PET parameter image generation method and device

By performing quality evaluation and correction processing on PET parameter images, the problems of poor artifacts, noise and accuracy in the imaging process of PET parameter images are solved, which improves imaging efficiency and reduces costs.

CN119067905APending Publication Date: 2024-12-03SHANGHAI UNITED IMAGING HEALTHCARE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310620627.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-29
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

PET parameter images are prone to problems such as artifacts, noise and poor accuracy during imaging, resulting in low imaging efficiency and increased cost. The prior art can only choose to abandon reshoots and cannot be effectively solved.

Method used

By acquiring multi-frame PET dynamic images of the part to be inspected, quality evaluation is performed, and if it fails, correction processing is performed, including motion correction, noise reduction processing and model adaptation correction, to generate high-quality PET parameter images.

Benefits of technology

Automatic online correction of PET parameter images is realized, avoiding reshooting, improving imaging efficiency and saving imaging costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067905A_ABST
    Figure CN119067905A_ABST
Patent Text Reader

Abstract

The invention relates to a PET parameter image generation method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring multiple frames of PET dynamic images of a to-be-detected part, determining a quality evaluation result of a PET parameter image of the to-be-detected part according to each PET dynamic image, and performing correction processing on the PET dynamic images or the PET parameter images under the condition that the quality evaluation result of the PET parameter images is unqualified to obtain the corrected PET parameter images. Wherein the multi-frame PET dynamic image of the to-be-detected part is used for constructing a PET parameter image of the to-be-detected part. By means of the method, automatic correction is carried out under the condition that the quality evaluation result of the PET parameter image is unqualified so as to correct the image quality and obtain the corrected PET parameter image, re-shooting is effectively avoided through an online correction mode, the imaging efficiency is improved, and meanwhile the imaging cost is saved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and device for generating PET parameter images. Background Art

[0002] Positron emission tomography (PET) is a type of nuclear medicine imaging. Specifically, a tracer is injected into the subject to be examined, PET dynamic images are collected over a continuous period of time, and dynamic parameter analysis of the PET dynamic images is performed using a pharmacokinetic model to obtain PET parameter images including tissue and organ function parameters.

[0003] The imaging process is affected by many factors, which may cause PET parameter images to have many quality problems such as artifacts, noise, poor accuracy, etc. In the related art, PET parameter images with quality problems can only be abandoned and retaken, resulting in low imaging efficiency and increased imaging costs. Summary of the invention

[0004] Based on this, it is necessary to provide a PET parameter image generation method, apparatus, computer equipment, computer readable storage medium and computer program product to address the above technical problems.

[0005] In a first aspect, the present application provides a PET parameter image generation method, comprising:

[0006] Acquire multiple frames of PET dynamic images of the part to be examined; wherein the multiple frames of PET dynamic images of the part to be examined are used to construct a PET parameter image of the part to be examined;

[0007] Determine the quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image;

[0008] If the quality assessment result of the PET parameter image is unqualified, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image.

[0009] In one embodiment, determining the quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image includes:

[0010] Performing a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, and determining a quality assessment result of the PET parameter image according to the quality assessment result of the PET dynamic image; or,

[0011] A PET parameter image is constructed according to all PET dynamic images, and a secondary quality assessment is performed on the PET parameter image to obtain a quality assessment result of the PET parameter image.

[0012] In one embodiment, the first-level quality assessment includes a first-level motion assessment, and the first-level quality assessment is performed on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, including:

[0013] Obtaining organ segmentation results of each PET dynamic image, and determining the position offset of the same organ region according to the organ segmentation results of adjacent PET dynamic images;

[0014] If the position offset is greater than a preset offset threshold, the quality assessment result of the PET dynamic image is determined to be motion unqualified;

[0015] If the position offset is less than or equal to the offset threshold, the quality assessment result of the PET dynamic image is determined to be motion qualified.

[0016] In one embodiment, the secondary quality assessment includes secondary motion assessment, and the PET parameter image includes a first parameter image and a second parameter image representing different types of parameters; performing the secondary quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image includes:

[0017] Performing lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and performing lesion segmentation on the second parameter image to obtain a second lesion segmentation result;

[0018] Obtaining a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result;

[0019] If the similarity coefficient is greater than a preset coefficient threshold, the quality assessment result of the PET parameter image is determined to be motion qualified;

[0020] If the similarity coefficient is less than or equal to the coefficient threshold, the quality assessment result of the PET parameter image is determined to be motion unqualified.

[0021] In one embodiment, the first-level quality assessment includes a first-level noise assessment, and the first-level quality assessment is performed on the PET dynamic image to obtain a quality assessment result of the PET dynamic image, including:

[0022] Determine the target organ area in PET dynamic images;

[0023] Determine the image variation coefficient of the target organ region according to the pixel value of each pixel point in the target organ region;

[0024] If the image variation coefficient is greater than a preset variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for noise;

[0025] If the image variation coefficient is less than or equal to the variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be noise qualified.

[0026] In one embodiment, the secondary quality assessment includes a secondary noise assessment, and the secondary quality assessment is performed on the PET parameter image to obtain a quality assessment result of the PET parameter image, including:

[0027] Obtain the drug concentration time variation curve of each pixel in each PET dynamic image;

[0028] The drug concentration time variation curve is input into the noise evaluation model, and the noise evaluation model outputs the noise level parameter of the corresponding pixel point in the PET parameter image;

[0029] The quality assessment result of the PET parameter image is determined according to the noise level parameters of the corresponding pixel points in the PET parameter image.

[0030] In one embodiment, determining the quality assessment result of the PET parameter image according to the noise level parameter of the corresponding pixel point in the parameter image includes:

[0031] Obtaining an average value of the noise level parameters of all pixels in the PET parameter image as a target noise level parameter of the PET parameter image;

[0032] If the target noise level parameter is greater than a preset noise threshold, the quality assessment result of the PET parameter image is determined to be unqualified for noise;

[0033] If the target noise level parameter is less than or equal to the noise threshold, the quality assessment result of the PET parameter image is determined to be noise qualified.

[0034] In one embodiment, the secondary quality assessment includes a model adaptation assessment, and the secondary quality assessment is performed on the PET parameter image to obtain a quality assessment result of the PET parameter image, including:

[0035] Obtaining the goodness-of-fit parameter of the pharmacokinetic model for each pixel in the PET parameter image;

[0036] Determine whether there is a pixel area in the PET parameter image that does not match the pharmacokinetic model based on the goodness of fit parameters of each pixel point;

[0037] If it exists, the quality assessment result of the PET parameter image is determined to be unqualified in model adaptation; if it does not exist, the quality assessment result of the PET parameter image is determined to be qualified in model adaptation.

[0038] In one embodiment, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image, including:

[0039] If the quality assessment result of the PET parameter image is that the motion is unqualified, motion correction is performed on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image;

[0040] If the quality assessment result of the PET parameter image is that the noise is unqualified, the PET dynamic image or the PET parameter image is subjected to noise reduction processing to obtain a corrected PET parameter image;

[0041] If the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image, and a corrected PET parameter image is obtained according to the new pharmacokinetic model.

[0042] In a second aspect, the present application also provides a PET parameter image generating device, comprising:

[0043] A data acquisition module is used to acquire multiple frames of PET dynamic images of the part to be inspected; wherein the multiple frames of PET dynamic images of the part to be inspected are used to construct a PET parameter image of the part to be inspected;

[0044] A quality assessment module, used to determine the quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image;

[0045] The imaging correction module is used to correct the PET dynamic image or the PET parameter image if the quality assessment result of the PET parameter image is unqualified, so as to obtain a corrected PET parameter image.

[0046] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program.

