversion relationship.

[0099] In this step, key points can be selected on the reference segmented image and the floating segmented image, and then similarity measurement is performed on each key point on the reference segmented image and the floating segmented image to obtain matching feature point pairs on the reference segmented image and the floating segmented image. Next, the spatial coordinate transformation parameters between the reference segmented image and the floating segmented image are obtained through the relative position relationship between the matching feature point pairs, which is recorded as the second transformation relationship here.

[0100] Alternatively, key points may be selected on the reference segmentation mask image and the floating segmentation mask image, and then similarity measurement may be performed on each key point on the reference segmentation mask image and the floating segmentation mask image to obtain matching feature point pairs on the reference segmentation mask image and the floating segmentation mask image. Next, the spatial coordinate transformation parameters between the reference segmentation mask image and the floating segmentation mask image are obtained through the relative position relationship between the matching feature point pairs, which is also recorded as the second transformation relationship here.

[0101] It should be noted that, taking the conversion relationship as a conversion matrix as an example, when the first conversion relationship and the second conversion relationship are conversion matrices, the conversion matrices corresponding to the two types of conversion relationships usually include the same number of rows and columns, for example, both are matrices with three rows and five columns.

[0102] A3, performing fusion processing on the first conversion relationship and the second conversion relationship to determine the conversion relationship between the reference original image and the floating original image.

[0103] In this step, different weights can be set in advance for different conversion relationships. Then, after the first conversion relationship and the second conversion relationship are obtained as described above, optionally, a first weight corresponding to the first conversion relationship and a second weight corresponding to the second conversion relationship can be obtained; based on the first weight and the second weight, the first conversion relationship and the second conversion relationship are fused to determine the conversion relationship between the reference original image and the floating original image.

[0104] Here, the first transformation relationship can be the deformation field between each point on the reference original image and each point on the floating original image; the second transformation relationship can be the deformation field between each point on the reference segmentation image and each point on the floating segmentation image, or it can be the deformation field between each point on the reference segmentation mask image and each point on the floating segmentation mask image.

[0105] In addition, the first conversion relationship is usually a deformation field, and the second conversion relationship can be one or more deformation fields. The deformation field includes the transformation vectors of each point on the original image, that is, the deformation field includes the transformation vectors of multiple points, and each point corresponds to the transformation vector one by one.

[0106] For example, taking the human lung as an organ, the lung may include a left lung and a right lung. Taking the second conversion relationship as an example, the corresponding reference segmentation mask image may be a left lung reference segmentation mask image and a right lung reference segmentation mask image, and the corresponding floating segmentation mask image may be a left lung floating segmentation mask image and a right lung floating segmentation mask image. A deformation field may be obtained by aligning the left lung reference segmentation mask image and the left lung floating segmentation mask image, and a deformation field may also be obtained by aligning the right lung reference segmentation mask image and the right lung floating segmentation mask image. That is to say, two deformation fields may be obtained here, and both deformation fields are the second conversion relationship.

[0107] Furthermore, the first weight here can be a first weight vector, the first weight matrix includes multiple first weight elements, each first weight element corresponds to the transformation vector of a point in the first transformation relationship. Similarly, the second weight can also be a second weight vector, the second weight matrix includes multiple second weight elements, each second weight element corresponds to the transformation vector of a point in the second transformation relationship.

[0108] Here, the first conversion relationship and the second conversion relationship are both a deformation field. When the first conversion relationship and the second conversion relationship are fused using the first weight and the second weight, that is, when the first conversion relationship and the second conversion relationship are fused using the first weight vector and the second weight vector, each first weight element can be multiplied by its corresponding conversion vector in the first conversion relationship. Similarly, each second weight element can be multiplied by its corresponding conversion vector in the second conversion relationship. After that, the two conversion vectors after multiplication of the points at the same position in the two conversion relationships can be summed and other operations can be performed to obtain the conversion vectors after the points at each position are fused, and the conversion vectors after the points at each position are fused are combined, which is the conversion relationship between the reference original image and the floating original image.

[0109] In addition, the size of each first weight element and each second weight element can be set according to actual conditions. Each first weight element can be equal or unequal, and similarly, each second weight element can be equal or unequal. The first weight element and / or the second weight element can be a number greater than 1 or a number less than or equal to 1.

