Apparatus and method for enabling quantitative thyroid SPECT without CT using a neural network

Abstract

An apparatus and method are disclosed that enable quantitative thyroid SPECT without CT using a neural network according to an embodiment. An apparatus that enables quantitative thyroid SPECT without CT using a neural network according to an embodiment includes a generation unit that inputs a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid into a first model to generate an attenuation map; a segmentation unit that segments a thyroid map by inputting the attenuation map and the first SPECT image into a second model; and a calculation unit that calculates an uptake amount of a radiopharmaceutical of the thyroid using a first SPECT image sinogram, a second SPECT image sinogram, the attenuation map, and the thyroid map.

Description

Technical Field

[0001] The present disclosure relates to a technique that enables quantitative thyroid SPECT without CT using a neural network.

Background Art

[0002] Thyroid function is evaluated by measuring the amount (% uptake) of a radiopharmaceutical injected intravenously into a thyroid patient that is taken up by the thyroid. Typically, measurement of the uptake amount of a radiopharmaceutical is mainly evaluated by a thyroid uptake system, but there is a limitation in that the accuracy is low.

[0003] Therefore, SPECT / CT (Single-Photon Emission Computed Tomography / Computed Tomography), which has been reported to have higher accuracy in measuring the uptake amount than the thyroid uptake system, has attracted attention.

[0004] SPECT / CT is a nuclear medicine imaging method that provides quantitative information using a radiopharmaceutical, for example, Tc-99m pertechnetate. SPECT / CT uses CT for quantitative measurement. Here, CT performs attenuation correction to accurately measure the uptake amount of the thyroid and is used for thyroid segmentation.

[0005] However, the method for evaluating thyroid function using SPECT / CT induces a radiation exposure of about 3.34 mSv in total, combining the radiation exposure due to CT of about 1.12 mSv and the radiation exposure due to SPECT of about 2.22 mSv. Furthermore, there is a problem that thyroid segmentation is troublesome because it needs to be segmented by an expert each time.

Summary of the Invention

[0006] The disclosed embodiments aim to enable quantitative thyroid SPECT without CT using a neural network. [Means for solving the problem]

[0007] An apparatus that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment comprises: a generation unit that inputs a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid gland into a first model to generate an attenuation map; a segmentation unit that inputs the attenuation map and the first SPECT image into a second model to segment the thyroid map; and A first SPECT sinogram corresponding to the raw data of the first SPECT image, and a second SPECT sinogram corresponding to the raw data of the second SPECT image. The system includes a calculation unit that calculates the amount of radiopharmaceutical ingested by the thyroid gland using the attenuation map and the thyroid map.

[0008] The first SPECT image may be a primary emission SPECT image.

[0009] The second SPECT image may be a scattering SPECT image.

[0010] The generation unit uses the attenuation map The aforementioned first SPECT sinogram Apply attenuation correction (AC) to the signal. The second SPECT sinogram Using The aforementioned first SPECT sinogram Scatter Correction (SC) and Resolution Recovery (RR) are performed. The aforementioned first SPECT sinogram A third SPECT image may be generated by applying this to [the specified method].

[0011] The third SPECT image may be a quantitative ACSCRR SPECT image.

[0012] The calculation unit may combine the third SPECT image and the thyroid map to calculate the amount of radiopharmaceutical absorbed by the thyroid gland.

[0013] The calculation unit may calculate the amount of radiopharmaceutical absorbed by the thyroid gland by combining the third SPECT image and the thyroid map and counting matching voxels.

[0014] The second model described above may be trained to segment the thyroid map in an input image based on multiple thyroid segmentation maps drawn and labeled along the outline of the thyroid gland in a CT (Computed Tomography) image.

[0015] A method for enabling quantitative thyroid SPECT without CT using a neural network according to one embodiment is a method performed using an apparatus that enables quantitative thyroid SPECT without CT using a neural network, the apparatus comprising one or more processors and a memory storing one or more programs executed by the one or more processors, comprising the steps of: inputting a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid gland into a first model to generate an attenuation map; and inputting the attenuation map and the first SPECT image into a second model to segment the thyroid map. A first SPECT sinogram corresponding to the raw data of the first SPECT image, and a second SPECT sinogram corresponding to the raw data of the second SPECT image. The process includes the step of calculating the amount of radiopharmaceutical ingested by the thyroid gland using the attenuation map and the thyroid map.