[0047] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any of the above methods when the computer program is executed by a processor.

[0048] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of any of the above methods when executed by a processor.

[0049] The above-mentioned PET parameter image generation method, device, computer equipment, storage medium and computer program product obtain multiple frames of PET dynamic images of the part to be inspected to determine the quality evaluation result of the PET parameter image of the part to be inspected according to each PET dynamic image, and in the case where the quality evaluation result of the PET parameter image is unqualified, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image. Among them, the multiple frames of PET dynamic images of the part to be inspected are used to construct the PET parameter image of the part to be inspected. Through the above-mentioned method, automatic correction is achieved in the case where the quality evaluation result of the PET parameter image is unqualified to correct the image quality and obtain the corrected PET parameter image. The online correction method effectively avoids reshooting, improves imaging efficiency and saves imaging costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is an internal structure diagram of a computer device in one embodiment;

[0051] Figure 2 A schematic diagram of a process of generating a PET parameter image in one embodiment;

[0052] Figure 3 A schematic diagram of a process for obtaining a corrected PET parameter image in one embodiment;

[0053] Figure 4 A schematic diagram of a process of first-level motion assessment in one embodiment;

[0054] Figure 5 is a schematic diagram of a process of secondary motion assessment in one embodiment;

[0055] Figure 6 A schematic diagram of a process of first-level noise assessment in one embodiment;

[0056] Figure 7 is a schematic diagram of a process of secondary noise assessment in one embodiment;

[0057] Figure 8 A schematic diagram of a flow chart of secondary noise assessment in another embodiment;

[0058] Fig. 9 A schematic diagram of a flow chart of model adaptation evaluation in one embodiment;

[0059] Fig.10 FIG. 4 is a structural block diagram of a PET parameter image generating device in one embodiment. DETAILED DESCRIPTION

[0060] 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.

[0061] The PET parameter image generation method provided in the embodiment of the present application can be applied to Figure 1 The computer device shown in FIG. 1 may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 1 As shown, it 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, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a PET parameter image generation 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 shell, or an external keyboard, touchpad or mouse, etc.

[0062] 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.

[0063] In one embodiment, Figure 2 As shown, a PET parameter image generation method is provided, and the method is applied to Figure 1 The computer device in the example is used to illustrate, including the following steps:

[0064] S210, acquiring multiple frames of PET dynamic images of the part to be examined.

[0065] The multiple frames of PET dynamic images of the part to be examined are used to construct a PET parameter image of the part to be examined.

[0066] It should be noted that the computer device in this embodiment may be a PET scanning device, and the PET dynamic image is a series of scanning image sets that change with time and are obtained during the scanning process of the PET scanning device, and is used to reflect the distribution of the tracer injected into the subject over time. The PET scanning device reconstructs the PET dynamic image obtained by scanning to obtain a PET parameter image.

[0067] Optionally, the computer device performs a PET scan on the part to be examined where the tracer is injected, so as to obtain multiple frames of PET dynamic images of the part to be examined.

[0068] S220 , determining a quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image.

[0069] Optionally, the computer device may use a predictive evaluation to perform a quality evaluation on the PET parameter image to determine the quality evaluation result of the PET parameter image. For example, a quality evaluation is performed on each PET dynamic image, and the quality evaluation result of the PET parameter image is predicted based on the quality evaluation result of the PET dynamic image. The computer device may also use a retrospective evaluation to perform a quality evaluation on the PET parameter image to determine the quality evaluation result of the PET parameter image. For example, the PET dynamic image is input into the pharmacokinetic model to obtain the PET parameter image, and the quality evaluation of the PET parameter image is performed to directly obtain the quality evaluation result of the PET parameter image.

[0070] S230: If the quality assessment result of the PET parameter image is unqualified, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image.

[0071] Optionally, after obtaining the quality assessment result of the PET parameter image, the computer device may output the quality assessment result, and may also correct the PET dynamic image when the quality assessment result of the PET parameter image is unqualified, and use the corrected PET dynamic image to input into the pharmacokinetic model to calculate the PET parameter image. The computer device may also directly correct the PET parameter image to obtain the corrected PET parameter image when the quality assessment result of the PET parameter image is unqualified.

[0072] Optionally, when the quality assessment result of the PET parameter image is qualified, the computer device may save the PET parameter image, and may also generate and output a quality assessment result report including the PET parameter image.

[0073] In this embodiment, the computer device obtains multiple frames of PET dynamic images of the part to be inspected to determine the quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image, and when the quality assessment result of the PET parameter image is unqualified, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image. Among them, the multiple frames of PET dynamic images of the part to be inspected are used to construct the PET parameter image of the part to be inspected. Through the above method, automatic correction is achieved when the quality assessment result of the PET parameter image is unqualified to correct the image quality and obtain a corrected PET parameter image. The online correction method effectively avoids rescanning, improves imaging efficiency and saves imaging costs.

[0074] The quality assessment result of the PET parameter image can be estimated by performing quality assessment on the PET dynamic image, or can be directly obtained by performing quality assessment on the PET parameter image. Based on this, in one embodiment, the above S220, determining the quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image, includes:

[0075] Performing a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, and determining a quality assessment result of the PET parameter image according to the quality assessment result of the PET dynamic image; or,

[0076] A PET parameter image is constructed according to the PET dynamic image, and a secondary quality assessment is performed on the PET parameter image to obtain a quality assessment result of the PET parameter image.

[0077] Optionally, the PET dynamic images are used to construct PET parameter images, and the computer device can perform a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, and estimate the quality assessment result of the PET parameter image based on the quality assessment result of the PET dynamic image.

[0078] Optionally, if the quality assessment result of the PET dynamic image is unqualified, the computer device determines that the quality assessment result of the PET parameter image is unqualified, and performs quality assessment and correction of the dynamic image in advance before generating the PET parameter image, thereby improving the generation efficiency of the PET parameter image; conversely, if the quality assessment result of the PET dynamic image is qualified, the computer device determines that the quality assessment result of the PET parameter image is qualified, and may also perform further quality assessment on the PET parameter image to determine whether the quality assessment result of the PET parameter image is qualified.

[0079] In order to improve the accuracy of the quality assessment results of the PET dynamic images, the computer device can determine the quality assessment results of the PET dynamic images according to the first-level quality assessment results of each frame of the PET dynamic images. For example, the computer device can obtain the first-level quality assessment results of each PET dynamic image, determine the number of unqualified images, and determine that the quality assessment results of the PET dynamic images are unqualified when the number of unqualified images is greater than the percentage threshold; conversely, when the number of unqualified images is less than or equal to the percentage threshold, the quality assessment results of the PET dynamic images are determined to be qualified.

[0080] Optionally, the computer device may also construct a PET parameter image based on the PET dynamic image, and perform a secondary quality assessment on the PET parameter image to directly obtain a quality assessment result of the PET parameter image.

[0081] The primary quality assessment and the secondary quality assessment may include multiple aspects of quality assessment, for example, an assessment on motion to determine artifacts in an image; an assessment on noise to determine the noise level of an image.

[0082] It should be noted that PET dynamic images need to be combined with an adapted pharmacokinetic model to complete the construction of PET parameter images. The secondary quality assessment is a direct assessment of PET parameter images. Therefore, the secondary quality assessment can also include an assessment of the pharmacokinetic model to determine whether the model is adapted.

[0083] In this embodiment, the computer device can perform a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, so as to determine the quality assessment result of the PET parameter image based on the quality assessment result of the PET dynamic image. In this method, the quality assessment result of the PET parameter image is estimated by performing a quality assessment on the PET dynamic image, thereby realizing a pre-construction estimate assessment, so that the PET dynamic image can be corrected before construction, and the corrected PET parameter image can be obtained more efficiently and quickly. The computer device can also construct a PET parameter image based on all PET dynamic images, and perform a second-level quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image. In this method, the quality assessment result of the PET parameter image is directly obtained by performing a quality assessment on the PET parameter image, thereby improving the authenticity and reliability of the result.

