JP2022080094A - Nuclear medicine diagnosis device and nuclear medicine data processing device - Google Patents

Abstract

To determine a quantitative value related to a metabolic function based on collection in one bed.SOLUTION: A nuclear medicine diagnosis device according to an embodiment includes: an acquisition unit; a generation unit; and a calculation unit. The acquisition unit acquires the list mode data based on collection in one bed. The generation unit generates a plurality of nuclear medicine data different in the collection timing based on the list mode data. The calculation unit calculates a quantitative value related to a metabolic function based on the plurality of nuclear medicine data.SELECTED DRAWING: Figure 1

Description

The embodiments disclosed herein and in the drawings relate to a nuclear medicine diagnostic apparatus and a nuclear medicine data processing apparatus.

Nuclear medicine diagnostic equipment such as SPECT (Single Photon Emission Computed Tomography) equipment and PET (Positron Emission Tomography) equipment contains chemicals (blood flow markers, tracers) containing radioisotopes (hereinafter referred to as RI) in vivo. Utilize the property of being selectively taken up by specific tissues and organs. The nuclear medicine diagnostic device detects gamma rays emitted from RI distributed in the living body with a gamma ray detector arranged outside the living body, collects data based on the detected gamma rays, and uses this data. Generate nuclear medicine images.

For example, SUV (standardized uptake value) is one of the main quantitative indexes in PET examination using F18-FDG (fluorodeoxyglucose, hereinafter referred to as FDG). The SUV value is an index used for quantifying the distribution of RI (Radio Isotope) tracers such as FDG in the body, and it is considered that the accumulation of FDG is relatively strong if the SUV value is large. However, FDG accumulates not only in malignant tumors but also in active inflammatory sites and normal physiological sites. Therefore, it is difficult to clearly determine the threshold value for determining whether the tumor is benign or malignant.

As a method for identifying a malignant tumor, a method of obtaining a quantitative value regarding the metabolic function of a tissue can be considered by comparing each SUV of nuclear medicine data (raw data or nuclear medicine image) collected at different timings. .. As a method for acquiring nuclear medicine data collected at different timings, there are a method using delayed imaging and a method of performing dynamic imaging immediately after FDG administration.

However, in the method using delayed imaging and the method of performing dynamic imaging immediately after FDG administration, the examination time becomes long, so that the exposure dose of the subject increases, the burden on the subject is heavy, and the user's burden is heavy. The burden is also heavy.

Special Table 2020-517946

One of the challenges to be solved by the embodiments disclosed herein and in the drawings is to determine quantitative values for metabolic function based on collection in one bed. However, the problems to be solved by the embodiments disclosed in the present specification and the drawings are not limited to the above problems. The problem corresponding to each effect by each configuration shown in the embodiment described later can be positioned as another problem.

The nuclear medicine diagnostic apparatus according to the embodiment includes an acquisition unit, a generation unit, and a calculation unit. The acquisition unit acquires list mode data based on the collection in one bed. The generation unit generates a plurality of nuclear medicine data having different collection timings based on the list mode data. The calculation unit calculates quantitative values related to metabolic function based on a plurality of nuclear medicine data.

A block diagram showing a configuration example of a nuclear medicine diagnostic system including a nuclear medicine diagnostic device according to an embodiment. (A) is a diagram for explaining a method using conventional delayed imaging, and (b) is a diagram for explaining a method for performing dynamic imaging immediately after conventional FDG administration. The figure for demonstrating an example of the operation of the nuclear medicine data generation function which concerns on this embodiment. In (a), the time change of the SUV value of the malignant tumor and the normal tissue, and the collection timing of the first image and the second image generated from the first half and the second half of the list mode data based on the collection in one bed, respectively. An explanatory diagram showing an example of the relationship with the above, (b) shows the time variation of the SUV values of the malignant tumor and the normal tissue, and the list mode data based on the collection in one bed from the first minute and the last minute, respectively. Explanatory drawing which shows an example of the relationship with the collection timing of each of the generated 1st image and 2nd image. A flowchart showing an example of a procedure for obtaining a quantitative value related to metabolic function based on one-bed collection (one step-and-shoot imaging) by a processing circuit.

