Method, device, equipment and medium for determining kinetic parameters based on PET imaging

Abstract

The application discloses a method and device for determining kinetic parameters based on PET imaging, equipment and medium, the method comprises the following steps: determining the tracer output function according to the blood input function and the two-compartment dynamic model; according to the dynamic PET image set and the tracer output function, the model dynamic parameters of the two-compartment dynamic model are obtained by iterative solution; according to the model dynamic parameters, the kinetic parameter image is determined; wherein, the dynamic PET image set represents the PET image set within a preset time after tracer injection, the preset time is greater than or equal to 30 minutes and less than 60 minutes, the two-compartment dynamic model corresponds to the blood compartment and the tissue compartment, the blood input function represents the activity value of the tracer in the blood compartment at different scanning times, the tracer output function represents the sum of the activity value of the tracer in the blood compartment and the activity value of the tracer in the tissue compartment, which improves the inspection throughput and use efficiency of the PET imaging equipment.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to a method, apparatus, device, and medium for determining dynamic parameters based on PET imaging. Background Technology

[0002] In recent years, dynamic positron emission tomography (PET) imaging has been widely used in clinical examinations and disease diagnosis. Dynamic PET imaging, in particular, refers to the continuous acquisition of PET images over a period of time to provide a data basis for determining dynamic parameters.

[0003] The two main scanning protocols currently used are late-stage dynamic scanning, such as 30-60 minutes after tracer injection, and two short-duration dynamic scanning, such as 0-6 minutes and 60-75 minutes after tracer injection, or 0-10 minutes and 55-60 minutes after tracer injection.

[0004] Although the two scanning protocols mentioned above have shorter scanning times, the required imaging time is still relatively long, resulting in low throughput and low efficiency of PET imaging equipment. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for determining dynamic parameters based on PET imaging, in order to solve the problem of long imaging time required for scanning protocols in dynamic PET imaging, thereby improving the throughput and efficiency of PET imaging equipment.

[0006] According to one embodiment of the present invention, a method for determining dynamic parameters based on PET imaging is provided, the method comprising:

[0007] Acquire a dynamic PET image set and obtain the blood input function;

[0008] Based on the blood input function and the two-compartment dynamic model, determine the tracer output function;

[0009] Based on the dynamic PET image set and the tracer output function, the dynamic parameters of the two-compartment dynamic model are obtained by iteratively solving the two-compartment dynamic model.

[0010] Based on the model's dynamic parameters, determine the dynamic parameter image;

[0011] The dynamic PET image set refers to the set of PET images within a preset time period after tracer injection, wherein the preset time period is greater than or equal to 30 minutes and less than 60 minutes; the two-compartment dynamic model corresponds to the blood compartment and the tissue compartment; the blood input function represents the activity value of the tracer in the blood compartment at different scanning times; and the tracer output function represents the sum of the activity values ​​of the tracer in the blood compartment and the tracer in the tissue compartment.

[0012] According to another embodiment of the present invention, a device for determining dynamic parameters based on PET imaging is provided, the device comprising:

[0013] The blood input function acquisition module is used to acquire dynamic PET image sets and obtain blood input functions;

[0014] The tracer output function determination module is used to determine the tracer output function based on the blood input function and the two-compartment dynamic model.

[0015] The model dynamic parameter solving module is used to iteratively solve the two-compartment dynamic model to obtain the model dynamic parameters of the two-compartment dynamic model based on the dynamic PET image set and the tracer output function.

[0016] The dynamic parameter image determination module is used to determine the dynamic parameter image based on the model's dynamic parameters;

[0017] The dynamic PET image set refers to the set of PET images within a preset time period after tracer injection, wherein the preset time period is greater than or equal to 30 minutes and less than 60 minutes; the two-compartment dynamic model corresponds to the blood compartment and the tissue compartment; the blood input function represents the activity value of the tracer in the blood compartment at different scanning times; and the tracer output function represents the sum of the activity values ​​of the tracer in the blood compartment and the tracer in the tissue compartment.

[0018] According to another embodiment of the present invention, an electronic device is provided, the electronic device comprising:

[0019] At least one processor; and

[0020] A memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the method for determining dynamic parameters based on PET imaging as described in any embodiment of the present invention.

[0022] According to another embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the method for determining dynamic parameters based on PET imaging as described in any embodiment of the present invention.

[0023] According to another embodiment of the present invention, a computer program product is provided, including a computer program that, when executed by a processor, implements the method for determining dynamic parameters based on PET imaging as described in any embodiment of the present invention.

[0024] The technical solution of this invention determines the tracer output function based on the blood input function and the two-compartment dynamic model. Iteratively solving the two-compartment dynamic model using the dynamic PET image set and the tracer output function yields the model dynamic parameters. Based on these parameters, a kinetic parameter image is determined. The dynamic PET image set represents the PET images taken within a preset time after tracer injection, with the preset time being greater than or equal to 30 minutes and less than 60 minutes. This approach ensures the accuracy of the kinetic parameter image while shortening the imaging time of dynamic PET imaging, solving the problem of long imaging times required by the scanning protocol of dynamic PET imaging, and improving the throughput and efficiency of PET imaging equipment.