[0110] For example, it is assumed that the first transformation relationship includes transformation vectors of three points, which are The first weight elements corresponding to the three points are 2, 3, and 4 respectively. Then, when fusion is performed, the first conversion relationship can be fused first to obtain the fused first conversion relationship: for Correspondingly, assuming that the second transformation relationship also includes the transformation vectors of three points, the corresponding ones are The second weight elements corresponding to the three points are 0.5, 0.7, and 0.9 respectively. Therefore, when fusion is performed, the second conversion relationship can be fused first to obtain the fused second conversion relationship for Then the conversion relationship between the reference original image and the floating original image It can be expressed as the following formula:

[0111]

[0112] Among them, a and b are weights of the first conversion relationship and the second conversion relationship respectively, and their sizes can be determined according to actual conditions, for example, both can be 0.5, or one can be 0.3 and the other can be 0.7, etc. In other words, each point within the first conversion relationship and the second conversion relationship has a weight, and the first conversion relationship and the second conversion relationship as a whole also have a weight.

[0113] The image quantization method provided in this embodiment can obtain the reference segmentation mask image corresponding to the reference segmentation image and the floating segmentation mask image corresponding to the floating segmentation image, and determine the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image. In this embodiment, since the conversion relationship between the segmentation mask image or the segmentation image can be combined on the basis of the conversion relationship of the original image, more conversion relationships can be combined, that is, the combined information is richer, so that the conversion relationship between the reference original image and the floating original image obtained in the end can be more accurate, and the accuracy of image quantization and comparison can be further improved.

[0114] In another embodiment, another image quantization method is provided. This embodiment involves the above-mentioned reference segmented image including a first type of reference segmented image and a corresponding second reference segmented image, and the converted segmented image including a first type of floating segmented image and a corresponding second floating segmented image. Then, how to perform quantization processing on the reference segmented image and the converted segmented image respectively to obtain the specific process of the first quantization result and the second quantization result. Based on the above embodiment, Figure 4 As shown, the above S208 may include the following steps:

[0115] S402, quantizing the first type of reference segmented image and the second reference segmented image to determine a first quantization result corresponding to the reference segmented image; the first quantization result is used to characterize the distribution and quality of the region of interest in the second reference segmented image.

[0116] In this step, optionally, the first type of reference segmented image includes multiple first reference segmented images, each first reference segmented image includes a first region of interest, and the second reference segmented image includes a second region of interest, wherein the first reference segmented image can be a first region of interest mask image, and the second reference segmented image can be a second region of interest mask image.

[0117] Taking the lung as an example, the first type of reference segmentation image here can be considered as a reference segmentation image of the lung area, which can be a mask image of the lung area. Then the first reference segmentation image can be a mask image of the lung field area (the first region of interest can be the whole lung area, the left lung area, and the right lung area), a mask image of the lobe area (the first region of interest can be the three lobe areas of the right lung and the two lobe areas of the left lung), and a mask image of the segment area (the first region of interest can be 10 segments of each of the left and right lungs).

[0118] The second reference segmented image may be a lung lesion reference segmented image, which may be a lung lesion region mask image, and the second region of interest may be the lung lesion region thereon.

[0119] After obtaining the first region of interest mask image and the second region of interest mask image, quantization processing can be performed. Optionally, the following steps B1-B4 can be used to obtain the first quantization result:

[0120] B1, calculating the volume occupied by the second region of interest on each first region of interest.

[0121] In this step, the volume occupied by the second region of interest on each first region of interest may be the volume of the intersection area of ​​the second region of interest and each first region of interest. Then, when calculating here, optionally, the reconstruction interval corresponding to the reference segmentation image may be obtained, and the unit volume corresponding to each voxel in the reference segmentation image may be determined according to the reconstruction interval; the intersection area between the second region of interest and each first region of interest may be calculated, and the volume occupied by the second region of interest on each first region of interest may be determined according to each intersection area and the unit volume.