Hereinafter, embodiments of the nuclear medicine diagnostic apparatus and the nuclear medicine data processing apparatus will be described in detail with reference to the drawings.

The nuclear medicine diagnostic apparatus according to the present embodiment may be any as long as it can collect list mode data based on imaging by step-and-shoot, and a SPECT apparatus, a PET apparatus, or the like can be used. In the following description, an example of using a nuclear medicine diagnostic device capable of PET examination using FDG as the nuclear medicine diagnostic device according to the present embodiment will be shown.

FIG. 1 is a block diagram showing a configuration example of a nuclear medicine diagnostic system 1 including a nuclear medicine diagnostic device 10 according to an embodiment. The nuclear medicine diagnostic apparatus 10 may be any as long as it can collect list mode data based on imaging by step-and-shoot, and a SPECT apparatus, a PET apparatus, or the like can be used. In the following description, an example of using a nuclear medicine diagnostic device capable of PET examination using FDG as the nuclear medicine diagnostic device 10 according to the present embodiment will be shown.

The nuclear medicine diagnostic system 1 includes a nuclear medicine diagnostic device 10 and an image server 101.

The nuclear medicine diagnostic apparatus 10 includes a gamma ray detector 11, a data acquisition circuit 12, and a console 13 as an example of the nuclear medicine data processing apparatus.

The gamma ray detector 11 detects gamma rays emitted from the radioactive chemicals taken into the subject under the control of the console 13.

When a PET device is used as the nuclear medicine diagnostic device 10, the gamma ray detector 11 is a detector that detects gamma rays contained in a drug such as FDG and emitted from a radioactive drug administered to a subject. As the gamma ray detector 11, a semiconductor type detector may be used, or a scintillator type detector may be used.

When the gamma ray detector 11 is configured by using a semiconductor type detector, the gamma ray detector 11 is a collimeter, and a plurality of semiconductor elements for gamma ray detection arranged in two dimensions for detecting collimated gamma rays (hereinafter referred to as “semiconductor elements”). It has a semiconductor element) and an electronic circuit for semiconductors. The semiconductor element is composed of, for example, CdTe or CdZnTe (CZT).

The semiconductor electronic circuit generates position information and intensity information based on the output of the semiconductor element every time an event in which gamma rays are incident occurs, and outputs the position information and the intensity information to the data acquisition circuit 12. This position information is information indicating which semiconductor element among the plurality of semiconductor elements is incident.

On the other hand, when the gamma ray detector 11 is configured by using a scintillator type detector, the gamma ray detector 11 is a collimeter for defining the incident angle of the gamma ray, and a scintillator that emits a momentary flash when the collimated gamma ray is incident. , A light guide, a plurality of photomultiplier tubes arranged in two dimensions for detecting the light emitted from the scintillator, an electronic circuit for the scintillator, and the like. The scintillator is composed of, for example, thallium-activated sodium iodide NaI (Tl).

In the electronic circuit for scintillator, each time an event in which gamma rays are incident occurs, information on the incident position of gamma rays in a detection surface composed of a plurality of photomultiplier tubes based on the outputs of the plurality of photomultiplier tubes. (Position information) and intensity information are generated and output to the data acquisition circuit 12. This position information may be information on two-dimensional coordinates in the detection surface, or the detection surface is virtually divided into a plurality of division areas (hereinafter referred to as primary cells) in advance, and which primary cell is used. It may be information indicating whether or not there was an incident on.

When a SPECT device is used as the nuclear medicine diagnostic device 10, the gamma ray detector 11 emits gamma rays emitted from radioisotopes such as technetium-99m and thallium-201, which are contained in a drug and administered to the subject. It is a detector to detect. In this case as well, the semiconductor type detector may be used or the scintillator type detector may be used as the gamma ray detector 11, and the configuration of the semiconductor type detector and the scintillator type detector is as the nuclear medicine diagnostic apparatus 10. This is the same as when a PET device is used.

That is, the gamma ray detector 11 detects the gamma rays emitted from the radioactive chemicals controlled by the console 13 and administered to the subject, and outputs the position information and the intensity information for each event. Hereinafter, an example in which a PET device is used as the nuclear medicine diagnostic device 10 and a PET examination using FDG is performed will be described. For example, if one imaging (one step-and-shoot imaging) is performed while the positional relationship with the subject is fixed, the gamma-ray detector 11 will perform the one-step imaging of the subject. Detects gamma rays emitted from.