[0025] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A flowchart illustrating a method for determining dynamic parameters based on PET imaging, as provided in one embodiment of the present invention;

[0028] Figure 2 A model architecture diagram of a two-compartment dynamic model provided in one embodiment of the present invention;

[0029] Figure 3 A flowchart illustrating another method for determining dynamic parameters based on PET imaging, provided in one embodiment of the present invention;

[0030] Figure 4A K provided as an embodiment of the present invention i -30 minutes and K i The corresponding ROC curves for each of the 65-minute intervals;

[0031] Figure 5 This is a schematic diagram of a device for determining dynamic parameters based on PET imaging, provided in one embodiment of the present invention.

[0032] Figure 6 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product or device.

[0035] Figure 1 This is a flowchart illustrating a method for determining dynamic parameters based on PET imaging, provided in one embodiment of the present invention. This embodiment is applicable to situations where dynamic parameters in dynamic PET imaging need to be estimated. The method can be executed by a device for determining dynamic parameters based on PET imaging, which can be implemented in hardware and / or software and can be configured in a terminal device. Figure 1 As shown, the method includes:

[0036] S110. Obtain the dynamic PET image set and obtain the blood input function.

[0037] In this embodiment, the dynamic PET image set refers to the set of PET images within a preset time period after tracer injection, where the preset time period is greater than or equal to 30 minutes and less than 60 minutes.

[0038] In this context, a tracer refers to a radioactive substance that undergoes decay and emits specific rays that can be detected by a detector. Examples of tracers include, but are not limited to, 18F-FDG (2-deoxy-2-[18F]fluoro-D-glucose), 11C-choline, 18F-FMISO (18F-FMISO), or 18F-FLT (3'-deoxy-3'-18F-fluorothymidine), etc. The specific tracer used is not limited here and can be customized according to actual needs.

[0039] In one optional embodiment, acquiring a dynamic PET image set includes: acquiring a dynamic PET image set captured by a PET / CT imaging device (Positron Emission Tomography / Computed Tomography); or reading a pre-stored dynamic PET image set from a cloud server.

[0040] In one optional embodiment, the acquisition object corresponding to the dynamic PET image set is the test tissue that is completely within the scanner's field of view and has positive tracer uptake, and whose PD-L1 (Programmed Death-Ligand 1) expression level meets the expression level condition.

[0041] For example, the tissue to be tested can be non-small cell lung cancer (NSCLC), but is not limited to the example scenario.

[0042] For example, PD-L1 expression levels can be assessed based on indicators such as the Tumor Proportion Score (TPS) and the Combined Positive Score (CPS). A TPS greater than 50% indicates high PD-L1 expression, a TPS between 1% and 50% indicates low PD-L1 expression, and a TPS less than 1% indicates negative PD-L1 expression. Similarly, a CPS greater than 10 indicates high PD-L1 expression, a CPS between 1 and 10 indicates low PD-L1 expression, and a CPS less than 1 indicates negative PD-L1 expression.

[0043] Of course, TPS and CPS can also be normalized separately, and the expression level of PD-L1 can be determined together based on the normalization results.

[0044] In an optional embodiment, the expression level condition is that the expression level of PD-L1 is high.

[0045] The determination method, classification method, and indicator basis of PD-L1 expression level are not limited here. They can be customized according to actual needs.

[0046] For the scanning protocol of dynamic PET imaging, taking 18F-FDG as the tracer and a preset duration of 30 minutes as an example, in some embodiments, the subjects should avoid strenuous exercise and fast for at least 6 hours before the PET / CT scan. At the time of 18F-FDG injection, the blood glucose level should be below 8.0 mmol / L. First, the subjects undergo a breath-hold chest CT scan and a whole-body CT scan (from head to mid-femur, supine position, arms raised) to obtain CT scan data. The CT scan parameters are as follows: tube voltage 120 kV, tube current 10-220 mA, pitch 1.375:1, noise figure 20. Subsequently, immediately after the 18F-FDG injection (via an indwelling intravenous catheter, mean injection dose ± standard deviation 288.78 ± 49.69 MBq, injection dose range 203.15-423.65 MBq), a dynamic PET scan of the chest region (axial field of view 20 cm) is performed to obtain dynamic PET data. The specific plan is as follows: 6×10 seconds, 4×30 seconds, 4×60 seconds, 4×120 seconds, and 3×300 seconds, lasting for a total of 30 minutes.