[0122] The reconstruction interval here refers to the distance between each layer of scan data when the scan data is used for image reconstruction, which can be preset before imaging. Taking 3D as an example, the reconstruction interval here can include the reconstruction intervals in the x, y, and z directions, assuming that it is denoted as s x 、s y 、s z , then the unit volume v corresponding to each voxel here s It can be expressed by the product of these three reconstruction intervals, as shown in the following formula: s =s x *s y *s z .

[0123] In addition, after obtaining the mask images of the first regions of interest and the mask images of the second regions of interest as described above, the number of non-zero elements in the mask images can be counted, that is, the number of voxels in each first region of interest and the number of voxels in the second region of interest can be obtained. Then, by placing the first regions of interest and the second regions of interest in the same spatial position, the intersection area of ​​the first regions of interest and the second regions of interest can be obtained, and similarly, the number of voxels in the intersection area of ​​the first regions of interest and the second regions of interest can be obtained.

[0124] Afterwards, by multiplying the number of voxels in each intersection region by the unit volume, the volume of each intersection region, ie, the volume occupied by the second region of interest on each first region of interest, can be obtained.

[0125] B2, according to the volume occupied by the second region of interest on each first region of interest, determine the volume percentage occupied by the second region of interest on each first region of interest and the region of interest mass corresponding to the second region of interest on each first region of interest.

[0126] In this step, when calculating the volume percentage, optionally, the volume of each first region of interest can be calculated based on the unit volume; based on the volume occupied by the second region of interest on each first region of interest and the volume of each first region of interest, the volume percentage of the second region of interest on each first region of interest can be obtained.

[0127] For example, continuing to take the lung as an example, assuming that sum(mask) represents the number of non-zero elements in the first reference segmented image or the second reference segmented image, mask1*mask2 represents the intersection area of ​​the regions of interest in the two reference segmented images, assuming that the second region of interest mask is m0, and the first region of interest mask of the lung field is m 1i(i=1, 2, 3, for the whole lung, left lung, and right lung, respectively), the first region of interest mask of the lung lobe is m 2j (j=1-5, three lobes of the right lung and two lobes of the left lung respectively), the mask of the first region of interest in the lung segment is m 3k (k=1-20, 10 lung segments for each left and right lung respectively).

[0128] Then the volume of the first region of interest in each lung field can be calculated as v 1i =sum(m 1i )*v s , the volume of the first region of interest of each lung lobe can also be calculated as v 2j =sum(m 2j )*v s Similarly, the volume of the first region of interest of each lung segment can be calculated as v 3k =sum(m 3k )*v s .

[0129] Similarly, the volume of the second region of interest on the first region of interest in each lung field can be calculated as v 01i =sum(m0*m 1i )*v s , the volume of the second region of interest on the first region of interest of each lobe is v 02j =sum(m0*m 2j )*v s , the volume of the second region of interest on the first region of interest in each lung segment is v 03k =sum(m0*m 3k )*v s .

[0130] The volume occupied by the second region of interest on each first region of interest is ratioed to the volume of each first region of interest to obtain the volume percentage of the second region of interest on each first region of interest. In other words, the volume percentage of the second region of interest on the first region of interest in each lung field can be calculated as r 01i =sum(m0*m 1i )*v s / v 1i =sum(m0*m 1i ) / sum(m 1i ), that is, the ratio of the number of voxels in the intersection of the second region of interest and the first region of interest to the number of voxels in the first region of interest; similarly, the volume ratio of the second region of interest on the first region of interest in each lobe can also be calculated as r 02j =sum(m0*m2j )*v s / v 2j =sum(m0*m 2j ) / sum(m 2j ), similarly, the volume ratio of the second region of interest to the first region of interest in each lung segment can be calculated as r 03k =sum(m0*m 3k )*v s / v 3k =sum(m0*m 3k ) / sum(m 3k ).

[0131] Further, when calculating the quality of the region of interest, optionally, the quality of the region of interest corresponding to the second region of interest on each first region of interest can be determined based on the volume occupied by the second region of interest on each first region of interest and the image value on the second region of interest.

[0132] Continuing with the above example, assuming that the quality of the region of interest corresponding to the second region of interest on each first region of interest is M p According to the above example, the maximum value of p here can be the sum of the numbers of all first regions of interest, that is, 3+5+20=28, then the range of p is 1-28.