[0084] Where primary and secondary quality assessments may include multiple aspects of quality assessment, such as Figure 3 As shown, in the above S230, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image, including:

[0085] S310: If the quality assessment result of the PET parameter image is that the motion is unqualified, motion correction is performed on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image.

[0086] Optionally, when the computer device determines that the quality assessment result of the PET parameter image is motion unqualified based on the quality assessment result of the PET dynamic image being motion unqualified, the computer device may accordingly perform motion correction on the PET dynamic image to construct a PET parameter image using the motion-corrected PET dynamic image, and the PET parameter image is the corrected PET parameter image.

[0087] Optionally, when the computer device directly determines that the quality assessment result of the PET parameter image is motion unqualified, the computer device may perform motion correction on the PET parameter image to directly obtain a corrected PET parameter image, or may perform motion correction on the PET dynamic image used to construct the PET parameter image to construct a new PET parameter image using the motion-corrected PET parameter image, and the new PET parameter image is the corrected PET parameter image.

[0088] Optionally, the computer device may use a corresponding registration algorithm to perform motion correction according to the detection part. For example, when the detection part is the head (usually produces rigid motion), a rigid registration algorithm may be used to perform motion correction; when the detection part is the liver or lung (performs respiratory motion), a non-rigid registration algorithm may be used to perform motion correction.

[0089] S320: If the quality assessment result of the PET parameter image is that the noise is unqualified, a noise reduction process is performed on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image.

[0090] Optionally, when the computer device determines that the quality assessment result of the PET parameter image is unqualified for noise based on the quality assessment result of the PET dynamic image being unqualified for noise, the computer device may accordingly perform noise reduction processing on the PET dynamic image to construct a PET parameter image using the PET dynamic image after noise reduction processing, and the PET parameter image is the corrected PET parameter image.

[0091] Optionally, when the computer device directly determines that the quality assessment result of the PET parameter image is unqualified for noise, the computer device may perform noise reduction processing on the PET parameter image to directly obtain a corrected PET parameter image.

[0092] S330: If the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image, and a corrected PET parameter image is obtained according to the new pharmacokinetic model.

[0093] It should be noted that in the process of computer equipment constructing PET parameter images using a preset pharmacokinetic model from multiple frames of PET dynamic images, there are cases where the pharmacokinetic model used in some areas is not suitable, that is, the quality assessment result of the corresponding PET parameter image is that the model adaptation is unqualified.

[0094] Optionally, when the computer device determines that the quality assessment result of the PET parameter image is that the model adaptation is unqualified, the computer device can replace a new pharmacokinetic model for the unmatched pixel area in the PET parameter image, and then construct a new PET parameter image from the PET dynamic image according to the new pharmacokinetic model, and use the PET parameter image as the corrected PET parameter image. Among them, for the area where the dynamic model does not match, pixel by pixel analysis can be performed to select a suitable pharmacokinetic model. The new pharmacokinetic model can be used to calculate the PET dynamic image corresponding to the unmatched pixel area in the PET parameter image, and the preset pharmacokinetic model can be used to continue to calculate the PET dynamic image corresponding to the matched pixel area in the PET parameter image, and finally obtain an updated PET parameter image.

[0095] In this embodiment, when the quality assessment result of the PET parameter image is that the motion is unqualified, the computer device performs motion correction on the PET dynamic image / PET parameter image to obtain a corrected PET parameter image; when the quality assessment result of the PET parameter image is that the noise is unqualified, the PET dynamic image / PET parameter image is subjected to noise reduction processing to obtain a corrected PET parameter image; when the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image to obtain a corrected PET parameter image according to the new pharmacokinetic model. In the above method, the correction process is to correct the specific unqualified content of the quality assessment result of the PET parameter image using a corresponding correction method, so as to achieve targeted correction of the unqualified content of the PET parameter image and improve the quality of the corrected PET parameter image.

[0096] The first-level quality assessment includes a first-level motion assessment, which is used to estimate the artifacts in the PET parameter image based on the motion information of the organ region in the PET dynamic image. Figure 4 As shown, the first-level quality assessment is performed on each PET dynamic image to obtain the quality assessment result of the PET dynamic image, including:

[0097] S410 , obtaining organ segmentation results of each PET dynamic image, and determining the position offset of the same organ region according to the organ segmentation results of adjacent PET dynamic images.

[0098] The multiple frames of PET dynamic images used to construct the PET parameter image are time-series continuous images.

[0099] Optionally, the computer device performs organ segmentation on each PET dynamic image to obtain the organ segmentation result of each PET dynamic image, and compares the organ segmentation results of two adjacent frames of PET dynamic images to determine the position offset of the same organ region, so as to estimate whether a large movement / deformation that causes artifacts in the PET parameter image has occurred in the part to be inspected based on the position offset, that is, to perform a motion-specific quality assessment on the PET dynamic image. For example, it is pre-set that the region with the largest area in the organ segmentation results of two adjacent frames of PET dynamic images represents the same organ, and the computer device can respectively obtain the position of the region with the largest area in the two adjacent frames of PET dynamic images, and obtain the position offset between the two.

[0100] S420: If the position offset is greater than a preset offset threshold, determine that the quality assessment result of the PET dynamic image is motion unqualified.

[0101] S430: If the position offset is less than or equal to the offset threshold, determine that the quality assessment result of the PET dynamic image is qualified for motion.

[0102] Optionally, the computer device can determine whether the quality assessment of the PET dynamic image for motion is qualified by threshold comparison according to the obtained position offset. If the position offset is greater than the preset offset threshold, it means that the object to be inspected has undergone a large movement / deformation during the scan. If the PET dynamic image obtained by scanning is used to construct a PET parameter image, the probability of motion artifacts in the PET parameter image is high. Based on this, the computer device determines that the quality assessment result of the PET dynamic image is unqualified for motion; conversely, if the position offset is less than or equal to the offset threshold, it means that the object to be inspected has not undergone a large movement / deformation during the scan. If the PET dynamic image obtained by scanning is used to construct a PET parameter image, the probability of motion artifacts in the PET parameter image is low. Based on this, the computer device determines that the quality assessment result of the PET dynamic image is qualified for motion.

[0103] In this embodiment, the computer device obtains the organ segmentation results of each PET dynamic image, and determines the position offset of the same organ region according to the organ segmentation results of adjacent PET dynamic images, so as to determine the quality assessment result of the PET dynamic image for motion according to the position offset. Wherein, if the position offset is greater than a preset offset threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for motion; if the position offset is less than or equal to the offset threshold, the quality assessment result of the PET dynamic image is determined to be qualified for motion. In the above method, the position offset of the same organ region can accurately determine whether a large movement / deformation occurs in the part to be inspected during scanning, which directly affects the probability of whether there is motion artifact in the constructed PET parameter image. Therefore, the position offset of the same organ region in the organ segmentation result based on the PET dynamic image can effectively improve the accuracy of motion estimation.

[0104] The secondary quality assessment includes a secondary motion assessment, which is used to directly determine the artifact situation in the PET parameter image based on the motion information of the lesion area in the PET parameter image. In one embodiment, Figure 5 As shown, the above-mentioned two-level quality assessment is performed on each PET parameter image to obtain the quality assessment result of the PET parameter image, including:

[0105] S510 , performing lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and performing lesion segmentation on the second parameter image to obtain a second lesion segmentation result.