[0047] Specifically, CT scan data is used to perform attenuation correction on dynamic PET data, and the attenuation-corrected dynamic PET data is then reconstructed to obtain a dynamic PET image set. The reconstruction algorithm can employ Block Sequential Regularized Expectation Maximization (BSREM), with reconstruction parameters including 25 iterations and 2 subsets, and a matrix size of 256×256.

[0048] In this embodiment, the blood input function represents the activity value of the tracer in the blood at different scan times. In an optional embodiment, the blood input function is a discrete array or a continuous curve. The array contains the measured activity values ​​corresponding to multiple scan times, and the curve contains the fitted activity values ​​corresponding to multiple scan times.

[0049] In one optional embodiment, obtaining the blood input function includes: determining the blood tracer count based on at least two scan times corresponding to the dynamic PET image set, and determining the blood input function based on the blood tracer count corresponding to each scan time.

[0050] The blood tracer count represents the statistical quantity of tracers in the blood at the time of scanning. In this embodiment, the blood tracer count is used to characterize the measured activity value of tracers in the blood at the time of scanning. For example, the blood tracer count can be obtained by measuring the blood sample of the collected object using a gamma counter.

[0051] S120. Determine the tracer output function based on the blood input function and the two-compartment dynamic model.

[0052] In this embodiment, the two-compartment dynamic model corresponds to the blood compartment and the tissue compartment. The two-compartment dynamic model is a mathematical model defined based on the kinetic characteristics of the tracer moving to different compartments per unit time. The rate of change of tracer activity in the blood compartment or the tissue compartment is called the rate constant, which can also be called the kinetic parameter.

[0053] Figure 2 This is a model architecture diagram of a two-compartment dynamic model provided in one embodiment of the present invention. Specifically, Figure 2 In this diagram, "blood" represents the blood compartment, "C1" and "C2" together represent the tissue compartment, K1 is the forward transport rate, which is the rate constant for the tracer to move from the blood compartment to the tissue compartment, k2 is the reverse transport rate, which is the rate constant for the tracer to move from the tissue compartment to the blood compartment, k3 is the phosphorylation rate, which is the rate constant for the free tracer in the tissue compartment to be converted into phosphorylated tracer, or it can also be expressed as the rate constant for the tracer to move from the unphosphorylated compartment to the phosphorylated compartment, and k4 is the dephosphorylation rate, which is the rate constant for the phosphorylated tracer in the tissue compartment to be converted into the free tracer, or it can also be expressed as the rate constant for the tracer to move from the phosphorylated compartment to the unphosphorylated compartment.

[0054] In this embodiment, the blood input function represents the activity value of the tracer in the blood compartment at different scan times, and the tracer output function represents the sum of the activity values ​​of the tracer in the blood compartment and the tracer in the tissue compartment at each scan time.

[0055] In one optional embodiment, the tracer output function is determined based on the blood input function and the two-compartment dynamic model, including: determining the tissue activity function based on the blood input function and the model dynamic parameters; and performing a weighted summation of the blood input function and the tissue activity function to obtain the tracer output function.

[0056] In this embodiment, the tissue activity function represents the activity value of the tracer in the tissue compartment at different scanning times, and the weight corresponding to the blood input function is the blood volume percentage.

[0057] In an optional embodiment, determining the tissue activity function based on the blood input function and model dynamic parameters includes: determining the instantaneous response function based on the model dynamic parameters; and using the convolution result of the blood input function and the instantaneous response function as the tissue activity function.

[0058] In this embodiment, the tracer is an irreversible tracer, and accordingly, the model dynamic parameters do not include the dephosphorylation rate, which represents the rate constant of phosphorylated tracer in tissue compartments being converted into free tracer.

[0059] For example, tracers include, but are not limited to, 18F-FDG or choline-based tracers. The tracer used is not limited here, and can be customized according to actual needs.

[0060] In this embodiment, the model dynamic parameters include forward transport rate K1, reverse transport rate k2, and phosphorylation rate k3.

[0061] For example, the tissue activity function C i (t) is expressed by the following formula:

[0062]

[0063] Where t represents the scan time, C p y(t) represents the blood input function, and y(t) represents the instantaneous response function.

[0064] For example, the tracer output function C T (t) is expressed by the following formula:

[0065] C T (t)=CBV×C p (t)+C i (t);

[0066] CBV represents the percentage of blood volume.

[0067] S130. Based on the dynamic PET image set and tracer output function, the dynamic parameters of the two-compartment dynamic model are obtained by iteratively solving the two-compartment dynamic model.

[0068] In one optional embodiment, the dynamic parameters of the two-compartment dynamic model are obtained by iteratively solving the dynamic PET image set and the tracer output function. This includes: determining the objective function based on the dynamic PET image set and the tracer output function; and using a least squares fitting algorithm to minimize the objective function to obtain the dynamic parameters of the two-compartment dynamic model.

[0069] In this embodiment, the objective function characterizes the sum of squared errors between the voxel values ​​of all pixels in the PET image corresponding to each scanning time in the dynamic PET image set and the activity value of the tracer output function at the scanning time.