[0133] Here we calculate the mass M of the region of interest p When the image value can be the HU value, the following formula can be used for calculation:

[0134] M p =V p *[(A mp +1000)*0.001]

[0135] Among them, V p It can be the above v 1i 、v 2j 、v 3k , A mp When the intersection area of ​​the second region of interest and each first region of interest corresponds to the position of the reference original image, the average HU value of the image at that position can be obtained on the reference original image and is a known value. Then, the region of interest quality corresponding to the second region of interest on each first region of interest can be calculated by the above formula.

[0136] B3: Calculate and obtain a distribution histogram corresponding to the second region of interest according to the image value on the second region of interest.

[0137] In this step, the image value can be a HU value, so the position of the second region of interest can be corresponded to the corresponding position in the reference original image, so that the image value of each pixel at the corresponding position can be obtained. Then these image values ​​are counted to obtain the maximum and minimum values ​​of these image values, forming an image value range, and the image value range is divided into multiple segments, so that the number of pixels in each image value range can be obtained, and then the number of pixels in each segment is divided by the total number of pixels to obtain the probability of each segment. The probability of each segment is counted using a cubic graph tool to obtain the corresponding distribution histogram, that is, the distribution histogram corresponding to the second region of interest here.

[0138] B4, determining the volume percentage, the quality of the region of interest and the distribution histogram as the first quantification result.

[0139] After obtaining the volume percentage of the second region of interest on each first region of interest, the region of interest mass corresponding to the second region of interest on each first region of interest, and the distribution histogram corresponding to the second region of interest, these results can be used as the first quantization result corresponding to the reference segmented image.

[0140] S404, quantizing the first type of floating segmented image and the second floating segmented image to determine a second quantization result corresponding to the floating segmented image; the second quantization result is used to characterize the distribution and quality of the region of interest in the second floating segmented image.

[0141] In this step, before quantization, optionally, each first reference segmented image can be transformed using a transformation relationship to obtain a floating segmented image corresponding to each first reference segmented image, and a first type of floating segmented image can be obtained based on the floating segmented image corresponding to each first reference segmented image; the second reference segmented image can be transformed using the transformation relationship to obtain a second floating segmented image.

[0142] After the conversion, multiple first floating segmented images and second floating segmented images can be obtained, corresponding to the above S402, where each first floating segmented image includes the first region of interest, and the second floating segmented image includes the second region of interest, wherein the first floating segmented image can be a mask image of the first region of interest, and the second floating segmented image can be a mask image of the second region of interest.

[0143] Taking the lung as an example, the first type of floating segmentation image here can be considered as a floating segmentation image of the lung area, which can be a mask image of the lung area. Then the first floating segmentation image can be a mask image of the lung field area (the first region of interest can be the whole lung area, the left lung area, and the right lung area), a mask image of the lobe area (the first region of interest can be the three lobe areas of the right lung and the two lobe areas of the left lung), and a mask image of the segment area (the first region of interest can be the 10 segment areas of the left and right lungs).

[0144] The second floating segmentation image may be a lung lesion floating segmentation image, which may be a lung lesion region mask image, and the second region of interest may be the lung lesion region thereon.

[0145] Furthermore, the method of S402 described above may also be used. The specific process will not be repeated here. In short, the volume percentage of the second region of interest in each first region of interest in the floating segmented image, the region of interest mass corresponding to the second region of interest on each first region of interest, and the distribution histogram corresponding to the second region of interest may be calculated. These results may then be used as the second quantization result corresponding to the floating segmented image.

[0146] After obtaining the first quantitative result and the second quantitative result as mentioned above, the same indicators (such as volume percentage, etc.) in the two types of quantitative results can be compared, and the rate of change of the indicator can be calculated to indicate the changing trend of the indicator. Finally, the quantitative results, changing trends and other indicators can also be displayed on the interface.