[0016] The first SPECT image may be a primary emission SPECT image.

[0017] The second SPECT image may be a scattering SPECT (Scattering SPECT) image.

[0018] The generating step uses the attenuation map The first SPECT sinogram to perform attenuation correction (Attenuation Correction; AC), The second SPECT sinogram uses The first SPECT sinogram to perform scatter correction (Scatter Correction; SC) on, and applies resolution recovery (Resolution Recovery; RR) to The first SPECT sinogram to generate a third SPECT image. The third SPECT image may be a quantitative ACSCRR SPECT image.

[0019] The calculating step may include a step of combining the third SPECT image and the thyroid map to calculate the uptake amount of the radiopharmaceutical in the thyroid.

[0020] The calculating step may include a step of combining the third SPECT image and the thyroid map and counting matching voxels to calculate the uptake amount of the radiopharmaceutical in the thyroid.

[0021] The second model may be learned to segment the thyroid map in the input image based on a plurality of thyroid segmentation maps drawn and labeled along the outline of the thyroid in a CT (Computed Tomography) image. [Effect of the Invention]

[0022] The disclosed embodiment can reduce the radiation exposure by CT based on the ALARA (As Low As Reasonably Achievable) principle by performing attenuation correction using a neural network instead of CT.

[0023] The disclosed embodiments can segment a thyroid map using a neural network and provide a thyroid map without human intervention. [Brief explanation of the drawing]

[0024] [Figure 1] This is a block diagram illustrating an apparatus that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment. [Figure 2] This is a block diagram illustrating a learning model for a device that enables quantitative thyroid SPECT without CT using a neural network, according to one embodiment. [Figure 3a-4c] This is an illustrative diagram showing the performance of a device that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment. [Figure 5] This is a flowchart illustrating a method for enabling quantitative thyroid SPECT without CT using a neural network. [Modes for carrying out the invention]

[0025] The terminology used herein has been selected to reflect its function and, to the greatest extent possible, to be widely used and common. However, this may change depending on the intentions and practices of engineers in the art or the emergence of new technologies. In certain cases, the applicant has also arbitrarily selected some terms, in which case their meanings will be described in the corresponding descriptive section of the specification. Therefore, it should be made clear that the terminology used herein is not merely a set of names, but should be interpreted based on the substantive meaning of the term and the overall content of this specification.

[0026] The terms "first" and "second" used in this application may be used to describe various components, but the components are not limited by such terms. The terms are used solely for the purpose of distinguishing one component from another; for example, without departing from the scope of rights under the concept of the present invention, the first component may be referred to as the second component, and similarly, the second component may be referred to as the first component.

[0027] In this application, singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as “includes,” “equip,” and “possess” indicate the existence of the components or combinations thereof described in the specification, and do not preclude the possibility of other components or features being present or added.

[0028] Furthermore, the embodiments described herein may have aspects that are entirely hardware, partially hardware and partially software, or entirely software. In this specification, "unit," "module," "device," "server," or "system" refers to computer-related entities such as hardware, a combination of hardware and software, or software. For example, a unit, module, device, server, or system may refer to hardware and / or software such as an application for driving said hardware that constitute part or all of a platform.

[0029] The following describes practical examples in detail with reference to the attached drawings and their contents, but the scope claimed is not limited or restricted by the examples.

[0030] Figure 1 is a block diagram illustrating a device 100 that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment.

[0031] Referring to Figure 1, the apparatus 100, which enables quantitative thyroid SPECT without CT using a neural network, includes a generation unit 110, a segmentation unit 120, and a calculation unit 130.

[0032] The generation unit 110, the segmentation unit 120, and the calculation unit 130 are implemented using one or more physically separated devices, but may also be implemented by one or more processors or a combination of one or more processors and software, and unlike the illustrated example, they may not be clearly separated in their specific operation.

[0033] The generation unit 110 inputs a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid gland into the first model to generate an attenuation map.

[0034] Single-photon emission computed tomography (SEM) is a nuclear medicine imaging technique that involves injecting a radiopharmaceutical that emits a single photon (gamma ray) into the body, and then measuring the gamma rays transmitted through the body to reconstruct the distribution of the single-photon-emitting nuclide within the body as an image. In particular, SEM can provide images of biological functions, which are biochemical phenomena in living organisms.