[0106] The PET parameter image includes a first parameter image and a second parameter image that characterize different types of parameters. For example, the PET dynamic image may include a Ki parameter image and an Intercept parameter image output based on the Patlak model.

[0107] Optionally, the computer device may perform lesion segmentation on the first parameter image to determine the lesion area in the first parameter image, and use the first parameter image with the lesion area segmented as the first lesion segmentation result, and simultaneously perform lesion segmentation on the second parameter image to determine the lesion area in the second parameter image, and use the second parameter image with the lesion area segmented as the second lesion segmentation result.

[0108] S520: Obtain a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result.

[0109] Optionally, the computer device may further obtain a similarity coefficient (DSC) between the first lesion segmentation result and the second lesion segmentation result to determine whether a large movement / deformation occurs in the lesion area in the PET parameter image based on the similarity coefficient, that is, to perform a motion-related quality assessment on the PET parameter image.

[0110] It should be noted that, when the part to be inspected does not undergo a large movement / deformation during scanning, the difference between the lesion segmentation results of the first parameter image and the second parameter image representing different types of parameters is small, and the similarity coefficient between the two is high. If the part to be inspected undergoes a large movement / deformation during scanning, the difference between the lesion segmentation results of the first parameter image and the second parameter image will become larger, resulting in a decrease in the similarity coefficient between the two.

[0111] S530: If the similarity coefficient is greater than a preset coefficient threshold, it is determined that the quality assessment result of the PET parameter image is motion qualified.

[0112] S540: If the similarity coefficient is less than or equal to the coefficient threshold, determine that the quality assessment result of the PET parameter image is motion unqualified.

[0113] Optionally, the computer device can determine whether the quality assessment of the PET parameter image for motion is qualified by threshold comparison based on the obtained similarity coefficient. If the similarity coefficient is greater than the preset coefficient threshold, it means that the object to be inspected has undergone a small movement / deformation during the scan, and there is no motion artifact in the obtained PET parameter image (or the degree of motion artifact is low and does not exceed the control standard). Based on this, the computer device determines that the quality assessment result of the PET parameter image is qualified for motion; conversely, if the similarity coefficient is less than or equal to the offset threshold, it means that the object to be inspected has undergone a large movement / deformation during the scan, and there is a motion artifact in the obtained PET parameter image and it exceeds the control standard. Based on this, the computer device determines that the quality assessment result of the PET parameter image is unqualified for motion.

[0114] In this embodiment, the computer device performs lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and performs lesion segmentation on the second parameter image to obtain a second lesion segmentation result, and then obtains the similarity coefficient between the first lesion segmentation result and the second lesion segmentation result, so as to determine the quality assessment result of the PET parameter image for motion according to the similarity coefficient. If the similarity coefficient is greater than a preset coefficient threshold, the quality assessment result of the PET parameter image is determined to be qualified for motion; if the similarity coefficient is less than or equal to the coefficient threshold, the quality assessment result of the PET parameter image is determined to be unqualified for motion. In the above method, the first parameter image and the second parameter image represent different types of parameters. Whether the part to be inspected undergoes large movement / deformation during scanning directly affects the similarity coefficient between the lesion segmentation results corresponding to the PET parameter images of different types of parameters, so the accuracy of the quality assessment result can be effectively improved based on the similarity coefficient.

[0115] The first-level quality assessment includes a first-level noise assessment for estimating the noise level of the PET parameter image based on the PET dynamic image. Figure 6 As shown, the first-level quality assessment is performed on each PET dynamic image to obtain the quality assessment result of the PET dynamic image, including:

[0116] S610: Determine the target organ region in the PET dynamic image.

[0117] The target organ region may be the region where a preset organ is located in the PET dynamic image, or may be the region where the largest organ in the PET dynamic image is located.

[0118] Optionally, the computer device performs organ segmentation on the PET dynamic image to obtain an organ segmentation result that divides different organs on the PET dynamic image. The computer device can determine the area where a preset organ is located in the organ segmentation result as the target organ area, and can also determine the area with the largest area in the organ segmentation result as the target organ area.

[0119] S620, determining the image variation coefficient of the target organ region according to the pixel value of each pixel point in the target organ region.

[0120] Among them, the image coefficient of variation (COV) is used to characterize the ratio between the standard deviation and the mean value of the data within the group. It should be noted that, ideally, the medium in the same organ area is evenly distributed, the difference between pixels is small, and the corresponding image coefficient of variation is also small, while the noise in the image will increase the image coefficient of variation.

[0121] Optionally, the computer device can calculate the image variation coefficient of the target organ area based on the pixel value of each pixel point in the target organ area, so as to determine the noise level of the PET dynamic image based on the image variation coefficient, that is, to perform a quality assessment on the PET dynamic image with respect to noise.

[0122] S630: If the image variation coefficient is greater than a preset variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for noise.

[0123] S640: If the image variation coefficient is less than or equal to the variation coefficient threshold, determine that the quality assessment result of the PET dynamic image is qualified for noise.

[0124] Optionally, on the basis that the noise in the image will increase the image variation coefficient, the computer device can determine whether the quality assessment of the PET dynamic image for noise is qualified by threshold comparison according to the calculated image variation coefficient. Among them, if the image variation coefficient is greater than the preset variation coefficient threshold, it means that the noise level in the PET dynamic image is high, and it is likely that the noise level in the constructed PET parameter image is also high, and the computer device determines that the quality assessment result of the PET dynamic image is unqualified for noise; on the contrary, if the image variation coefficient is less than or equal to the preset variation coefficient threshold, it means that the noise level in the PET dynamic image is low, and it is likely that the noise level in the constructed PE parameter image is also low, and the computer device determines that the quality assessment result of the PET dynamic image is qualified for noise.

[0125] In this embodiment, the computer device determines the target organ region in the PET dynamic image, and determines the image variation coefficient of the target organ region according to the pixel value of each pixel in the target organ region, so as to determine the quality assessment result of the PET dynamic image for noise according to the image variation coefficient. If the image variation coefficient is greater than the preset variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for noise; if the image variation coefficient is less than or equal to the variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be qualified for noise. In the above method, the image variation coefficient of the target organ region can accurately reflect the noise level in the PET dynamic image, which directly affects the noise level of the constructed PET parameter image. Therefore, the image variation coefficient of the target organ segmentation region in the PET dynamic image can effectively improve the accuracy of the noise assessment.

[0126] The secondary quality assessment includes a secondary noise assessment for determining the noise level of the PET parameter image directly based on the PET parameter image. Figure 7 As shown, the above-mentioned two-level quality assessment of the PET parameter image is performed to obtain the quality assessment result of the PET parameter image, including:

[0127] S710 , obtaining a drug concentration time variation curve for each pixel point in each PET dynamic image.

[0128] The drug concentration-time variation curve is the time-activity curve (TAC), which is a curve showing the variation of the radioactive tracer concentration in a specific reaction area (such as tissue or plasma) over time.

[0129] Optionally, the computer device determines a standardized uptake value (SUV) of each pixel in the PET dynamic image based on each PET dynamic image to obtain a SUV map corresponding to the PET dynamic image, and then determines the TAC of each pixel in the PET dynamic image based on the SUV map.

[0130] S720, inputting the drug concentration time variation curve into the noise evaluation model, and the noise evaluation model outputs the noise level parameters of the corresponding pixel points in the PET parameter image.

[0131] The noise assessment model may be a Gaussian denoising model.

[0132] Optionally, after obtaining the TAC of each pixel in the PET dynamic image, the computer device can input the TAC of each pixel into a noise assessment model, and the noise assessment model outputs the noise level parameters of the corresponding pixel in the PET parameter image constructed from the PET dynamic image.