[0147] The image quantization method provided in this embodiment can perform quantization processing on the first type of reference segmented image and the second reference segmented image to determine the first quantization result corresponding to the reference segmented image, and perform quantization processing on the first type of floating segmented image and the second floating segmented image to determine the second quantization result corresponding to the floating segmented image. In this embodiment, the number, volume, quality, distribution and other indicators of the region of interest in the image at different periods can be quantized, which can more objectively reflect the situation of the region of interest, and will also facilitate the subsequent comparative analysis of these indicators to obtain more accurate analysis results.

[0148] It should be understood that although Figure 2-4 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2-4At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0149] In one embodiment, Figure 5 As shown, an image quantization device is provided, comprising: a segmentation module 10, a conversion relationship determination module 11, a conversion module 12, a quantization module 13 and a comparison display module 14, wherein:

[0150] The segmentation module 10 is used to perform region of interest segmentation processing on the acquired reference original image and floating original image, and determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0151] A conversion relationship determination module 11 is used to determine the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0152] A conversion module 12, used for converting the floating segmented image using the conversion relationship to obtain a converted segmented image;

[0153] A quantization module 13 is used to perform quantization processing on the reference segmented image and the converted segmented image respectively to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0154] The comparison and display module 14 is used to compare the first quantization result and the second quantization result, and display the comparison result.

[0155] For specific limitations on the image quantization device, please refer to the limitations on the image quantization method above, which will not be repeated here.

[0156] In another embodiment, another image quantization device is provided. Before the above-mentioned conversion relationship determination module 11 determines the conversion relationship between the reference original image and the floating original image based on the reference original image and the floating original image, the above-mentioned device may also include an acquisition module, which is used to acquire a reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image; the above-mentioned conversion relationship determination module 11 is also used to determine the conversion relationship between the reference original image and the floating original image based on the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or based on the reference original image and the floating original image and the reference segmentation image and the floating segmentation image.

[0157] Optionally, the conversion relationship determination module 11 may include a first registration unit, a second registration unit and a fusion unit, wherein:

[0158] A first registration unit, used for registering the reference original image and the floating original image to obtain a first conversion relationship;

[0159] A second registration unit, used to register the reference segmentation image and the floating segmentation image, or to register the reference segmentation mask image and the floating segmentation mask image to obtain a second conversion relationship;

[0160] The fusion unit is used to fuse the first conversion relationship and the second conversion relationship to determine the conversion relationship between the reference original image and the floating original image.

[0161] Optionally, the fusion unit may include a weight acquisition subunit and a fusion subunit, wherein:

[0162] A weight acquisition subunit, used to acquire a first weight corresponding to the first conversion relationship and a second weight corresponding to the second conversion relationship;

[0163] The fusion subunit is used to fuse the first conversion relationship and the second conversion relationship according to the first weight and the second weight to determine the conversion relationship between the reference original image and the floating original image.

[0164] In another embodiment, another image quantization device is provided, wherein the reference segmented image includes a first type of reference segmented image and a corresponding second reference segmented image, and the converted segmented image includes a first type of floating segmented image and a corresponding second floating segmented image. Based on the above embodiment, the quantization module 13 may include a first quantization unit and a second quantization unit, wherein:

[0165] A first quantization unit is used to perform quantization processing on the first type of reference segmented image and the second reference segmented image to determine a first quantization result corresponding to the reference segmented image; the first quantization result is used to characterize the distribution and quality of the region of interest in the second reference segmented image;

[0166] The second quantization unit is used to perform quantization processing on the first type of floating segmented image and the second floating segmented image to determine a second quantization result corresponding to the floating segmented image; the second quantization result is used to characterize the distribution and quality of the region of interest in the second floating segmented image.

[0167] Optionally, the first type of reference segmented image includes multiple first reference segmented images, each first reference image includes a first region of interest, and the second reference segmented image includes a second region of interest. The first quantization unit may include a first calculation subunit, a quality determination subunit, a second calculation subunit, and a quantization determination subunit, wherein:

[0168] A first calculation subunit is used to calculate the volume occupied by the second region of interest on each first region of interest;

[0169] A quality determination subunit, used to determine the volume percentage of the second region of interest on each first region of interest and the region of interest mass corresponding to the second region of interest on each first region of interest according to the volume occupied by the second region of interest on each first region of interest;

[0170] A second calculation subunit, configured to calculate a distribution histogram corresponding to the second region of interest according to the image value on the second region of interest;

[0171] The quantification determination subunit is used to determine the volume percentage, the quality of the region of interest and the distribution histogram as the first quantification result.