The nuclear medicine diagnostic apparatus 10 according to the present embodiment can collect data on a pair annihilation event of a subject at each of a plurality of imaging sites in a PET examination. Specifically, the nuclear medicine diagnostic apparatus 10 images a subject at each of a plurality of imaging sites by a step-and-shoot method. In the step-and-shoot method, the nuclear medicine diagnostic apparatus 10 collects data at one imaging site, and then moves the top plate to move the position of the subject placed on the top plate, and then moves the position of the subject placed on the top plate to the next imaging site. Take an image with.

Here, the setting of a plurality of imaging sites is performed, for example, by a user who refers to a scanogram of a subject imaged by an X-ray CT (Computed Tomography) device and sets an imaging plan for a nuclear medicine image. The scanogram moves the subject along the body axis while irradiating X-rays from the X-ray tube with the rotating frame that rotatably supports the X-ray tube and the X-ray detector fixed. It is image data obtained by scanning the whole body of a subject along the body axis direction.

For example, when the height of the subject is "175 cm" and the width of the gamma ray detector 11 (width along the longitudinal direction of the top plate) is "25 cm" and a whole-body nuclear medicine image is taken. The user sets the whole body image while overlapping the subject by "12.5 cm" every "25 cm". Each imaging is called a bed position, or bed.

In the present embodiment, data collection by one step-and-shoot imaging is defined as one-bed collection. Collection in one bed corresponds to imaging of the area of the subject corresponding to the width of the gamma ray detector 11. For example, when imaging (multi-bed scanning) is performed at a plurality of imaging positions, data is collected per bed for, for example, 5 minutes.

The data acquisition circuit 12 is composed of, for example, a printed circuit board, collects the output of the gamma ray detector 11 in, for example, a list mode, and outputs list mode data based on the collection in one bed. In the list mode, the gamma ray detection position information, intensity information, information indicating the relative position between the gamma ray detector 11 and the subject (position and angle of the gamma ray detector 11, etc.), and the gamma ray detection time are gamma ray incident events. Collected for each.

The console 13 as an example of the nuclear medicine data processing device may be connected to the data acquisition circuit 12 so as to be able to transmit and receive data, and may not be provided in the same room or building as the data acquisition circuit 12.

Further, the nuclear medicine data processing apparatus having the same configuration and operation as the console 13 described below is connected to the nuclear medicine diagnostic apparatus 10 having at least the gamma ray detector 11 and the data acquisition circuit 12 via the network 100. List mode data based on the collection in one bed may be acquired from the data collection circuit 12. Further, the nuclear medicine data processing device is connected to the image server 101 via the network 100, and the list mode data based on the collection in one bed output from the data collection circuit 12 and stored in the image server 101 is imaged. It may be acquired from the server 101.

The console 13 has an input interface 21, a display 22, a storage circuit 23, a network connection circuit 24, and a processing circuit 25.

The input interface 21 is composed of general input devices such as a trackball, a switch button, a mouse, a keyboard, a ten-key, and a voice input device, and outputs an operation input signal corresponding to the user's operation to the processing circuit 25. The display 22 is composed of a general display output device such as a liquid crystal display or an OLED (Organic Light Emitting Diode) display.

The storage circuit 23 has a configuration including a recording medium readable by a processor such as a RAM (Random Access Memory), a semiconductor memory element such as a flash memory, a hard disk, an optical disk, and the like, and is a program used by the processing circuit 25. And stores data such as list mode data based on the collection of one bed. A part or all of the program and data in the recording medium of the storage circuit 23 may be downloaded by communication via the network 100, or may be given to the storage circuit 23 via a portable storage medium such as an optical disk. You may.

The network connection circuit 24 implements various information and communication protocols according to the form of the network 100. The network connection circuit 24 connects to other electric devices via the network 100 according to the various protocols. The network 100 means a general information communication network using telecommunications technology, and includes a wireless / wired LAN such as a hospital backbone LAN (Local Area Network) and an Internet network, as well as a telephone communication line network, an optical fiber communication network, and cable communication. Includes networks and satellite communications networks, etc.