[0133] Optionally, the noise assessment model may determine the noise level parameter α based on the following formula:

[0134]

[0135] Among them, N frame Indicates the number of frames of PET dynamic image, N parameters Indicates the number of parameters of the dynamic model, y i represents the time-activity curve, represents the time activity curve obtained by fitting, λ represents the decay constant, t i Indicates the time corresponding to the i-th frame, Δt i Indicates the scanning duration corresponding to the i-th frame, i is greater than or equal to 1 and less than or equal to N frame A positive integer.

[0136] S730 , determining a quality assessment result of the PET parameter image according to noise level parameters of corresponding pixel points in the PET parameter image.

[0137] Optionally, after obtaining the noise level parameters of the corresponding pixel points in the PET parameter image output by the noise assessment model, the computer device may perform data processing on the noise level parameters of the corresponding pixel points in the PET parameter image to comprehensively determine the noise level of the PET parameter image, that is, perform a quality assessment on the PET parameter image with respect to noise.

[0138] In an optional embodiment, if Figure 8 As shown, the above S730, determining the quality assessment result of the PET parameter image according to the noise level parameter of the corresponding pixel point in the PET parameter image, includes:

[0139] S810 , obtaining an average value of noise level parameters of all pixels in the PET parameter image as a target noise level parameter of the PET parameter image.

[0140] Optionally, in order to achieve an overall noise evaluation of the constructed PET parameter image, the computer device may obtain an average value of the noise level parameters of all pixels in the PET parameter image as a target noise level parameter of the PET parameter image, so as to determine the noise level of the PET parameter image according to the target noise level parameter.

[0141] S820: If the target noise level parameter is greater than a preset noise threshold, determine that the quality assessment result of the PET parameter image is unqualified for noise.

[0142] S830: If the target noise level parameter is less than or equal to the noise threshold, determine that the quality assessment result of the PET parameter image is noise qualified.

[0143] Optionally, after obtaining the target noise level parameter of the PET parameter image, the computer device may determine whether the quality assessment of the PET parameter image for noise is qualified by threshold comparison according to the target noise level parameter. If the target noise level parameter is greater than a preset noise threshold, it means that the noise level in the PET parameter image is high and exceeds the control standard, and the computer device determines that the quality assessment result of the PET parameter image is unqualified for noise; conversely, if the target noise level parameter is less than or equal to the preset noise threshold, it means that the noise level in the PET parameter image is low and does not exceed the control standard, and the computer device determines that the quality assessment result of the PET parameter image is qualified for noise.

[0144] In this embodiment, the computer device obtains the drug concentration time variation curve of each pixel point in each PET dynamic image, inputs the drug concentration time variation curve into the noise assessment model, and the noise assessment model outputs the noise level parameter of the corresponding pixel point in the PET parameter image, and then determines the quality assessment result of the PET parameter image according to the noise level parameter of the corresponding pixel point in the PET parameter image. Among them, the average value of the noise level parameter of all pixels in the PET parameter image can be obtained as the target noise level parameter of the PET parameter image. If the target noise level parameter is greater than the preset noise threshold, the quality assessment result of the PET parameter image is determined to be unqualified in noise; if the target noise level parameter is less than or equal to the noise threshold, the quality assessment result of the PET parameter image is determined to be qualified in noise. The above method can be used to determine the noise level of the PET parameter image itself, so as to directly obtain the quality assessment result of the PET parameter image for noise, thereby improving the authenticity and reliability of the result.

[0145] The secondary quality assessment includes a model adaptation assessment, which is used to determine whether the pharmacokinetic model used in the process of constructing the PET parameter image is adapted to the corresponding region. Based on this, in one embodiment, Fig. 9 As shown, the above-mentioned two-level quality assessment of the PET parameter image is performed to obtain the quality assessment result of the PET parameter image, including:

[0146] S910, obtaining a goodness-of-fit parameter of the pharmacokinetic model for each pixel in the PET parameter image.

[0147] Among them, the goodness of fit parameter is used to characterize the adaptability of the pharmacokinetic model to each pixel in the PET parameter image.

[0148] Optionally, the computer device inputs multiple frames of PET dynamic images into a preset pharmacokinetic model to obtain a PET parameter image. The computer device further determines a goodness of fit parameter between each pixel point in the PET parameter image and the preset pharmacokinetic model. Specifically, the pharmacokinetic model can output a time acquisition activity change curve for each pixel point, and compare the time acquisition activity change curve output by the pharmacokinetic model with the time acquisition change curve of each corresponding pixel point obtained by measurement to obtain the goodness of fit parameter.

[0149] S920: Determine whether there is a pixel region in the PET parameter image that does not match the pharmacokinetic model according to the goodness of fit parameter of each pixel point.

[0150] Optionally, after obtaining the goodness-of-fit parameters between each pixel point in the PET parameter image and a preset pharmacokinetic model, the computer device can determine whether there are pixels in the PET parameter image that do not match the pharmacokinetic model based on the goodness-of-fit parameters of each pixel point, and the pixel area formed by the set of mismatched pixels is the pixel area that does not match the pharmacokinetic model.

[0151] Optionally, the computer device may determine whether the quality assessment of the PET parameter image for model fitness is qualified by threshold comparison according to the calculated goodness of fit parameter between each pixel point in the PET parameter image and the preset pharmacokinetic model. If the goodness of fit parameter of the pixel is greater than the preset parameter threshold, the computer device determines that the corresponding pixel point in the PET parameter image matches the preset pharmacokinetic model; conversely, if the goodness of fit parameter of the pixel is less than or equal to the preset parameter threshold, the computer device determines that the corresponding pixel point in the PET parameter image does not match the preset pharmacokinetic model.

[0152] S930. If it exists, determine that the quality assessment result of the PET parameter image is that the model adaptation is unqualified; if it does not exist, determine that the quality assessment result of the PET parameter image is that the model adaptation is qualified.

[0153] Optionally, if the computer device determines that there is a pixel area in the PET parameter image that does not match the preset pharmacokinetic model, it determines that the quality assessment result of the PET parameter image is that the model adaptation is unqualified; conversely, if the computer device determines that there is no pixel area in the PET parameter image that does not match the preset pharmacokinetic model, it determines that the quality assessment result of the PET parameter image is that the model adaptation is qualified.

[0154] In this embodiment, the computer device obtains the goodness of fit parameter of the pharmacokinetic model for each pixel in the PET parameter image, and determines whether there is a pixel area in the PET parameter image that does not match the pharmacokinetic model according to the goodness of fit parameter of each pixel, and if there is, determines the quality assessment result of the PET parameter image as unqualified model adaptation; if not, determines the quality assessment result of the PET parameter image as qualified model adaptation. The above method can be used to implement the quality assessment of whether the pharmacokinetic model used in the construction process is adapted, and whether the model is adapted is an important factor affecting the quality of PET parameter image generation. Therefore, the quality assessment of whether the PET parameter image is adapted for the preset pharmacokinetic model can fundamentally and qualitatively determine the imaging quality of the PET parameter image, so as to clarify the subsequent correction direction and improve the imaging efficiency.

[0155] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0156] Based on the same inventive concept, the embodiment of the present application also provides a PET parameter image generation device for implementing the PET parameter image generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more PET parameter image generation device embodiments provided below can refer to the limitations of the PET parameter image generation method above, and will not be repeated here.

[0157] In one embodiment, Fig.10 As shown, a PET parameter image generation device is provided, comprising: a data acquisition module 1001, a quality assessment module 1002 and an imaging correction module 1003, wherein:

[0158] The data acquisition module 1001 is used to acquire multiple frames of PET dynamic images of the part to be inspected; wherein the multiple frames of PET dynamic images of the part to be inspected are used to construct a PET parameter image of the part to be inspected;

[0159] The quality assessment module 1002 is used to determine the quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image;

[0160] The imaging correction module 1003 is used to correct the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image if the quality assessment result of the PET parameter image is unqualified.