[0172] Optionally, the above-mentioned first calculation subunit is also used to obtain the reconstruction interval corresponding to the reference segmentation image, and determine the unit volume corresponding to each voxel in the reference segmentation image according to the reconstruction interval; calculate the intersection area between the second region of interest and each first region of interest, and determine the volume occupied by the second region of interest on each first region of interest based on each intersection area and the unit volume.

[0173] Optionally, the above-mentioned quality determination subunit is also used to calculate the volume of each first region of interest based on the unit volume; obtain the volume percentage of the second region of interest on each first region of interest based on the volume occupied by the second region of interest on each first region of interest and the volume of each first region of interest; determine the region of interest quality corresponding to the second region of interest on each first region of interest based on the volume occupied by the second region of interest on each first region of interest and the image value of the second region of interest.

[0174] Optionally, the conversion module 12 may include a first conversion unit and a second conversion unit, wherein:

[0175] A first conversion unit is used to convert each first reference segmented image using a conversion relationship to obtain a floating segmented image corresponding to each first reference segmented image, and obtain a first type of floating segmented image according to the floating segmented image corresponding to each first reference segmented image;

[0176] The second conversion unit is used to convert the second reference segmented image using the conversion relationship to obtain a second floating segmented image.

[0177] For specific limitations on the image quantization device, please refer to the limitations on the image quantization method above, which will not be repeated here.

[0178] Each module in the above-mentioned image quantization device can be implemented in whole or in part by software, hardware or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0179] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0180] Performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0181] Determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0182] The floating segmented image is transformed by using the transformation relationship to obtain a transformed segmented image;

[0183] The reference segmented image and the converted segmented image are quantized to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0184] The first quantization result and the second quantization result are compared, and the comparison result is displayed.

[0185] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0186] Obtain a reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image; determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image.

[0187] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0188] The reference original image and the floating original image are registered to obtain a first conversion relationship; the reference segmented image and the floating segmented image are registered, or the reference segmented mask image and the floating segmented mask image are registered to obtain a second conversion relationship; the first conversion relationship and the second conversion relationship are fused to determine the conversion relationship between the reference original image and the floating original image.

[0189] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0190] A first weight corresponding to the first conversion relationship and a second weight corresponding to the second conversion relationship are obtained; the first conversion relationship and the second conversion relationship are fused according to the first weight and the second weight to determine the conversion relationship between the reference original image and the floating original image.

[0191] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0192] The first type of reference segmented image and the second reference segmented image are quantized to determine a first quantization result corresponding to the reference segmented image; the first quantization result is used to characterize the distribution and quality of the region of interest in the second reference segmented image; the first type of floating segmented image and the second floating segmented image are quantized to determine a second quantization result corresponding to the floating segmented image; the second quantization result is used to characterize the distribution and quality of the region of interest in the second floating segmented image.

[0193] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0194] Calculate the volume occupied by the second region of interest on each first region of interest; determine the volume percentage occupied by the second region of interest on each first region of interest and the region of interest mass corresponding to the second region of interest on each first region of interest based on the volume occupied by the second region of interest on each first region of interest; calculate the distribution histogram corresponding to the second region of interest based on the image value on the second region of interest; determine the volume percentage, region of interest mass and distribution histogram as the first quantification result.

[0195] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0196] Obtain a reconstruction interval corresponding to the reference segmentation image, and determine the unit volume corresponding to each voxel in the reference segmentation image according to the reconstruction interval; calculate the intersection area between the second region of interest and each first region of interest, and determine the volume occupied by the second region of interest on each first region of interest according to each intersection area and the unit volume.

[0197] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0198] The volume of each first region of interest is calculated based on the unit volume; the volume percentage of the second region of interest on each first region of interest is obtained based on the volume occupied by the second region of interest on each first region of interest and the volume of each first region of interest; the mass of the region of interest corresponding to the second region of interest on each first region of interest is determined based on the volume occupied by the second region of interest on each first region of interest and the image value on the second region of interest.