The nuclear medicine diagnostic apparatus 10 is connected to each other via the image server 101 and the network 100 so as to be able to transmit and receive data.

Although FIG. 1 shows an example in which the nuclear medicine diagnostic apparatus 10 and the image server 101 are connected via the network 100, the nuclear medicine diagnostic device 10 and the image server 101 are directly connected by wireless communication such as wired communication or infrared communication without going through the network 100. May be connected.

The processing circuit 25 reads out and executes a medical information processing program stored in the storage circuit 23 to obtain a quantitative value related to the metabolic function based on one-bed collection (one step-and-shoot imaging). It is a processor that executes the processing of.

As shown in FIG. 1, the processor of the processing circuit 25 realizes an acquisition function 31, a nuclear medicine data generation function 32, a denoise function 33, and a quantitative value calculation function 34. Each of these functions is stored in the storage circuit 23 in the form of a program.

The acquisition function 31 acquires list mode data based on the collection in one bed from the data acquisition circuit 12. The acquisition function 31 may acquire list mode data that has been collected in advance and stored in the image server 101 via the network 100. The acquisition function 31 is an example of an acquisition unit.

Here, two conventional methods for acquiring nuclear medicine data collected at different timings will be described.

FIG. 2A is a diagram for explaining a method using conventional delayed imaging, and FIG. 2B is a diagram for explaining a method for performing dynamic imaging immediately after administration of conventional FDG.

In a general imaging procedure, after FDG is administered to a subject, after waiting for about 60 minutes, imaging is performed on the observation area of the subject. For this reason, in a general imaging procedure, only one nuclear medicine data acquisition is performed for one observation area. In addition, this general imaging procedure called "imaging 60 minutes after FDG administration" is adopted in consideration of throughput and physical half-life, and in the first place, the accumulation ratio of normal tissue and malignant tumor is the most appropriate, and both It cannot be said that it is the optimum time for identification of.

As a method for acquiring nuclear medicine data collected at different timings, as described above, a method using delayed imaging (see FIG. 2A) and a method of performing dynamic imaging immediately after FDG administration (FIG. 2 (Fig. 2)). b) See) is known.

However, in the method using delayed imaging, after a general imaging procedure, that is, after waiting for about 60 minutes after FDG administration to the subject, imaging of the observation area of the subject is performed, and then a predetermined image is further determined. By performing another imaging after the waiting time (for example, 20 to 30 minutes), nuclear medicine data is collected twice for one observation area at different timings (see FIG. 2A). Therefore, in this method, it is necessary to perform a plurality of imaging shots, which is very complicated and the examination time becomes long, so that the exposure dose of the subject increases. In addition, since the inspection time is long, the inspection frame is squeezed, and the burden on the subject is heavy and the burden on the user is also heavy.

In the method of performing dynamic imaging immediately after FDG administration, dynamic imaging is started immediately after FDG administration, and the top plate is continuously moved during the imaging to make one round trip in the observation area of the subject placed on the top plate. By doing so, nuclear medicine data is collected twice at different timings for one observation area (see FIG. 2 (b)). Therefore, this method requires a long-time dynamic imaging of about 90 minutes. Therefore, even with this method, the inspection time becomes long, the exposure dose of the subject increases, the frame of the inspection is squeezed, and the burden on the subject and the user is heavy.

Therefore, the nuclear medicine data generation function 32 of the processing circuit 25 according to the present embodiment generates a plurality of nuclear medicine data having different collection timings based on the list mode data acquired by the acquisition function 31 in one bed. do. The nuclear medicine data is the list mode data itself or a nuclear medicine image generated by the reconstruction process based on the list mode. The nuclear medicine data generation function 32 is an example of a generation unit.

FIG. 3 is a diagram for explaining an example of the operation of the nuclear medicine data generation function 32 according to the present embodiment. As shown in FIG. 3, the nuclear medicine diagnostic apparatus 10 according to the present embodiment can generate a plurality of nuclear medicine data having different collection timings based on the list mode data based on the collection in one bed.