[0161] In one embodiment, the quality assessment module 1002 includes:

[0162] The first evaluation submodule is used to perform a first-level quality evaluation on each PET dynamic image to obtain a quality evaluation result of the PET dynamic image, and determine a quality evaluation result of the PET parameter image according to the quality evaluation result of the PET dynamic image; or

[0163] The second evaluation submodule is used to construct a PET parameter image based on all PET dynamic images, and perform a secondary quality evaluation on the PET parameter image to obtain a quality evaluation result of the PET parameter image.

[0164] In one embodiment, the first-level quality assessment includes a first-level motion assessment, and the first assessment submodule includes:

[0165] An organ segmentation unit is used to obtain the organ segmentation results of each PET dynamic image, and determine the position offset of the same organ region according to the organ segmentation results of adjacent PET dynamic images;

[0166] The first motion unit is used to determine that the quality assessment result of the PET dynamic image is unqualified for motion if the position offset is greater than a preset offset threshold; if the position offset is less than or equal to the offset threshold, determine that the quality assessment result of the PET dynamic image is qualified for motion.

[0167] In one embodiment, the secondary quality assessment includes a secondary motion assessment, the PET parameter image includes a first parameter image and a second parameter image representing different types of parameters; and the second assessment submodule includes:

[0168] A lesion segmentation unit, configured to perform lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and to perform lesion segmentation on the second parameter image to obtain a second lesion segmentation result;

[0169] A similarity determination unit, used for obtaining a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result;

[0170] The second motion unit is used to determine that the quality assessment result of the PET parameter image is motion qualified if the similarity coefficient is greater than a preset coefficient threshold; if the similarity coefficient is less than or equal to the coefficient threshold, determine that the quality assessment result of the PET parameter image is motion unqualified.

[0171] In one embodiment, the first-level quality assessment includes a first-level noise assessment, and the first assessment submodule includes:

[0172] A target confirmation unit, used to determine the target organ area in the PET dynamic image;

[0173] An image variation unit, used for determining an image variation coefficient of the target organ region according to a pixel value of each pixel point in the target organ region;

[0174] The first noise unit is used to determine that the quality assessment result of the PET dynamic image is unqualified for noise if the image variation coefficient is greater than a preset variation coefficient threshold; if the image variation coefficient is less than or equal to the variation coefficient threshold, then determine that the quality assessment result of the PET dynamic image is qualified for noise.

[0175] In one embodiment, the secondary quality assessment includes a secondary noise assessment, and the second assessment submodule includes:

[0176] A concentration curve unit is used to obtain a drug concentration time variation curve for each pixel in each PET dynamic image;

[0177] A noise parameter unit, used for inputting the drug concentration time variation curve into the noise evaluation model, and the noise evaluation model outputs the noise level parameter of the corresponding pixel point in the PET parameter image;

[0178] The noise evaluation unit is used to determine the quality evaluation result of the PET parameter image according to the noise level parameters of the corresponding pixel points in the PET parameter image.

[0179] In one embodiment, the noise assessment unit comprises:

[0180] A parameter subunit, used for obtaining an average value of noise level parameters of all pixels in the PET parameter image as a target noise level parameter of the PET parameter image;

[0181] The noise subunit is used to determine that the quality assessment result of the PET parameter image is unqualified for noise if the target noise level parameter is greater than a preset noise threshold; and to determine that the quality assessment result of the PET parameter image is qualified for noise if the target noise level parameter is less than or equal to the noise threshold.

[0182] In one embodiment, the secondary quality assessment includes a model adaptation assessment, and the second assessment submodule includes:

[0183] A goodness of fit unit, used to obtain a goodness of fit parameter of the pharmacokinetic model for each pixel in the PET parameter image;

[0184] A model matching unit, used to determine whether there is a pixel area in the PET parameter image that does not match the pharmacokinetic model according to the goodness of fit parameters of each pixel point;

[0185] The result judgment unit is used to determine that the quality evaluation result of the PET parameter image is unqualified in model adaptation if it exists; if it does not exist, determine that the quality evaluation result of the PET parameter image is qualified in model adaptation.

[0186] In one embodiment, the imaging correction module 1003 includes:

[0187] A first correction unit is used for performing motion correction on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image if the quality assessment result of the PET parameter image is that the motion is unqualified;

[0188] A second correction unit is used for performing noise reduction processing on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image if the quality assessment result of the PET parameter image is that the noise is unqualified;

[0189] The third correction unit is used to replace a new pharmacokinetic model for the unmatched pixel area in the PET parameter image if the quality assessment result of the PET parameter image is that the model adaptation is unqualified, and obtain a corrected PET parameter image according to the new pharmacokinetic model.

[0190] Each module in the above-mentioned PET parameter image generation device can be implemented in whole or in part by software, hardware or a combination thereof. Each module 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 module above.

[0191] 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:

[0192] Acquire multiple frames of PET dynamic images of the part to be inspected; wherein the multiple frames of PET dynamic images of the part to be inspected are used to construct a PET parameter image of the part to be inspected; determine a quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image; if the quality assessment result of the PET parameter image is unqualified, perform correction processing on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image.

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

[0194] Perform a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, and determine the quality assessment result of the PET parameter image based on the quality assessment result of the PET dynamic image; or, construct a PET parameter image based on all PET dynamic images, and perform a second-level quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image.

[0195] In one embodiment, the first-level quality assessment includes a first-level motion assessment, and when the processor executes the computer program, the processor further implements the following steps:

[0196] The organ segmentation results of each PET dynamic image are obtained, and the position offset of the same organ area is determined according to the organ segmentation results of adjacent PET dynamic images; if the position offset is greater than a preset offset threshold, the quality assessment result of the PET dynamic image is determined to be motion unqualified; if the position offset is less than or equal to the offset threshold, the quality assessment result of the PET dynamic image is determined to be motion qualified.

[0197] In one embodiment, the secondary quality assessment includes a secondary motion assessment, and the PET parameter image includes a first parameter image and a second parameter image representing different types of parameters; when the processor executes the computer program, the following steps are also implemented:

[0198] Perform lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and perform lesion segmentation on the second parameter image to obtain a second lesion segmentation result; obtain a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result; if the similarity coefficient is greater than a preset coefficient threshold, determine that the quality assessment result of the PET parameter image is motion qualified; if the similarity coefficient is less than or equal to the coefficient threshold, determine that the quality assessment result of the PET parameter image is motion unqualified.

[0199] In one embodiment, the first-level quality assessment includes a first-level noise assessment, and when the processor executes the computer program, the following steps are further implemented:

[0200] The target organ region is determined in the PET dynamic image; the image variation coefficient of the target organ region is determined according to the pixel value of each pixel point in the target organ region; if the image variation coefficient is greater than a preset variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for noise; if the image variation coefficient is less than or equal to the variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be qualified for noise.

[0201] In one embodiment, the secondary quality assessment includes a secondary noise assessment, and when the processor executes the computer program, the following steps are further implemented:

[0202] Obtain the drug concentration time variation curve of each pixel point in each PET dynamic image; input the drug concentration time variation curve into the noise assessment model, and the noise assessment model outputs the noise level parameters of the corresponding pixel points in the PET parameter image; determine the quality assessment result of the PET parameter image according to the noise level parameters of the corresponding pixel points in the PET parameter image.