[0035] The first SPECT image may be a primary emission SPECT image generated using single-photon emission computed tomography.

[0036] A primary emission SPECT image may be a SPECT image for primary gamma rays, which are the most frequent type of gamma ray detected by single-photon emission computed tomography.

[0037] The second SPECT image may be a scattered SPECT image generated using single-photon emission computed tomography. A scattered SPECT image may be a SPECT image of gamma rays scattered during single-photon emission computed tomography.

[0038] At least one of the first SPECT image and the second SPECT image is used to improve accuracy by analyzing the collimator-detector response. In other words, resolution The SPECT image may be a SPECT image of NCRR corrected for recovery. Specifically, at least one of the first SPECT image and the second SPECT image may be an image corrected for resolution recovery using a Butterworth low-pass filter to reduce statistical noise. On the other hand, the first SPECT sinogram and the second SPECT sinogram may be raw data obtained from a single-photon emission tomography system, and may be used to reconstruct the first SPECT image and the second SPECT image, respectively.

[0039] At least one of the first SPECT image and the second SPECT image may be an image normalized by the maximum value of the combined image of the two SPECTs.

[0040] The generation unit 110 uses an attenuation map First SPECT sinogram corresponding to the raw data of the first SPECT image A third SPECT image may be generated by applying attenuation correction (AC) to the first image.

[0041] The generation unit 110 may generate a third SPECT image by performing at least one of the following using an attenuation map: attenuation correction, scatter correction (SC), and resolution recovery (RR).

[0042] In this case, the third SPECT image may be a quantitative ACSCRR SPECT image. A quantitative ACSCRR SPECT image is a numerical representation of the results for primary gamma rays. SPECT Cynogram The correction was applied. Later generated SPECT images are acceptable.

[0043] The segmentation unit 120 inputs the attenuation map and the first SPECT image into the second model to segment the thyroid map.

[0044] The calculation unit 130 is, The first SPECT sinogram, and the second SPECT sinogram corresponding to the raw data of the second SPECT image. The amount of radiopharmaceuticals absorbed by the thyroid gland is calculated using attenuation maps and thyroid maps.

[0045] The calculation unit 130 may combine the third SPECT image and the thyroid map to calculate the amount of radiopharmaceutical absorbed by the thyroid gland.

[0046] The calculation unit 130 may calculate the amount of radiopharmaceuticals absorbed by the thyroid gland by combining the third SPECT image and the thyroid map and counting the matching voxels.

[0047] The calculation unit 130 may calculate the amount of radiopharmaceutical absorbed by the thyroid by counting matching voxels between the third SPECT image and the thyroid map, treating negative voxel values ​​as 0.

[0048] Figure 2 is a block diagram illustrating the learning model of a device 100 that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment.

[0049] Referring to Figure 2, the first SPECT image 211 and the second SPECT image 212 are input to the first model 220. The second model 230 is input to the first SPECT image 211 and the attenuation map output by the first model 220.

[0050] Specifically, the first model 220 uses one or more neural networks to output an attenuation map provided by CT as an output value, even without CT, when the first SPECT image 211 and the second SPECT image 212 are input.

[0051] The first model 220 may include a loss function based on the following mathematical formula 1. [Mathematical formula 1] L(G(X), Y)=L error (G(X, Y)+L GDL (G(X), Y)

[0052] Here, the loss function is L(G(X), Y), where Y is the ground truth of the attenuation map, X is the input SPECT, G(X) is the generated attenuation map, L error L may be a first or second loss function that has already been defined. GDL This may be a GDL (Gradient Difference Loss) for the clarity of the generated attenuation map.

[0053] For example, the first loss function may be a function that calculates the sum of the differences between the ground truth of the attenuation map and the absolute value between the generated attenuation maps. In other words, the first loss function (L1) may be set based on the following mathematical formula 2. [Mathematical formula 2] JPEG0007902340000001.jpg6170

[0054] For example, the second loss function may be a function that calculates the sum of squared differences between the ground truth of the attenuation map and the generated attenuation map. In other words, the second loss function (L2) may be set based on the following mathematical formula 3. [Mathematical formula 3] JPEG0007902340000002.jpg6170

[0055] For example, LGDL may be used to compensate for imagining blur due to the loss effect of the second loss function, and LGDL may be set based on the following mathematical formula 4. [Mathematical formula 4] JPEG0007902340000003.jpg7170

[0056] In this case, the first model 220 may be trained based on a hyperfunction attenuation map obtained by imaging a thyroid gland judged to be hyperfunctioning using CT, a hypofunction attenuation map obtained by imaging a thyroid gland judged to be hypofunctioning using CT, and a standard function attenuation map obtained by imaging a thyroid gland judged to be of standard function using CT.