[0070] For example, the objective function can be expressed by the following formula:

[0071]

[0072] in, This represents the parameter set for iterative solution, which includes model dynamic parameters and blood volume percentage. C PET (t i ) represents the voxel value of the i-th pixel in the PET image corresponding to the scanning time t in the dynamic PET image set, and N represents the number of pixels in the PET image.

[0073] S140. Determine the dynamic parameter image based on the model's dynamic parameters.

[0074] In one alternative embodiment, the kinetic parameter image includes at least one of a forward transport rate image, a reverse transport rate image, a phosphorylation rate image, and a net uptake rate image.

[0075] In one specific embodiment, when the kinetic parameter image includes a net uptake rate image, the net uptake rate is determined based on the forward transport rate, reverse transport rate, and phosphorylation rate in the model kinetic parameters; and the net uptake rate image is determined based on the net uptake rate.

[0076] Among them, net uptake rate K i It can be used to reflect the net amount of tracer entering the tissue per unit time. For example, the net uptake rate K. i It is expressed by the following formula:

[0077]

[0078] The technical solution of this embodiment determines the tracer output function based on the blood input function and the two-compartment dynamic model. Iteratively solving the two-compartment dynamic model based on the dynamic PET image set and the tracer output function yields the model dynamic parameters. Based on these parameters, a kinetic parameter image is determined. The dynamic PET image set represents the PET images taken within a preset time after tracer injection. The preset time is greater than or equal to 30 minutes and less than 60 minutes. This approach ensures the accuracy of the kinetic parameter image while shortening the imaging time of dynamic PET imaging, solving the problem of long imaging times required by the scanning protocol of dynamic PET imaging and improving the throughput and efficiency of PET imaging equipment.

[0079] Figure 3 This is a flowchart illustrating another method for determining dynamic parameters based on PET imaging, provided in one embodiment of the present invention. This embodiment further refines the "acquiring blood input function" step in the above embodiment. In this embodiment, acquiring the blood input function includes: acquiring an early PET image set corresponding to the dynamic PET image set; wherein the early PET image set contains multiple PET images with scan times prior to the tracer injection; determining the blood image region corresponding to the blood compartment based on the early PET image set; determining the average voxel value corresponding to each PET image frame based on the dynamic PET image set and the blood image region; and constructing the blood input function based on the scan time and average voxel value of each PET image frame.

[0080] like Figure 3 As shown, the method includes:

[0081] S210, Obtain dynamic PET image set.

[0082] S210 in this embodiment is the same as that in the above embodiment. Figure 1 The S110 shown is the same or similar, and will not be described again in this embodiment.

[0083] S220. Obtain the early PET image set corresponding to the dynamic PET image set.

[0084] In this embodiment, the early PET image set refers to the set of PET images in the early time range after tracer injection, and the scanning time corresponding to each frame of PET image in the early PET image set is earlier than the scanning time of other PET images outside the early PET image set.

[0085] For example, the early time range can be [0, 30s] or [0, 60s]. There is no limitation on the early time range here, and it can be customized according to actual needs.

[0086] S230. Based on the early PET image set, determine the blood image region corresponding to the blood chamber.

[0087] Specifically, the blood image region represents the arterial region. In one optional embodiment, a preset segmentation algorithm is used to determine the blood image region corresponding to the blood compartments based on an early PET image set. Exemplary preset segmentation algorithms include, but are not limited to, threshold-based segmentation algorithms, region-based segmentation algorithms, edge-based segmentation algorithms, clustering-based segmentation algorithms, and deep learning algorithms, etc.

[0088] Threshold-based segmentation algorithms employ one or more pixel thresholds and divide the PET image into blood image regions based on the comparison between voxel values ​​and these thresholds. Region segmentation algorithms include region growing and region splitting / merging algorithms. Region growing selects a seed pixel in the PET image and merges neighboring pixels into the seed pixel's region based on similarity, continuing until a stopping condition is met to obtain the blood image region. Region splitting / merging divides the PET image into several smaller regions and determines whether adjacent regions should be merged or split into smaller regions based on similarity criteria, continuing until a stopping condition is met to obtain the blood image region. Edge-based segmentation algorithms detect discontinuities in voxel values ​​in the PET image to determine the edges of the blood image region. For example, the neural networks used in deep learning algorithms include, but are not limited to, fully convolutional neural networks or U-net networks.

[0089] The preset segmentation algorithm used here is not limited; it can be customized according to actual needs.

[0090] In another alternative embodiment, when the blood image region is the ascending aorta region and the blood image region is manually drawn, the region size of the blood image region is 10mm × 10mm × 20mm.

[0091] S240. Based on the blood image region, determine the average voxel value corresponding to each frame of PET image in the dynamic PET image set.