[0199] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0200] Each first reference segmented image is transformed using the transformation relationship to obtain a floating segmented image corresponding to each first reference segmented image, and a first type of floating segmented image is obtained based on the floating segmented image corresponding to each first reference segmented image; the second reference segmented image is transformed using the transformation relationship to obtain a second floating segmented image.

[0201] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0202] Performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image;

[0203] Determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image;

[0204] The floating segmented image is transformed by using the transformation relationship to obtain a transformed segmented image;

[0205] The reference segmented image and the converted segmented image are quantized to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image;

[0206] The first quantization result and the second quantization result are compared, and the comparison result is displayed.

[0207] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0208] Obtain a reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image; determine a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image and the reference segmentation mask image and the floating segmentation mask image, or according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image.

[0209] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0210] The reference original image and the floating original image are registered to obtain a first conversion relationship; the reference segmented image and the floating segmented image are registered, or the reference segmented mask image and the floating segmented mask image are registered to obtain a second conversion relationship; the first conversion relationship and the second conversion relationship are fused to determine the conversion relationship between the reference original image and the floating original image.

[0211] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0212] A first weight corresponding to the first conversion relationship and a second weight corresponding to the second conversion relationship are obtained; the first conversion relationship and the second conversion relationship are fused according to the first weight and the second weight to determine the conversion relationship between the reference original image and the floating original image.

[0213] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0214] The first type of reference segmented image and the second reference segmented image are quantized to determine a first quantization result corresponding to the reference segmented image; the first quantization result is used to characterize the distribution and quality of the region of interest in the second reference segmented image; the first type of floating segmented image and the second floating segmented image are quantized to determine a second quantization result corresponding to the floating segmented image; the second quantization result is used to characterize the distribution and quality of the region of interest in the second floating segmented image.

[0215] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0216] Calculate the volume occupied by the second region of interest on each first region of interest; determine the volume percentage occupied by the second region of interest on each first region of interest and the region of interest mass corresponding to the second region of interest on each first region of interest based on the volume occupied by the second region of interest on each first region of interest; calculate the distribution histogram corresponding to the second region of interest based on the image value on the second region of interest; determine the volume percentage, region of interest mass and distribution histogram as the first quantification result.

[0217] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0218] Obtain a reconstruction interval corresponding to the reference segmentation image, and determine the unit volume corresponding to each voxel in the reference segmentation image according to the reconstruction interval; calculate the intersection area between the second region of interest and each first region of interest, and determine the volume occupied by the second region of interest on each first region of interest according to each intersection area and the unit volume.

[0219] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0220] The volume of each first region of interest is calculated based on the unit volume; the volume percentage of the second region of interest on each first region of interest is obtained based on the volume occupied by the second region of interest on each first region of interest and the volume of each first region of interest; the mass of the region of interest corresponding to the second region of interest on each first region of interest is determined based on the volume occupied by the second region of interest on each first region of interest and the image value on the second region of interest.

[0221] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0222] Each first reference segmented image is transformed using the transformation relationship to obtain a floating segmented image corresponding to each first reference segmented image, and a first type of floating segmented image is obtained based on the floating segmented image corresponding to each first reference segmented image; the second reference segmented image is transformed using the transformation relationship to obtain a second floating segmented image.

[0223] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0224] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0225] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. An image quantization method, characterized in that: The method comprises: Performing region of interest segmentation processing on the acquired reference original image and floating original image to determine a reference segmented image corresponding to the reference original image and a floating segmented image corresponding to the floating original image; Determining a conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image, and a reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image, or according to the reference original image and the floating original image, and the reference segmentation image and the floating segmentation image; Using the conversion relationship to convert the floating segmented image, to obtain a converted segmented image; Performing quantization processing on the reference segmented image and the transformed segmented image respectively to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the transformed segmented image; comparing the first quantization result and the second quantization result, and displaying the comparison result; The determining of the conversion relationship between the reference original image and the floating original image according to the reference original image and the floating original image and the reference segmentation mask image corresponding to the reference segmentation image and the floating segmentation mask image corresponding to the floating segmentation image, or the determining of the conversion relationship between the reference original image and the floating original image and the reference segmentation image and the floating segmentation image according to the reference original image and the floating original image and the reference segmentation image and the floating segmentation image comprises: Registering the reference original image and the floating original image to obtain a first conversion relationship; Registering the reference segmented image and the floating segmented image, or registering the reference segmented mask image and the floating segmented mask image to obtain a second conversion relationship; Performing a fusion process on the first conversion relationship and the second conversion relationship to determine a conversion relationship between the reference original image and the floating original image; The step of fusing the first conversion relationship and the second conversion relationship to determine the conversion relationship between the reference original image and the floating original image includes: Obtaining a first weight corresponding to the first conversion relationship and a second weight corresponding to the second conversion relationship; The first conversion relationship and the second conversion relationship are fused according to the first weight and the second weight to determine the conversion relationship between the reference original image and the floating original image.