[0057] CT scans may be performed to cover the entire axial field of view of the SPECT scan, or they may be performed to cover only 1 / 2 to 2 / 3 of the SPECT axial field of view to reduce unnecessary radiation exposure from the CT scan. It is preferable that SPECT and CT scans be performed in the same axial field of view from the mid-skull to the upper mediastinum.

[0058] CT or SPECT images may be labeled with a clinical diagnosis by a nuclear medicine specialist for at least one of the following conditions: Graves' disease / hyperthyroidism, painless / subacute thyroiditis, SNG (Single Nodular Goiter) / MNG (Multi Nodular Goiter), drug-induced thyroiditis, and lingual thyroid.

[0059] The second model 230 may be designed to use one or more neural networks to segment a thyroid map as an output value when the first SPECT image 211 and the attenuation map output by the first model 220 are input.

[0060] The second model 230 may include a loss function based on categorical cross-entropy (CCE).

[0061] Specifically, the second model 230 may include a loss function based on mathematical formula 5. [Mathematical formula 5] JPEG0007902340000004.jpg10170

[0062] At this time, y i is ground truth, y h 'n' may be synthetic, and 'n' may be the number of classes in the second Model 230.

[0063] For example, y i is the ground truth, represented by 0 or 1, and y h n is a probability value between 0 and 1, and n can be set to one of three values, such as background, left thyroid gland, or right thyroid gland.

[0064] In this case, the second model 230 may be trained to segment the thyroid maps in the input image based on multiple thyroid maps drawn and labeled along the outline of the thyroid gland in the CT image.

[0065] At least one of the first model 220 and the second model 230 may be a U-net structure, preferably a 3D U-net substrate structure. In other words, the first model 220 and the second model 230 may be designed with a structure that skips the contraction path and the segmentation path.

[0066] As a specific example, at least one of the first model 220 and the second model 230 may be designed to include 64 initial neurons and 4 skip connections. At least one of the first model 220 and the second model 230 may be a U-net structure-based neural network that skip-connects a contraction path including one or more folding blocks including a 3x3 convolutional layer, batch normalization, and a ReLU function, and a segmentation path including one or more of the folding blocks.

[0067] In this case, the folding blocks in the contraction path may be subjected to maximum pooling, preferably 2x2x2 stride maximum pooling, and the folding blocks in the segmentation path may be subjected to transposed convolution, preferably 2x2x2 upconvolution.

[0068] At least one of the first model 220 and the second model 230 may further include a 1×1×1 convolutional layer as the final output layer.

[0069] The first model 220 and the second model 230 may be designed to be end-to-end learned.

[0070] On the other hand, while the first model 220 and the second model 230 were described as U-net structures widely used in the medical field, these are merely illustrative examples. They can be designed based on a variety of already known structures such as Seg-NET, MFU-net, and Generative Adversarial Networks (GANs), and are not necessarily limited to U-net structures.

[0071] Figures 3a to 4c are illustrative diagrams showing the performance of a learning model for a device 100 that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment.

[0072] Figure 3a is an illustrative diagram showing a comparison between ground truth and synthetic attenuation maps.

[0073] The ground truth map on the left is an attenuation map generated by CT scanning, while the synthetic map on the right is an attenuation map generated by a device 100 that enables quantitative thyroid SPECT without CT scans according to one embodiment.

[0074] Figure 3b shows the correlation between the attenuation coefficients of the ground truth and synthetic attenuation maps in Figure 3a. Figure 3b shows a high correlation between the ground truth and synthetic attenuation maps.

[0075] Figure 4a shows the apparatus 100 that enables quantitative thyroid SPECT without CT according to one embodiment, and the amount of radiopharmaceuticals absorbed by the thyroid as measured by an existing SPECT / CT.

[0076] In Figure 4a, the horizontal axis represents the amount of radiopharmaceuticals absorbed by the thyroid gland as measured by SPECT / CT, and the vertical axis represents the amount of radiopharmaceuticals absorbed by the device 100, which enables quantitative thyroid SPECT without CT according to one embodiment.