[0092] In one optional embodiment, determining the average voxel value corresponding to each frame of PET image in the dynamic PET image set based on the blood image region includes: for each frame of PET image in the dynamic PET image set, obtaining a blood segmentation image in the PET image corresponding to the blood image region, and determining the average voxel value of the PET image based on the blood segmentation image.

[0093] In one optional embodiment, determining the average voxel value of the PET image based on the blood segmentation image includes: using the average voxel value corresponding to all pixels in the blood segmentation image as the average voxel value of the PET image; or, using the voxel value of a specified pixel in the blood segmentation image as the average voxel value of the PET image.

[0094] S250. Construct a blood input function based on the scanning time and average voxel value of each PET image frame.

[0095] In one optional embodiment, a blood input function is constructed based on the scanning time and average voxel value of each PET image frame, including: sorting the average voxel values ​​corresponding to each PET image frame according to the scanning time of each PET image frame to obtain the blood input function; or, using a preset fitting function, fitting the blood input function based on the scanning time and average voxel value of each PET image frame to obtain the blood input function.

[0096] In this embodiment, the average voxel value is used to characterize the measured activity value of the tracer in the blood at the scanning time. For example, the preset fitting function can be an exponential function; however, the specific preset fitting function used is not limited here and can be customized according to actual needs.

[0097] S260. Determine the tracer output function based on the blood input function and the two-compartment dynamic model.

[0098] S270. Based on the dynamic PET image set and tracer output function, the dynamic parameters of the two-compartment dynamic model are obtained by iteratively solving the two-compartment dynamic model.

[0099] S280. Determine the dynamic parameter image based on the model's dynamic parameters.

[0100] S260-S280 in this embodiment are the same as those in the above embodiments. Figure 1 The S120-S130 shown are the same or similar, and will not be described again in this embodiment.

[0101] Based on the above embodiments, optionally, after determining the kinetic parameter image according to the model dynamic parameters, the method further includes: when the kinetic parameter image includes a net uptake rate constant image, determining the tissue image region corresponding to the tissue compartment according to the dynamic PET image set; cropping the net uptake rate constant image according to the tissue image region to obtain a tissue segmentation image; and determining the classification label of the tissue to be tested corresponding to the tissue segmentation image according to the tissue segmentation image.

[0102] In one optional embodiment, determining the tissue image region corresponding to the tissue compartment based on the dynamic PET image set includes: acquiring a late PET image set corresponding to the dynamic PET image set; and determining the tissue image region corresponding to the tissue compartment based on the late PET image set.

[0103] In this embodiment, the late PET image set refers to the set of PET images in the late time range after tracer injection, and the scanning time corresponding to each PET image in the late PET image set is later than the scanning time of other PET images outside the late PET image set.

[0104] For example, when the preset duration is 30 minutes, the late time range can be [20 minutes, 30 minutes] or [25 minutes, 30 minutes]. There is no limitation on the late time range here, and it can be customized according to actual needs.

[0105] For example, the method for determining tissue image regions is similar to that for determining blood image regions. For instance, assuming the preset segmentation algorithm is a threshold-based segmentation algorithm, the pixel threshold can be the product of the maximum SUV (Standardized Uptake Value) corresponding to the late PET image set and a preset ratio, which can be 40%. Here, SUV represents the degree of tissue uptake of the tracer, and SUV = voxel value / injection dose / body weight of the sampled subject.

[0106] In one optional embodiment, determining the classification label of the tissue to be tested corresponding to the tissue segmentation image based on the tissue segmentation image includes: using the average voxel value corresponding to all pixels in the tissue segmentation image as a quantitative parameter value; if the quantitative parameter value is greater than a quantitative threshold, then setting the classification label of the tissue to be tested corresponding to the tissue segmentation image as a first label; if the quantitative parameter value is less than or equal to the quantitative threshold, then setting the classification label of the tissue to be tested corresponding to the tissue segmentation image as a second label.

[0107] The technical solution of this embodiment obtains an early PET image set corresponding to a dynamic PET image set, determines the blood image region corresponding to the blood compartment based on the early PET image set, determines the average voxel value corresponding to each frame of PET image in the dynamic PET image set based on the blood image region, and constructs a blood input function based on the scanning time and average voxel value of each frame of PET image. This solves the problem of the invasiveness of constructing the blood input function and achieves the purpose of non-invasive determination of dynamic parameters.

[0108] In some instances, K i -30 minutes represents the net uptake rate constant image determined from a 30-minute dynamic PET image set, K i -65 minutes represents the net uptake rate constant image determined based on a 65-minute dynamic PET image set. For example, based on a 30-minute dynamic PET image set, seven more PET images are acquired at 300-second intervals to form a 65-minute dynamic PET image set. K i -30 minutes and K i The procedures for determining the net uptake rate constant graphs corresponding to each of the -65 minutes are similar or identical.