For example, the nuclear medicine data generation function 32 extracts a plurality of list mode data belonging to different collection periods from the list mode data based on the collection in one bed acquired by the acquisition function 31, thereby performing a plurality of nuclear medicine. Generate data. In this case, the nuclear medicine data is the list mode data itself (raw data). Further, the nuclear medicine data generation function 32 may generate a plurality of nuclear medicine images by performing a reconstruction process based on each of the extracted list mode data belonging to different collection periods. In this case, the nuclear medicine data is a nuclear medicine image. It is recommended that collection periods different from each other do not overlap with each other.

As described above, the nuclear medicine data generation function 32 can generate a plurality of nuclear medicine data having different collection timings from the list mode data based on the collection in one bed.

FIG. 4A shows the time variation of the SUV values of the malignant tumor and the normal tissue, and the first image and the second image generated from the first half and the second half of the list mode data based on the collection in one bed, respectively. It is an explanatory diagram showing an example of the relationship with the collection timing of, and (b) shows the time change of the SUV value of each of the malignant tumor and the normal tissue, and the first minute of the list mode data based on the collection in one bed. It is explanatory drawing which shows an example of the relationship with the collection timing of each of the 1st image and the 2nd image generated from the last 1 minute respectively.

Since malignant tumors are metabolized, they consume a lot of energy (glucose), and FDG gradually accumulates in malignant tumors. Therefore, the SUV value of the pixel corresponding to the malignant tumor increases with time. On the other hand, in normal tissues, FDG gradually disappears, or FDG accumulation is slower than that of malignant tumors. Therefore, the time change of SUV value is much smaller in normal tissue than in malignant tumor.

Therefore, even in the collection time of one bed (for example, 5 minutes), the images reconstructed based on the different collection periods of 5 minutes have different contrasts.

That is, it is possible to calculate a quantitative value related to metabolic function by comparing the nuclear medicine data (list mode data itself or the nuclear medicine image) corresponding to each of the different collection periods in the collection time of one bed. be.

However, when a plurality of nuclear medicine data having different collection timings are generated from the list mode data based on the collection in one bed, the count of each nuclear medicine data may be insufficient and the statistical noise may increase.

Therefore, the denoise function 33 performs denoise processing on each of a plurality of nuclear medicine data having different collection timings. By performing the denoising process, it is possible to clarify the difference between the plurality of nuclear medicine data even if the plurality of nuclear medicine data having a relatively short collection time generated based on the collection in one bed. The denoise function 33 is an example of a denoise unit.

Various types of denoise processing have been conventionally known, and any of these can be used. Preferably, the denoise function 33 performs the denoise processing described in, for example, Japanese Patent Application Laid-Open No. 2019-21147. In this case, the denoise function 33 generates a plurality of nuclear medicine data having different collection timings generated by the nuclear medicine data generation function 32 with respect to the trained model that outputs the denoised nuclear medicine data by inputting the nuclear medicine data. It is advisable to output a plurality of denoised nuclear medicine data by inputting each of the above.

At this time, the trained model has a loss function so as to minimize the difference between the nuclear medicine data of the first noise level and the nuclear medicine data of a plurality of second noise levels higher than the first noise level. It is built by optimizing. According to the denoise processing using this trained model, the ambient noise can be removed while maintaining the accumulation of the accumulation sites, so that the contrast is maintained and the quantification of the nuclear medicine data is guaranteed.

In constructing this trained model, specifically, the methods of deep learning (DL, Deep Learning) network and convolutional neural network (CNN, Convolutional Neural Network) are applied, and training is performed using the loss function. Repeat the parameters of the DL-CNN network (eg, convolutional and pooling layer weights and biases) until the outage criteria are met (eg, convergence of the parameters to a given threshold) to generate the DL-CNN network. adjust.

That is, learning is performed on a plurality of first noise level nuclear medicine data by simultaneously minimizing the loss function, and learning is performed while repeatedly adjusting the adjustable parameters. The loss function combines high quality data as the first noise level nuclear medical data with multiple noise level nuclei such that the current version of the DL-CNN network has a higher noise level than the first noise level nuclear medical data. Low quality data as medical data applies)).

In the case of PET imaging, the high quality data and the low quality data are nuclear medicine data with high / good image quality and low / poor image quality, respectively. The DL-CNN network becomes robust against changes in noise level by being trained using different samples of low quality data with different noise levels. In general, the signal-to-noise ratio (SNR) becomes smaller enough to utilize smaller data sets (eg, due to shorter scan times or other factors that result in less concurrency counts) when reconstructing the image. ..