[0077] The correlation between the intake of radiopharmaceuticals of the thyroid gland measured by existing SPECT / CT and the intake of radiopharmaceuticals of the thyroid gland output by the device 100, which enables quantitative thyroid SPECT without CT according to one embodiment, shows r=0.9980, R^2=0.9959, and p<0.0001 in the intake range of 0 to 30. Here, r is the correlation coefficient in the Pearson correlation, and p represents significance.

[0078] Figure 4b is a Bland-altman plot chart showing the difference between the intake of thyroid radiopharmaceuticals measured by SPECT / CT and the intake of thyroid radiopharmaceuticals output by the device 100, which enables quantitative thyroid SPECT without CT according to one embodiment.

[0079] The horizontal axis of Figure 4b represents the average value between the amount of radiopharmaceutical intake of the thyroid gland measured by SPECT / CT and the amount of radiopharmaceutical intake of the thyroid gland output by the device 100 that enables quantitative thyroid SPECT without CT according to one embodiment. The vertical axis represents the difference between the amount of radiopharmaceutical intake of the thyroid gland measured by SPECT / CT and the amount output by the device 100 that enables quantitative thyroid SPECT without CT according to one embodiment.

[0080] The Brand-Altman plot chart in Figure 4b shows a bias of -0.99% and a non-significant systematic deviation, demonstrating the high accuracy of the device 100, which enables quantitative thyroid SPECT without CT according to one embodiment.

[0081] Figure 4c is a graph showing the intake of radiopharmaceuticals into the thyroid gland by SPECT / CT for different thyroid diseases and by a device 100 that enables quantitative thyroid SPECT without CT according to one example.

[0082] In this case, the amount of radiopharmaceuticals absorbed by the thyroid gland in SPECT / CT for each thyroid disease and in the device 100 that enables quantitative thyroid SPECT without CT according to one embodiment is shown as mean ± standard deviation.

[0083] Thyroid diseases can be distinguished by the absorption rate of the thyroid gland. Referring to Figure 4b, the intake of radiopharmaceuticals to the thyroid gland by SPECT / CT and a device 100 that enables quantitative thyroid SPECT without CT in one embodiment shows the intake of radiopharmaceuticals for, for example, Graves' disease / hyperthyroidism, painless / subacute thyroiditis, SNG (Single Nodular Goiter) / MNG (Multi Nodular Goiter), and others (drug-induced thyroiditis, lingual thyroid).

[0084] Figure 5 is a flowchart illustrating a method for enabling quantitative thyroid SPECT without CT using a neural network according to one embodiment.

[0085] Referring to Figure 5, the method for enabling quantitative thyroid SPECT without CT using a neural network according to one embodiment may be performed using the quantitative thyroid SPECT device without CT using a neural network shown in Figure 1.

[0086] First, a quantitative thyroid SPECT system without CT using a neural network inputs a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid gland into a first model to generate an attenuation map (510).

[0087] Next, a quantitative thyroid SPECT system without CT using a neural network inputs the attenuation map and the first SPECT image into a second model to segment the thyroid map (520).

[0088] Next, a quantitative thyroid SPECT device without CT using a neural network was developed. The first SPECT sinogram corresponds to the raw data of the first SPECT image, and the second SPECT sinogram corresponds to the raw data of the second SPECT image. The amount of radiopharmaceuticals absorbed by the thyroid gland is calculated using the attenuation map and the thyroid map (530).

[0089] The method illustrated in Figure 5 was described with reference to the flowchart presented in the same figure. For illustrative purposes, the method was illustrated and described in a series of blocks, but the present invention is not limited to the order of the blocks, and some blocks may be executed in a different order or simultaneously with other blocks than those illustrated and described herein, and a variety of other branches, flow paths, and block orders may be realized to achieve the same or similar results. Furthermore, not all blocks illustrated are required to realize the method described herein.

[0090] Furthermore, a method according to one embodiment of the present invention may be embodied in the form of a computer program for performing a series of steps, the computer program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specially configured to store and execute program instructions, such as ROMs, RAMs, and flash memory.

[0091] As described above with reference to examples, those skilled in the art will understand that the present invention can be modified and altered in various ways without departing from the spirit and scope of the invention as described in the following claims. [Industrial applicability]

[0092] An apparatus and method that enables quantitative thyroid SPECT without CT using a neural network according to one embodiment can be used in the medical diagnostics industry in the field of nuclear medicine by performing attenuation correction using a neural network instead of CT.

