Multi-nuclide data processing method and device, storage medium and digital pet system

Abstract

The application discloses a kind of multinuclei data processing method, device, storage medium and digital PET system.The multinuclei data processing method includes: obtaining nuclide category corresponding to coincidence event according to prior information;According to the coincidence event and the nuclide category, the sum of all nuclide decay activity is obtained by multi-index fitting;According to the sum of all nuclide decay activity and the sum of all nuclide initial activity, decay correction factor is calculated, and the decay correction factor is used to decay correction for each coincidence event when image reconstruction is carried out.The application can solve the quantization error problem of traditional single-index model in multinuclei scene, process proton therapy or heavy ion therapy or multiple tracer PET and other complex scenes of multiple nuclides coexistence.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a multi-nuclide data processing method and device, a computer storage medium and a digital PET system. BACKGROUND

[0002] During proton or heavy ion therapy, the interaction of incident radiation with tissue can generate a variety of positron emitting nuclides, such as 15 O, 11 C, 13 N, 10 C, etc., for treatment verification and dose monitoring. Currently, most proton therapy centers use PET imaging after treatment to capture nuclide signals with a half-life of several minutes (about 2-20 minutes), which requires a long acquisition time and is accompanied by biological clearance effects, resulting in delayed quantitative information. On the contrary, the "beam on PET" technique attempts to acquire decay signals of nuclides with extremely short half-lives (<19 seconds) in real time, which can provide timely dose feedback during treatment. However, whether it is post-treatment or real-time imaging, the PET image often contains decay signals of multiple positron nuclides at the same time, the 511 keV photon pairs produced by them are superimposed on each other, and the decay constants of different nuclides differ greatly. Traditional PET imaging usually uses only one tracer at a time and uses a single exponential decay model for correction, which cannot correctly process mixed signals of multiple nuclides. When multiple nuclides coexist, simple single exponential decay correction will introduce quantitative errors. Existing research shows that multi-tracer PET imaging has great potential, but also faces serious challenges. Traditional imaging systems are limited to a single nuclide, and repeated scanning is costly. At present, there is still a lack of effective methods that can simultaneously achieve multi-nuclide separation and decay correction at the level of list mode data. Therefore, there is an urgent need for new multi-nuclide imaging and signal separation techniques, especially new multi-nuclide decay correction schemes for actual imaging systems and list mode data. SUMMARY

[0003] Therefore, it is necessary to provide a multi-nuclide data processing method, device, computer storage medium and digital PET system to solve at least one technical problem existing in the traditional scheme.

[0004] According to a first aspect of the present application, a multi-nuclide data processing method is provided, comprising: obtaining a nuclide category corresponding to a coincidence event according to prior information; performing multi-exponential fitting according to the coincidence event and the nuclide category to obtain a sum of decay activities of all nuclides; and calculating a decay correction factor according to the sum of decay activities of all nuclides and a sum of initial activities of all nuclides, the decay correction factor being used for decay correction of each coincidence event during image reconstruction.

[0005] According to one embodiment of the present application, the multi-nuclide data processing method further comprises: obtaining to-be-processed coincidence events, and correcting the to-be-processed coincidence events to obtain coincidence events.

[0006] According to one embodiment of the present application, the correcting the to-be-processed coincidence events comprises: performing random correction, background correction, or both random correction and background correction on the to-be-processed coincidence events.

[0007] According to one embodiment of the present application, the obtaining to-be-processed coincidence events, and performing random correction, background correction, or both random correction and background correction on the to-be-processed coincidence events to obtain coincidence events comprises: obtaining to-be-processed coincidence events and processing to form a first list set, determining a start time of beam-off, delaying the to-be-processed coincidence events and processing to form a second list set; and performing random correction, background correction, or both random correction and background correction on the to-be-processed coincidence events according to the start time of beam-off, the first list set and the second list set to obtain coincidence events.

[0008] According to one embodiment of the present application, the obtaining to-be-processed coincidence events and processing to form a first list set, determining a start time of beam-off, delaying the to-be-processed coincidence events and processing to form a second list set comprises: obtaining a first average time of each to-be-processed coincidence event, taking the first average time as a first reference time of the corresponding to-be-processed coincidence event, and dividing all the to-be-processed coincidence events into a plurality of first bins at equal intervals according to the first reference time; obtaining a second average time of each coincidence event, taking the second average time as a second reference time of the corresponding to-be-processed coincidence event, and dividing all the to-be-processed coincidence events into a plurality of second bins at equal intervals after arranging according to the second reference time, the number of the first bins being the same as the number of the second bins, and corresponding one by one.