[0109] For K i -30 minutes and K iImage quality is evaluated over a period of 65 minutes, including but not limited to artifact reduction, noise suppression, contrast preservation, tissue segmentation, and overall quality. i -30 minutes and K i The image quality corresponding to both the -65-minute and -65-minute images met clinical requirements. Since tissue segmentation relies on the later PET image set from the dynamic PET image set, in this example, the number of tissue image regions segmented from the 30-minute dynamic PET image set was the same as the number segmented from the 65-minute dynamic PET image set. Furthermore, by analyzing K... i -30 minutes and K i A difference test was performed on the image quality at -65 minutes, and no significant difference was found. This was achieved by analyzing K... i -30 minutes and K i Correlation analysis was performed on the image quality corresponding to each image at -65 minutes, and the correlation was found to exist. The correlation coefficients used in the correlation analysis included Pearson correlation coefficient and / or Spearman rank correlation coefficient.

[0110] Figure 4 A K provided as an embodiment of the present invention i -30 minutes and K i The ROC curves for each of the -65 minutes are as follows. Specifically, the ROC curve (Receiver Operating Characteristic curve) represents a curve plotted with the false positive rate (FPR) on the x-axis and the true positive rate (TPR) on the y-axis for different pixel thresholds. Figure 4 In this context, "sensitivity" refers to the true positive rate, and "specificity" refers to the false positive rate. The true positive rate is calculated as 100 - specificity.

[0111] from Figure 4 We can obtain K i The optimal pixel threshold corresponding to -30 minutes is 0.018 ml / g / min, with an AUC (Area Under Curve) of 0.816 (95% confidence interval: 0.743–0.875), a sensitivity of 69.50%, and a specificity of 83.30%. i The optimal pixel threshold corresponding to -65 minutes was 0.022 ml / g / min, with an AUC of 0.816 (95% confidence interval: 0.744–0.876), a sensitivity of 66.40%, and a specificity of 83.30%.

[0112] Furthermore, regarding K i -30 minutes and K i A Delong test was performed at -65 minutes, and the test results showed that K... i -30 minutes and K i There was no significant difference in classification accuracy between -65 minutes and 65 minutes.

[0113] The following are embodiments of the device for determining dynamic parameters based on PET imaging provided in this invention. This device and the method for determining dynamic parameters based on PET imaging described above belong to the same inventive concept. For details not described in detail in the embodiments of the device for determining dynamic parameters based on PET imaging, please refer to the content of the method for determining dynamic parameters based on PET imaging described above.

[0114] Figure 5 This is a schematic diagram of a device for determining dynamic parameters based on PET imaging, provided as an embodiment of the present invention. Figure 5 As shown, the device includes: a blood input function acquisition module 310, a tracer output function determination module 320, a model dynamic parameter solution module 330, and a dynamic parameter image determination module 340.

[0115] The blood input function acquisition module 310 is used to acquire a dynamic PET image set and acquire the blood input function.

[0116] The tracer output function determination module 320 is used to determine the tracer output function based on the blood input function and the two-compartment dynamic model;

[0117] The model dynamic parameter solving module 330 is used to iteratively solve the two-compartment dynamic model to obtain the model dynamic parameters of the two-compartment dynamic model based on the dynamic PET image set and the tracer output function.

[0118] The dynamic parameter image determination module 340 is used to determine the dynamic parameter image based on the model's dynamic parameters;

[0119] The dynamic PET image set represents the set of PET images within a preset time after tracer injection, with the preset time being greater than or equal to 30 minutes and less than 60 minutes. The two-compartment dynamic model corresponds to the blood compartment and the tissue compartment. The blood input function represents the activity value of the tracer in the blood compartment at different scanning times, and the tracer output function represents the sum of the activity values ​​of the tracer in the blood compartment and the tracer in the tissue compartment.

[0120] The technical solution of this embodiment determines the tracer output function based on the blood input function and the two-compartment dynamic model. Iteratively solving the two-compartment dynamic model based on the dynamic PET image set and the tracer output function yields the model dynamic parameters. Based on these parameters, a kinetic parameter image is determined. The dynamic PET image set represents the PET images taken within a preset time after tracer injection. The preset time is greater than or equal to 30 minutes and less than 60 minutes. This approach ensures the accuracy of the kinetic parameter image while shortening the imaging time of dynamic PET imaging, solving the problem of long imaging times required by the scanning protocol of dynamic PET imaging and improving the throughput and efficiency of PET imaging equipment.

[0121] In an optional embodiment, the tracer output function determination module 320 includes:

[0122] The tissue activity function determination unit is used to determine the tissue activity function based on the blood input function and model dynamic parameters;

[0123] The tracer output function determination unit is used to perform a weighted summation of the blood input function and the tissue activity function to obtain the tracer output function;

[0124] Among them, the tissue activity function represents the activity value of the tracer in the tissue compartment at different scanning times, and the weight corresponding to the blood input function is the blood volume percentage.