When using reconstructed images as nuclear medicine data, high quality images can be generated using all of the simultaneous counts from the PET scan of the first subject, for example, the highest image quality achievable. PET image is generated. Next, lower quality images reconstructed from the scan of the first subject, 155 (1,1), 155 (1,2), ..., 155 (1, k) are selected from the entire dataset. It can be generated using various subsets of the simultaneous count data.

Each of these low quality images corresponds to a different number of counts, thus making it possible to reconstruct an image with a range of noise levels. Similarly, low quality images 155 (2,1) from a subset of the entire dataset generated from PET scans of the second subject and all other subjects up to the last subject. ), 155 (2,2), ..., 155 (2, k) (ie, for the Lth subject, the low quality image is 155 (L, 1), 155 (L, 2). ), ..., 155 (L, k), and the high quality image is 153 (L)).

The quantitative value calculation function 34 calculates a quantitative value related to the metabolic function based on a plurality of nuclear medicine data. For example, the quantitative value calculation function 34 obtains an SUV (Standardized Uptake Value) value for each predetermined subregion (for example, for each pixel) for each of a plurality of nuclear medicine data, and each SUV value of the plurality of nuclear medicine data. (See FIGS. 4A and 4B), a quantitative value relating to the metabolic function is calculated based on the change in the SUV value obtained by comparing the above. Changes in SUV values are represented, for example, by differences or ratios of SUV values. Further, the quantitative value calculation function 34 may calculate a quantitative value relating to the metabolic function based on a plurality of nuclear medicine data denoised by the denoise function 33. The quantitative value calculation function 34 is an example of a calculation unit.

FIG. 5 is a flowchart showing an example of a procedure for obtaining a quantitative value related to metabolic function based on one-bed collection (one step-and-shoot imaging) by the processing circuit 25. In FIG. 5, reference numerals with numbers attached to S indicate each step in the flowchart.

First, FDG is administered to the subject (step S1), and the subject is allowed to stand by at rest for a predetermined time (for example, 60 minutes).

Next, in step S3, imaging by step and shoot is performed once, and the acquisition function 31 acquires list mode data based on the collection in one bed from the data acquisition circuit 12.

Next, in step S4, the nuclear medicine data generation function 32 generates a plurality of nuclear medicine data having different collection timings based on the list mode data acquired by the acquisition function 31 in one bed (FIG. 4). (A), see FIG. 4 (b)).

Next, in step S5, denoise processing is performed on each of the plurality of nuclear medicine data having different collection timings.

Then, in step S6, the quantitative value calculation function 34 obtains an SUV (Standardized Uptake Value) value for each of the plurality of nuclear medicine data, for example, for each pixel, and compares the SUV values of the plurality of nuclear medicine data. A quantitative value relating to the metabolic function is calculated based on the change in the SUV value obtained in.

By the above procedure, the quantitative value regarding the metabolic function can be obtained based on the collection in one bed (one step-and-shoot imaging).

According to the processing circuit 25 according to the present embodiment, it is possible to generate a plurality of nuclear medicine data having different collection timings based on the collection in one bed. Therefore, it is possible to obtain a quantitative value regarding the metabolic function based on a plurality of nuclear medicine data having different collection timings only by collecting in one bed. Therefore, it is much simpler than the conventional method using delayed imaging (see FIG. 2 (a)) and the method of performing dynamic imaging immediately after FDG administration (see FIG. 2 (b)), and significantly reduces the examination time. Can be shortened. Therefore, according to the processing circuit 25 according to the present embodiment, the exposure dose of the subject can be reduced, and the burden on the subject and the burden on the user can be greatly reduced.

In the above embodiment, an example of generating two nuclear medicine data from the list mode data based on the collection in one bed is shown, but three or more nuclear medicine data having different collection timings may be generated. In this case, the quantitative value calculation function 34 may calculate the quantitative value related to the metabolic function based on the change in the SUV value of the nuclear medicine data of 3 or more.

According to at least one embodiment described above, quantitative values relating to metabolic function can be obtained based on collection in one bed.