[0203] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0204] The average value of the noise level parameters of all pixels in the PET parameter image is obtained as the target noise level parameter of the PET parameter image; if the target noise level parameter is greater than a preset noise threshold, the quality assessment result of the PET parameter image is determined to be unqualified for noise; if the target noise level parameter is less than or equal to the noise threshold, the quality assessment result of the PET parameter image is determined to be qualified for noise.

[0205] In one embodiment, the secondary quality assessment includes a model adaptation assessment, and when the processor executes the computer program, the following steps are also implemented:

[0206] Obtain the goodness-of-fit parameter of the pharmacokinetic model for each pixel in the PET parameter image; determine whether there is a pixel area in the PET parameter image that does not match the pharmacokinetic model based on the goodness-of-fit parameter of each pixel; if so, determine that the quality assessment result of the PET parameter image is unqualified in model adaptation; if not, determine that the quality assessment result of the PET parameter image is qualified in model adaptation.

[0207] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0208] If the quality assessment result of the PET parameter image is that the motion is unqualified, the PET dynamic image or the PET parameter image is motion corrected to obtain a corrected PET parameter image; if the quality assessment result of the PET parameter image is that the noise is unqualified, the PET dynamic image or the PET parameter image is denoised to obtain a corrected PET parameter image; if the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image, and a corrected PET parameter image is obtained according to the new pharmacokinetic model.

[0209] 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:

[0210] Acquire multiple frames of PET dynamic images of the part to be inspected; wherein the multiple frames of PET dynamic images of the part to be inspected are used to construct a PET parameter image of the part to be inspected; determine a quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image; if the quality assessment result of the PET parameter image is unqualified, perform correction processing on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image.

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

[0212] Perform a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, and determine the quality assessment result of the PET parameter image based on the quality assessment result of the PET dynamic image; or, construct a PET parameter image based on all PET dynamic images, and perform a second-level quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image.

[0213] In one embodiment, the first-level quality assessment includes a first-level motion assessment, and when the computer program is executed by the processor, the following steps are further implemented:

[0214] The organ segmentation results of each PET dynamic image are obtained, and the position offset of the same organ area is determined according to the organ segmentation results of adjacent PET dynamic images; if the position offset is greater than a preset offset threshold, the quality assessment result of the PET dynamic image is determined to be motion unqualified; if the position offset is less than or equal to the offset threshold, the quality assessment result of the PET dynamic image is determined to be motion qualified.

[0215] In one embodiment, the secondary quality assessment includes a secondary motion assessment, and the PET parameter image includes a first parameter image and a second parameter image representing different types of parameters; when the computer program is executed by the processor, the following steps are also implemented:

[0216] Perform lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and perform lesion segmentation on the second parameter image to obtain a second lesion segmentation result; obtain a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result; if the similarity coefficient is greater than a preset coefficient threshold, determine that the quality assessment result of the PET parameter image is motion qualified; if the similarity coefficient is less than or equal to the coefficient threshold, determine that the quality assessment result of the PET parameter image is motion unqualified.

[0217] In one embodiment, the first-level quality assessment includes a first-level noise assessment, and when the computer program is executed by the processor, the following steps are further implemented:

[0218] The target organ region is determined in the PET dynamic image; the image variation coefficient of the target organ region is determined according to the pixel value of each pixel point in the target organ region; if the image variation coefficient is greater than a preset variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for noise; if the image variation coefficient is less than or equal to the variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be qualified for noise.

[0219] In one embodiment, the secondary quality assessment includes a secondary noise assessment, and the computer program, when executed by a processor, further implements the following steps:

[0220] Obtain the drug concentration time variation curve of each pixel point in each PET dynamic image; input the drug concentration time variation curve into the noise assessment model, and the noise assessment model outputs the noise level parameters of the corresponding pixel points in the PET parameter image; determine the quality assessment result of the PET parameter image according to the noise level parameters of the corresponding pixel points in the PET parameter image.

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

[0222] The average value of the noise level parameters of all pixels in the PET parameter image is obtained as the target noise level parameter of the PET parameter image; if the target noise level parameter is greater than a preset noise threshold, the quality assessment result of the PET parameter image is determined to be unqualified for noise; if the target noise level parameter is less than or equal to the noise threshold, the quality assessment result of the PET parameter image is determined to be qualified for noise.

[0223] In one embodiment, the secondary quality assessment includes a model adaptation assessment, and when the computer program is executed by a processor, the following steps are further implemented:

[0224] Obtain the goodness-of-fit parameter of the pharmacokinetic model for each pixel in the PET parameter image; determine whether there is a pixel area in the PET parameter image that does not match the pharmacokinetic model based on the goodness-of-fit parameter of each pixel; if so, determine that the quality assessment result of the PET parameter image is unqualified in model adaptation; if not, determine that the quality assessment result of the PET parameter image is qualified in model adaptation.

[0225] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0226] If the quality assessment result of the PET parameter image is that the motion is unqualified, the PET dynamic image or the PET parameter image is motion corrected to obtain a corrected PET parameter image; if the quality assessment result of the PET parameter image is that the noise is unqualified, the PET dynamic image or the PET parameter image is denoised to obtain a corrected PET parameter image; if the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image, and a corrected PET parameter image is obtained according to the new pharmacokinetic model.

[0227] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0228] Acquire multiple frames of PET dynamic images of the part to be inspected; wherein the multiple frames of PET dynamic images of the part to be inspected are used to construct a PET parameter image of the part to be inspected; determine a quality assessment result of the PET parameter image of the part to be inspected according to each PET dynamic image; if the quality assessment result of the PET parameter image is unqualified, perform correction processing on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image.

[0229] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0230] Perform a first-level quality assessment on each PET dynamic image to obtain a quality assessment result of the PET dynamic image, and determine the quality assessment result of the PET parameter image based on the quality assessment result of the PET dynamic image; or, construct a PET parameter image based on all PET dynamic images, and perform a second-level quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image.

[0231] In one embodiment, the first-level quality assessment includes a first-level motion assessment, and when the computer program is executed by the processor, the following steps are further implemented:

[0232] The organ segmentation results of each PET dynamic image are obtained, and the position offset of the same organ area is determined according to the organ segmentation results of adjacent PET dynamic images; if the position offset is greater than a preset offset threshold, the quality assessment result of the PET dynamic image is determined to be motion unqualified; if the position offset is less than or equal to the offset threshold, the quality assessment result of the PET dynamic image is determined to be motion qualified.

[0233] In one embodiment, the secondary quality assessment includes a secondary motion assessment, and the PET parameter image includes a first parameter image and a second parameter image representing different types of parameters; when the computer program is executed by the processor, the following steps are also implemented:

[0234] Perform lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and perform lesion segmentation on the second parameter image to obtain a second lesion segmentation result; obtain a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result; if the similarity coefficient is greater than a preset coefficient threshold, determine that the quality assessment result of the PET parameter image is motion qualified; if the similarity coefficient is less than or equal to the coefficient threshold, determine that the quality assessment result of the PET parameter image is motion unqualified.

[0235] In one embodiment, the first-level quality assessment includes a first-level noise assessment, and when the computer program is executed by the processor, the following steps are further implemented:

[0236] The target organ region is determined in the PET dynamic image; the image variation coefficient of the target organ region is determined according to the pixel value of each pixel point in the target organ region; if the image variation coefficient is greater than a preset variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be unqualified for noise; if the image variation coefficient is less than or equal to the variation coefficient threshold, the quality assessment result of the PET dynamic image is determined to be qualified for noise.