[0125] In one optional embodiment, the tissue activity function determination unit is specifically used for:

[0126] Determine the instantaneous response function based on the model's dynamic parameters;

[0127] The convolution result of the blood input function and the transient response function is used as the tissue activity function;

[0128] In this model, the tracer is an irreversible tracer, and correspondingly, the model dynamic parameters do not include the dephosphorylation rate, which represents the rate constant of phosphorylated tracer being converted into free tracer in tissue compartments.

[0129] In an optional embodiment, the model dynamic parameter solving module 330 is specifically used for:

[0130] The objective function is determined based on the dynamic PET image set and the tracer output function;

[0131] The least squares fitting algorithm is used to minimize the objective function to obtain the model dynamic parameters of the two-compartment dynamic model;

[0132] The objective function represents the sum of squared errors between the voxel values ​​of all pixels in the dynamic PET image set at each scan time and the activity value of the tracer output function at the scan time.

[0133] In an optional embodiment, the blood input function acquisition module 310 is specifically used for:

[0134] Obtain the early PET image set corresponding to the dynamic PET image set; wherein, the scan time corresponding to each PET image in the early PET image set is earlier than the scan time of other PET images outside the early PET image set;

[0135] Based on early PET image sets, the blood image regions corresponding to blood compartments were determined;

[0136] Based on the blood image region, determine the average voxel value corresponding to each frame of PET image in the dynamic PET image set;

[0137] A blood input function is constructed based on the scan time and average voxel value of each PET image frame.

[0138] In an optional embodiment, the device further includes:

[0139] The classification label determination module is used to determine the tissue image region corresponding to the tissue compartment based on the dynamic PET image set when the dynamic parameter image includes a net uptake rate constant image after determining the kinetic parameter image based on the model dynamic parameters.

[0140] Based on the tissue image region, the net uptake rate constant image is cropped to obtain the tissue segmentation image;

[0141] Based on the tissue segmentation image, determine the classification label of the tissue to be tested corresponding to the tissue segmentation image.

[0142] The device for determining dynamic parameters based on PET imaging provided in this embodiment of the invention can execute the method for determining dynamic parameters based on PET imaging provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0143] Figure 6This is a schematic diagram of an electronic device provided according to one embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0144] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor 11. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0145] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information or data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0146] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for determining kinetic parameters based on PET imaging provided in the above embodiments.

[0147] In some embodiments, the method for determining PET-based imaging dynamic parameters provided in the above embodiments can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for determining PET-based imaging dynamic parameters described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for determining PET-based imaging dynamic parameters by any other suitable means (e.g., by means of firmware).

[0148] Various embodiments of the systems and techniques described above herein can be implemented in the following systems or combinations thereof: digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0149] Computer programs for implementing the method for determining kinetic parameters based on PET imaging of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0150] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable storage medium. Examples of machine-readable storage media include, based on an electrical connection of at least one wire, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0151] To provide interaction with a user, the systems and techniques described herein can be implemented on a terminal device having: a display device for displaying information to the user (e.g., a cathode-ray tube (CRT) or liquid crystal display (LCD) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the terminal device. Other types of devices can also provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0152] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0153] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0154] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0155] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining dynamic parameters based on PET imaging, characterized in that, include: Acquire a dynamic PET image set and obtain the blood input function; Based on the blood input function and the two-compartment dynamic model, determine the tracer output function; Based on the dynamic PET image set and the tracer output function, the dynamic parameters of the two-compartment dynamic model are obtained by iteratively solving the two-compartment dynamic model. Based on the model's dynamic parameters, determine the dynamic parameter image; Wherein, the dynamic PET image set represents the PET image set within a preset time period after tracer injection, the preset time period being greater than or equal to 30 minutes and less than 60 minutes, the two-compartment dynamic model corresponding to the blood compartment and the tissue compartment, the blood input function representing the activity value of the tracer in the blood compartment at different scanning times, and the tracer output function representing the sum of the activity values ​​of the tracer in the blood compartment and the activity values ​​of the tracer in the tissue compartment; The step of determining the tracer output function based on the blood input function and the two-compartment dynamic model includes: The tissue activity function is determined based on the blood input function and the model dynamic parameters; The tracer output function is obtained by weighted summation of the blood input function and the tissue activity function; Wherein, the tissue activity function represents the activity value of the tracer in the tissue compartment at different scanning times, and the weight corresponding to the blood input function is the blood volume percentage; The tracer output function Represented as: ； Wherein, CBV represents the percentage of blood volume. This represents the blood input function. Represents the tissue activity function; After determining the dynamic parameter image based on the model's dynamic parameters, the method further includes: When the kinetic parameter image includes a net uptake rate constant image, the tissue image region corresponding to the tissue compartment is determined based on the dynamic PET image set; Based on the tissue image region, the net uptake rate constant image is cropped to obtain a tissue segmentation image; Based on the tissue segmentation image, determine the classification label of the tissue to be tested corresponding to the tissue segmentation image; The step of determining the classification label of the tissue to be tested corresponding to the tissue segmentation image includes: The average voxel value corresponding to all pixels in the tissue segmentation image is used as the quantitative parameter value. If the quantitative parameter value is greater than the quantitative threshold, the classification label of the tissue to be tested corresponding to the tissue segmentation image is set as the first label. If the quantitative parameter value is less than or equal to the quantitative threshold, the classification label of the tissue to be tested corresponding to the tissue segmentation image is set as the second label.