[0009] According to one embodiment of the present application, the obtaining to-be-processed coincidence events and processing to form a first list set, determining a start time of beam-off comprises: counting the number of single events in each first bin, selecting a first bin whose number of single events exceeds a predetermined threshold, and regarding continuous first bins as a same interval, and taking the end time of the last interval or the first bin as the start time of beam-off.

[0010] According to one embodiment of the present application, the random correction, the background correction or the random correction and the background correction of the to-be-processed coincidence events according to the start time of the off-beam, the first list set and the second list set to obtain coincidence events comprises: subtracting the number of single events in the corresponding second bin from the number of single events in the first bin to realize random correction; obtaining the minimum value of the number of single events in all first bins as a background level, and subtracting the background level from the number of single events in the first bin to realize background correction.

[0011] According to one embodiment of the present application, the random correction, the background correction or the random correction and the background correction of the to-be-processed coincidence events according to the start time of the off-beam, the first list set and the second list set to obtain coincidence events comprises: eliminating coincidence events before the start time of the off-beam to perform multi-exponential fitting on the remaining coincidence events.

[0012] According to one embodiment of the present application, the multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities comprises: taking the coincidence events as input, performing multi-exponential fitting by using a non-linear least square method, and estimating an optimal square approximation to obtain the sum of all nuclide decay activities.

[0013] According to one embodiment of the present application, the multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities comprises: setting the sum of all nuclide decay activities of all categories to represent total radioactivity in an image, and adopting a fixed half-life fitting or a variable half-life fitting.

[0014] According to one embodiment of the present application, the multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities comprises:

[0015] The model function of the multi-exponential fitting is:

[0016] ,

[0017] wherein, is a parameter vector, , is an initial activity coefficient of the mth nuclide at a reference time, and M is the total number of nuclides, is a decay constant of the mth nuclide, is a time span of the coincidence events after background correction.

[0018] According to one embodiment of the present application, the multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities comprises:

[0019] The fixed half-life and least square method are used for fitting, and the objective function is:

[0020] ,

[0021] wherein χ2 represents the residual sum of squares between the observed value and the model predicted value, represents the number of bins of coincidence events, represents the number of single events in the jth interval, represents the time span of coincidence events in the jth interval after background correction.

[0022] According to one embodiment of the present application, the sum of all nuclide decay activities is obtained by multi-exponential fitting according to the coincidence events and the nuclide categories, including:

[0023] The variable half-life and least square method are used for fitting, and the model function is:

[0024] ,

[0025] The objective function is:

[0026] ,

[0027] wherein, is a parameter vector, , represents the number of bins of coincidence events, represents the number of single events in the jth interval, is the initial activity coefficient of the mth nuclide at the reference time, and M is the total number of nuclides, is the decay constant of the mth nuclide, is the time span of coincidence events after background correction, represents the time span of coincidence events in the jth interval after background correction.

[0028] According to one embodiment of the present application, the decay constant is limited in the range of the maximum and minimum values of the theoretical value.

[0029] According to one embodiment of the present application, a decay correction factor is calculated according to the sum of all nuclide decay activities and the sum of all nuclide initial activities, and the decay correction factor is used for decay correction of each coincidence event during image reconstruction, including: the decay correction factor is the ratio of the sum of all nuclide initial activities to the sum of all nuclide decay activities, and the weight of each coincidence event is decay corrected by using the decay correction factor during image reconstruction.

[0030] According to one embodiment of the present application, the formula of the decay correction factor is:

[0031] ,

[0032] wherein, denotes the decay correction factor at time t, is the initial activity coefficient of the mth nuclide at the reference time, and M is the total number of nuclides, is the decay constant of the mth nuclide.

[0033] According to one embodiment of the present application, in the image reconstruction, the decay correction factor is used to correct the weight of each coincidence event, comprising: using the following formula to correct the weight of each coincidence event:

[0034] ,

[0035] wherein, is the weight of coincidence event i before correction, is the weight after correction, denotes the decay correction factor of coincidence event i at time t.

[0036] According to a second aspect of the present application, a multi-nuclide data processing device is provided, comprising: a nuclide category obtaining module configured to obtain the nuclide category corresponding to a coincidence event according to prior information; a fitting module configured to perform multi-exponential fitting according to the coincidence event and the nuclide category to obtain the sum of decay activities of all nuclides; and a second correction module configured to calculate a decay correction factor according to the sum of decay activities of all nuclides and the sum of initial activities of all nuclides, the decay correction factor being used to correct each coincidence event in the image reconstruction.