[0237] In one embodiment, the secondary quality assessment includes a secondary noise assessment, and the computer program, when executed by a processor, further implements the following steps:

[0238] Obtain the drug concentration time variation curve of each pixel point in each PET dynamic image; input the drug concentration time variation curve into the noise assessment model, and the noise assessment model outputs the noise level parameters of the corresponding pixel points in the PET parameter image; determine the quality assessment result of the PET parameter image according to the noise level parameters of the corresponding pixel points in the PET parameter image.

[0239] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0240] The average value of the noise level parameters of all pixels in the PET parameter image is obtained as the target noise level parameter of the PET parameter image; if the target noise level parameter is greater than a preset noise threshold, the quality assessment result of the PET parameter image is determined to be unqualified for noise; if the target noise level parameter is less than or equal to the noise threshold, the quality assessment result of the PET parameter image is determined to be qualified for noise.

[0241] In one embodiment, the secondary quality assessment includes a model adaptation assessment, and when the computer program is executed by a processor, the following steps are further implemented:

[0242] Obtain the goodness-of-fit parameter of the pharmacokinetic model for each pixel in the PET parameter image; determine whether there is a pixel area in the PET parameter image that does not match the pharmacokinetic model based on the goodness-of-fit parameter of each pixel; if so, determine that the quality assessment result of the PET parameter image is unqualified in model adaptation; if not, determine that the quality assessment result of the PET parameter image is qualified in model adaptation.

[0243] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0244] If the quality assessment result of the PET parameter image is that the motion is unqualified, the PET dynamic image or the PET parameter image is motion corrected to obtain a corrected PET parameter image; if the quality assessment result of the PET parameter image is that the noise is unqualified, the PET dynamic image or the PET parameter image is denoised to obtain a corrected PET parameter image; if the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image, and a corrected PET parameter image is obtained according to the new pharmacokinetic model.

[0245] Those skilled 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 the memory, database or other medium used in the embodiments provided in the present 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, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. 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). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0246] 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.

[0247] The above-described 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 present application. 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 present application shall be subject to the attached claims.

Claims

1. A PET parameter image generation method, characterized in that: The method comprises: Acquire multiple frames of PET dynamic images of the part to be examined; wherein the multiple frames of PET dynamic images of the part to be examined are used to construct a PET parameter image of the part to be examined; Determine the quality assessment result of the PET parameter image of the part to be inspected according to each of the PET dynamic images; If the quality assessment result of the PET parameter image is unqualified, the PET dynamic image or the PET parameter image is corrected to obtain a corrected PET parameter image.

2. The method according to claim 1, characterized in that The step of determining the quality assessment result of the PET parameter image of the part to be inspected according to each of the PET dynamic images comprises: Performing a first-level quality assessment on each of the PET dynamic images to obtain a quality assessment result of the PET dynamic image, and determining a quality assessment result of the PET parameter image according to the quality assessment result of the PET dynamic image; or, The PET parameter image is constructed according to all the PET dynamic images, and a secondary quality assessment is performed on the PET parameter image to obtain a quality assessment result of the PET parameter image.

3. The method according to claim 2, characterized in that The first-level quality assessment includes a first-level motion assessment, and the first-level quality assessment is performed on each of the PET dynamic images to obtain a quality assessment result of the PET dynamic image, including: Acquire the organ segmentation results of each of the PET dynamic images, and determine the position offset of the same organ region according to the organ segmentation results of adjacent PET dynamic images; If the position offset is greater than a preset offset threshold, determining that the quality assessment result of the PET dynamic image is motion unqualified; If the position offset is less than or equal to the offset threshold, it is determined that the quality assessment result of the PET dynamic image is qualified for motion.

4. The method according to claim 2, characterized in that: The secondary quality assessment includes a secondary motion assessment, and the PET parameter image includes a first parameter image and a second parameter image representing different types of parameters; performing the secondary quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image includes: Performing lesion segmentation on the first parameter image to obtain a first lesion segmentation result, and performing lesion segmentation on the second parameter image to obtain a second lesion segmentation result; Acquire a similarity coefficient between the first lesion segmentation result and the second lesion segmentation result; If the similarity coefficient is greater than a preset coefficient threshold, determining that the quality assessment result of the PET parameter image is motion qualified; If the similarity coefficient is less than or equal to the coefficient threshold, it is determined that the quality assessment result of the PET parameter image is motion unqualified.

5. The method according to claim 2, characterized in that: The first-level quality assessment includes a first-level noise assessment, and the first-level quality assessment is performed on the PET dynamic image to obtain a quality assessment result of the PET dynamic image, including: determining a target organ region in the PET dynamic image; Determine the image variation coefficient of the target organ region according to the pixel value of each pixel point in the target organ region; If the image variation coefficient is greater than a preset variation coefficient threshold, determining that the quality assessment result of the PET dynamic image is unqualified for noise; If the image variation coefficient is less than or equal to the variation coefficient threshold, it is determined that the quality assessment result of the PET dynamic image is qualified for noise.

6. The method according to claim 2, characterized in that The secondary quality assessment includes a secondary noise assessment, and performing the secondary quality assessment on the PET parameter image to obtain a quality assessment result of the PET parameter image includes: Obtaining a drug concentration time variation curve for each pixel in each of the PET dynamic images; Inputting the drug concentration time variation curve into a noise evaluation model, and outputting the noise level parameter of the corresponding pixel point in the PET parameter image by the noise evaluation model; The quality assessment result of the PET parameter image is determined according to the noise level parameters of the corresponding pixel points in the PET parameter image.

7. The method according to claim 6, characterized in that Determining the quality assessment result of the PET parameter image according to the noise level parameter of the corresponding pixel point in the parameter image includes: Obtaining an average value of noise level parameters of all pixels in the PET parameter image as a target noise level parameter of the PET parameter image; If the target noise level parameter is greater than a preset noise threshold, determining that the quality assessment result of the PET parameter image is unqualified for noise; If the target noise level parameter is less than or equal to the noise threshold, it is determined that the quality assessment result of the PET parameter image is noise qualified.

8. The method according to claim 2, characterized in that: The secondary quality assessment includes a model adaptation assessment, and the secondary quality assessment of the PET parameter image is performed to obtain a quality assessment result of the PET parameter image, including: Obtaining a goodness-of-fit parameter of a pharmacokinetic model for each pixel in the PET parameter image; Determining whether there is a pixel region in the PET parameter image that does not match the pharmacokinetic model according to the goodness of fit parameter of each pixel point; If it exists, the quality assessment result of the PET parameter image is determined to be unqualified in model adaptation; if it does not exist, the quality assessment result of the PET parameter image is determined to be qualified in model adaptation.

9. The method according to any one of claims 1 to 8, characterized in that The step of performing correction processing on the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image includes: If the quality assessment result of the PET parameter image is that the motion is unqualified, performing motion correction on the PET dynamic image or the PET parameter image to obtain the corrected PET parameter image; If the quality assessment result of the PET parameter image is that the noise is unqualified, performing noise reduction processing on the PET dynamic image or the PET parameter image to obtain the corrected PET parameter image; If the quality assessment result of the PET parameter image is that the model adaptation is unqualified, a new pharmacokinetic model is replaced for the unmatched pixel area in the PET parameter image, and the corrected PET parameter image is obtained according to the new pharmacokinetic model.

10. A PET parameter image generating device, characterized in that: The device comprises: A data acquisition module, used to acquire multiple frames of PET dynamic images of the part to be inspected; wherein the multiple frames of PET dynamic images of the part to be inspected are used to construct a PET parameter image of the part to be inspected; A quality assessment module, used for determining a quality assessment result of the PET parameter image of the part to be inspected according to each of the PET dynamic images; The imaging correction module is used to correct the PET dynamic image or the PET parameter image to obtain a corrected PET parameter image if the quality assessment result of the PET parameter image is unqualified.