Claims

1. A generation unit that inputs a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid gland into a first model to generate an attenuation map; A segmentation unit that inputs the attenuation map and the first SPECT image into a second model to segment the thyroid map; and An apparatus that enables quantitative thyroid SPECT without CT using a neural network, comprising a calculation unit that calculates the amount of radiopharmaceutical ingested by the thyroid using a first SPECT sinogram corresponding to the raw data of the first SPECT image, a second SPECT sinogram corresponding to the raw data of the second SPECT image, the attenuation map, and the thyroid map.

2. The apparatus for enabling quantitative thyroid SPECT without CT using a neural network according to claim 1, wherein the first SPECT image is a primary emission SPECT image and the second SPECT image is a scattering SPECT image.

3. The apparatus for enabling quantitative thyroid SPECT without CT using a neural network, as described in claim 1, wherein the generation unit generates a third SPECT image by performing attenuation correction (AC) on the first SPECT sinogram using the attenuation map, scattering correction (SC) on the first SPECT sinogram using the second SPECT sinogram, and applying resolution recovery (RR) to the first SPECT sinogram.

4. The apparatus for enabling quantitative thyroid SPECT without CT using a neural network according to claim 3, wherein the third SPECT image is a quantitative ACSCRR SPECT image.

5. The device for enabling quantitative thyroid SPECT without CT using a neural network according to claim 3, wherein the calculation unit combines the third SPECT image and the thyroid map to calculate the amount of radiopharmaceutical ingested by the thyroid gland.

6. The device for enabling quantitative thyroid SPECT without CT using a neural network according to claim 5, wherein the calculation unit calculates the amount of radiopharmaceutical ingested by the thyroid gland by combining the third SPECT image and the thyroid map and counting matching voxels.

7. The second model is a device that enables quantitative thyroid SPECT without CT using a neural network according to claim 1, which is trained to segment thyroid maps in an input image based on a plurality of thyroid segmentation maps drawn and labeled along the outline of the thyroid gland in a CT (Computed Tomography) image.

8. One or more processors, and A method that enables quantitative thyroid SPECT without CT using a neural network having a memory that stores one or more programs executed by the one or more processors, The process involves inputting a first SPECT (Single Photon Emission Computed Tomography) image and a second SPECT image of the thyroid gland into a first model to generate an attenuation map, The steps include inputting the attenuation map and the first SPECT image into the second model to segment the thyroid map, A method for enabling quantitative thyroid SPECT without CT using a neural network, comprising the steps of calculating the amount of radiopharmaceutical intake by the thyroid using a first SPECT sinogram corresponding to the raw data of the first SPECT image, a second SPECT sinogram corresponding to the raw data of the second SPECT image, the attenuation map, and the thyroid map.

9. A method for enabling quantitative thyroid SPECT without CT using a neural network according to claim 8, wherein the first SPECT image is a primary emission SPECT image and the second SPECT image is a scattering SPECT image.

10. A method for enabling quantitative thyroid SPECT without CT using a neural network, as described in claim 8, comprising the steps of generating a third SPECT image by performing attenuation correction (AC) on the first SPECT sinogram using the attenuation map, performing scattering correction (SC) on the first SPECT sinogram using the second SPECT sinogram, and applying resolution recovery (RR) to the first SPECT sinogram.

11. The method for enabling quantitative thyroid SPECT without CT using a neural network according to claim 10, wherein the third SPECT image is a quantitative ACSCRR SPECT image.

12. The method for enabling quantitative thyroid SPECT without CT using a neural network according to claim 11, wherein the calculation step includes the step of combining the third SPECT image and the thyroid map to calculate the amount of radiopharmaceutical ingested by the thyroid gland.

13. The method for enabling quantitative thyroid SPECT without CT using a neural network according to claim 12, wherein the calculation step includes the step of calculating the amount of radiopharmaceutical intake by the thyroid by combining the third SPECT image and the thyroid map and counting matching voxels.

14. A method for enabling quantitative thyroid SPECT without CT using a neural network according to claim 8, wherein the second model is trained to segment a thyroid map in an input image based on a plurality of thyroid segmentation maps drawn and labeled along the outline of the thyroid gland in a CT (Computed Tomography) image.

Images