2. The method according to claim 1, characterized in that Before determining the conversion relationship between the reference original image and the floating original image, the method further includes: A reference segmentation mask image corresponding to the reference segmentation image and a floating segmentation mask image corresponding to the floating segmentation image are acquired.

3. The method according to claim 1 or 2, characterized in that: The reference segmented image includes a first type of reference segmented image and a corresponding second reference segmented image, the converted segmented image includes a first type of floating segmented image and a corresponding second floating segmented image, and the reference segmented image and the converted segmented image are quantized respectively to obtain a first quantization result corresponding to the reference segmented image and a second quantization result corresponding to the converted segmented image, including: Quantizing the first type of reference segmented image and the second reference segmented image to determine a first quantization result corresponding to the reference segmented image; the first quantization result is used to characterize the distribution and quality of the region of interest in the second reference segmented image; The first type of floating segmented image and the second floating segmented image are quantized to determine a second quantization result corresponding to the floating segmented image; the second quantization result is used to characterize the distribution and quality of the region of interest in the second floating segmented image.

4. The method according to claim 3, characterized in that The first type of reference segmented images includes a plurality of first reference segmented images, each of the first reference segmented images includes a first region of interest, and the second reference segmented images include a second region of interest, and the quantization processing is performed on the first type of reference segmented images and the second reference segmented images to determine a first quantization result corresponding to the reference segmented images, including: calculating the volume occupied by the second region of interest on each of the first regions of interest; Determine, according to the volume occupied by the second region of interest on each of the first regions of interest, the volume percentage occupied by the second region of interest on each of the first regions of interest and the region of interest mass corresponding to the second region of interest on each of the first regions of interest; Calculating a distribution histogram corresponding to the second region of interest according to the image value on the second region of interest; The volume percentage, the quality of the region of interest and the distribution histogram are determined as the first quantification result.

5. The method according to claim 4, characterized in that The calculating the volume occupied by the second region of interest on each of the first regions of interest comprises: Acquire a reconstruction interval corresponding to the reference segmented image, and determine a unit volume corresponding to each voxel in the reference segmented image according to the reconstruction interval; An intersection area between the second region of interest and each of the first regions of interest is calculated, and a volume occupied by the second region of interest on each of the first regions of interest is determined according to each of the intersection areas and the unit volume.

6. The method according to claim 5, characterized in that The determining, according to the volume occupied by the second region of interest on each of the first regions of interest, a volume percentage occupied by the second region of interest on each of the first regions of interest and a region of interest mass corresponding to the second region of interest on each of the first regions of interest comprises: Calculating the volume of each first region of interest according to the unit volume; Obtaining a volume percentage of the second region of interest on each of the first regions of interest according to the volume of the second region of interest on each of the first regions of interest and the volume of each of the first regions of interest; The region of interest quality corresponding to the second region of interest on each of the first regions of interest is determined according to the volume occupied by the second region of interest on each of the first regions of interest and the image value on the second region of interest.

7. The method according to claim 4, characterized in that The step of converting the floating segmented image using the conversion relationship to obtain a converted segmented image includes: Using the conversion relationship to convert each of the first reference segmented images, obtain a floating segmented image corresponding to each first reference segmented image, and obtain the first type of floating segmented image according to the floating segmented image corresponding to each first reference segmented image; The second reference segmented image is transformed using the transformation relationship to obtain the second floating segmented image.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.