In the above embodiment, the term "processor" refers to, for example, a dedicated or general-purpose CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an integrated circuit for a specific application (Application Specific Integrated Circuit: ASIC). Circuits such as programmable logic devices (eg, Simple Programmable Logic Device (SPLD), Complex Programmable Logic Device (CPLD), and Field Programmable Gate Array (FPGA)) means. When the processor is, for example, a CPU, the processor realizes various functions by reading and executing a program stored in a storage circuit. Further, when the processor is, for example, an ASIC, instead of storing the program in the storage circuit, the function corresponding to the program is directly incorporated as a logic circuit in the circuit of the processor. In this case, the processor realizes various functions by hardware processing that reads and executes a program embedded in the circuit. Alternatively, the processor can also combine software processing and hardware processing to realize various functions.

Further, in the above embodiment, an example in which a single processor of the processing circuit realizes each function has been shown, but a processing circuit is configured by combining a plurality of independent processors, and each processor realizes each function. May be good. When a plurality of processors are provided, the storage circuit for storing the program may be provided individually for each processor, or one storage circuit collectively stores the programs corresponding to the functions of all the processors. May be good.

Although some embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other embodiments, and various omissions, replacements, changes, and combinations of embodiments can be made without departing from the gist of the invention. These embodiments and variations thereof are included in the scope of the invention described in the claims and the equivalent scope thereof, as are included in the scope and gist of the invention.

10 Nuclear medicine diagnostic device 11 Gamma ray detector 12 Data acquisition circuit 13 Console (nuclear medicine data processing device)
25 Processing circuit 31 Acquisition function 32 Nuclear medicine data generation function 33 Denoise function 34 Quantitative value calculation function

Claims (8)

An acquisition unit that acquires list mode data based on collection in one bed,
A generator that generates multiple nuclear medicine data with different collection timings based on the list mode data,
A calculation unit that calculates quantitative values related to metabolic function based on the plurality of nuclear medicine data,
Equipped with nuclear medicine diagnostic equipment. The generator is
The plurality of nuclear medicine data are generated by extracting a plurality of list mode data belonging to different collection periods from the list mode data acquired by the acquisition unit, or the plurality of data belonging to the different collection periods. By performing the reconstruction process based on each of the list mode data, a plurality of nuclear medicine images are generated as the plurality of nuclear medicine data.
The nuclear medicine diagnostic apparatus according to claim 1. The collection periods that differ from each other do not overlap with each other.
The nuclear medicine diagnostic apparatus according to claim 2. The calculation unit
The metabolic function is related to the change in the SUV value obtained by obtaining the SUV (Standardized Uptake Value) value of each of the plurality of nuclear medicine data and comparing the SUV values of the plurality of nuclear medicine data. Calculate a quantitative value,
The nuclear medicine diagnostic apparatus according to any one of claims 1 to 3. A denoise unit that performs denoise processing on each of the plurality of nuclear medicine data with different collection timings.
Further prepare
The calculation unit
Quantitative values for the metabolic function are calculated based on the denoised nuclear medicine data.
The nuclear medicine diagnostic apparatus according to any one of claims 1 to 4. The denoise section is
For a trained model that outputs nuclear medicine data denoised by inputting nuclear medicine data, the denoise is performed by inputting each of the plurality of nuclear medicine data generated by the generation unit at different collection timings. The plurality of nuclear medicine data are output to the calculation unit, and the data is output to the calculation unit.
The trained model is
It is constructed by optimizing the loss function to minimize the difference between the nuclear medicine data of the first noise level and the nuclear medicine data of a plurality of second noise levels higher than the first noise level. Ru,
The nuclear medicine diagnostic apparatus according to claim 5. A gamma ray detector that detects gamma rays emitted from a radioactive chemical administered to the subject while fixing the positional relationship with the subject in the one bed.
A data acquisition circuit that collects the output of the gamma ray detector in the list mode and gives the list mode data based on the collection in the one bed to the acquisition unit.
The nuclear medicine diagnostic apparatus according to any one of claims 1 to 6, further comprising. An acquisition unit that acquires list mode data based on collection in one bed,
A generator that generates multiple nuclear medicine data with different collection timings based on the list mode data,
A calculation unit that calculates quantitative values related to metabolic function based on the plurality of nuclear medicine data,
Equipped with nuclear medicine data processing equipment.

Images