2. The method according to claim 1, characterized in that, The step of determining the tissue activity function based on the blood input function and the model dynamic parameters includes: Determine the instantaneous response function based on the model's dynamic parameters; The convolution result of the blood input function and the instantaneous response function is used as the tissue activity function; Wherein, the tracer is an irreversible tracer, and correspondingly, the model dynamic parameters do not include the dephosphorylation rate, which represents the rate constant of the conversion of phosphorylated tracer in the tissue compartment to free tracer.

3. The method according to claim 1, characterized in that, The step of iteratively solving the two-compartment dynamic model based on the dynamic PET image set and the tracer output function to obtain the model dynamic parameters of the two-compartment dynamic model includes: The target function is determined based on the dynamic PET image set and the tracer output function; The least squares fitting algorithm is used to minimize the objective function to obtain the model dynamic parameters of the two-compartment dynamic model; The objective function represents the sum of squared errors between the voxel values ​​of all pixels in the PET image corresponding to each scanning time in the dynamic PET image set and the activity value of the tracer output function at the scanning time.

4. The method according to claim 1, characterized in that, The function for obtaining blood input includes: Obtain the early PET image set corresponding to the dynamic PET image set; wherein, the scanning time of each PET image in the early PET image set is earlier than the scanning time of other PET images outside the early PET image set; Based on the early PET image set, determine the blood image region corresponding to the blood compartment; Based on the blood image region, determine the average voxel value corresponding to each frame of PET image in the dynamic PET image set; A blood input function is constructed based on the scan time and average voxel value of each PET image frame.

5. A device for determining dynamic parameters based on PET imaging, characterized in that, include: The blood input function acquisition module is used to acquire dynamic PET image sets and obtain blood input functions; The tracer output function determination module is used to determine the tracer output function based on the blood input function and the two-compartment dynamic model. The model dynamic parameter solving module is used to iteratively solve the two-compartment dynamic model to obtain the model dynamic parameters of the two-compartment dynamic model based on the dynamic PET image set and the tracer output function. The dynamic parameter image determination module is used to determine the dynamic parameter image based on the model's dynamic parameters; Wherein, the dynamic PET image set represents the PET image set within a preset time period after tracer injection, the preset time period being greater than or equal to 30 minutes and less than 60 minutes, the two-compartment dynamic model corresponding to the blood compartment and the tissue compartment, the blood input function representing the activity value of the tracer in the blood compartment at different scanning times, and the tracer output function representing the sum of the activity values ​​of the tracer in the blood compartment and the activity values ​​of the tracer in the tissue compartment; The tracer output function determination module includes: The tissue activity function determination unit is used to determine the tissue activity function based on the blood input function and model dynamic parameters; The tracer output function determination unit is used to perform a weighted summation of the blood input function and the tissue activity function to obtain the tracer output function; Among them, the tissue activity function represents the activity value of the tracer in the tissue compartment at different scanning times, and the weight corresponding to the blood input function is the blood volume percentage; The tracer output function Represented as: ； Wherein, CBV represents the percentage of blood volume. This represents the blood input function. Represents the tissue activity function; The classification label determination module is used to determine the tissue image region corresponding to the tissue compartment based on the dynamic PET image set when the kinetic parameter image includes a net uptake rate constant image; Based on the tissue image region, the net uptake rate constant image is cropped to obtain a tissue segmentation image; Based on the tissue segmentation image, determine the classification label of the tissue to be tested corresponding to the tissue segmentation image; The classification label determination module is specifically used to take the average voxel value corresponding to all pixels in the tissue segmentation image as a quantitative parameter value; if the quantitative parameter value is greater than the quantitative threshold, the classification label of the tissue to be tested corresponding to the tissue segmentation image is set as the first label; if the quantitative parameter value is less than or equal to the quantitative threshold, the classification label of the tissue to be tested corresponding to the tissue segmentation image is set as the second label.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, which enables the at least one processor to perform the method for determining kinetic parameters based on PET imaging as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for determining the dynamic parameters based on PET imaging as described in any one of claims 1-4.

8. A computer program product comprising a computer program that, when executed by a processor, implements a method for determining kinetic parameters based on PET imaging according to any one of claims 1-4.

Images