[0037] According to one embodiment of the present application, the multi-nuclide data processing device further comprises: a first correction module configured to obtain a to-be-processed coincidence event, and correct the to-be-processed coincidence event to obtain a coincidence event.

[0038] According to one embodiment of the present application, the first correction module is configured to obtain a to-be-processed coincidence event, and correct the to-be-processed coincidence event by random correction, background correction, or both random correction and background correction to obtain a coincidence event.

[0039] According to one embodiment of the present application, the first correction module is configured to: obtain a to-be-processed coincidence event and process to form a first list set, determine the start time of beam-off, delay the to-be-processed coincidence event and process to form a second list set; and correct the to-be-processed coincidence event by random correction, background correction, or both random correction and background correction according to the start time of beam-off, the first list set and the second list set to obtain a coincidence event.

[0040] According to one embodiment of the present application, the first correction module is configured to: obtain a first average time of each of the to-be-processed coincidence events, take the first average time as a first reference time of the corresponding to-be-processed coincidence event, divide all the to-be-processed coincidence events into a plurality of first bins according to the first reference time at equal intervals; obtain a second average time of each of the coincidence events, take the second average time as a second reference time of the corresponding to-be-processed coincidence event, and divide all the to-be-processed coincidence events into a plurality of second bins according to the second reference time at equal intervals, the number of the first bins is the same as the number of the second bins, and the first bins and the second bins correspond to each other one by one.

[0041] According to one embodiment of the present application, the first correction module is configured to: count the number of single events in each first bin, select a first bin whose number of single events exceeds a predetermined threshold, and regard the continuous first bins as one interval, and take the end time of the last interval or the first bin as the start time of the off-beam.

[0042] According to one embodiment of the present application, the first correction module is configured to: subtract the number of single events in the corresponding second bin from the number of single events in the first bin to achieve random correction, and obtain the minimum value of the number of single events in all the first bins as a background level, and subtract the background level from the number of single events in the first bin to achieve background correction.

[0043] According to one embodiment of the present application, the first correction module is configured to: eliminate the coincidence events before the start time of the off-beam, and perform multi-exponential fitting on the remaining coincidence events.

[0044] According to one embodiment of the present application, the fitting module is configured to: take the coincidence events as input, perform multi-exponential fitting by using a non-linear least square method, and estimate the sum of the decay activities of all the nuclides by using the best square approximation.

[0045] According to one embodiment of the present application, the fitting module is configured to: set the sum of the decay activities of all the nuclides as the total radioactivity in the image, and perform fitting by using a fixed half-life or by using a variable half-life.

[0046] According to one embodiment of the present application, the model function of the multi-exponential fitting is:

[0047] ,

[0048] wherein, is a parameter vector, , is an initial activity coefficient of the mth nuclide at a reference time, and M is the total number of the nuclides, is a decay constant of the mth nuclide, The time span of the coincidence events after background correction.

[0049] According to one embodiment of the present application, the fitting module is configured to perform fitting using a fixed half-life and least square method, and a target function is:

[0050] ,

[0051] wherein χ2 represents a residual sum of squares between the observation values and the model prediction values, represents the number of bins of coincidence events, represents the number of single events in the jth interval, represents the time span of the coincidence events in the jth interval after background correction.

[0052] According to one embodiment of the present application, the fitting module is configured to perform fitting using a variable half-life and least square method, and a model function is:

[0053] ,

[0054] a target function is:

[0055] ,

[0056] wherein, is a parameter vector, , represents the number of bins of coincidence events, represents the number of single events in the jth interval, is an initial activity coefficient of the mth nuclide at a reference time, and M is the total number of nuclides, is a decay constant of the mth nuclide, is a time span of the coincidence events after background correction, represents the time span of the coincidence events in the jth interval after background correction.

[0057] According to one embodiment of the present application, the decay constant is limited in a range of a maximum value and a minimum value of a theoretical value.

[0058] According to one embodiment of the present application, the decay correction factor is a ratio of a sum of initial activities of all nuclides to a sum of decay activities of all nuclides, and the second correction module is configured to perform decay correction on a weight of each coincidence event using the decay correction factor when reconstructing an image.

[0059] According to one embodiment of the present application, a formula of the decay correction factor is:

[0060] ,

[0061] wherein, denotes the decay correction factor at time t, is the initial activity coefficient of the mth nuclide at the reference time, is the decay constant of the mth nuclide.

[0062] According to one embodiment of the present application, in image reconstruction, the decay correction factor is used to correct the weight of each coincidence event, including: using the following formula to correct the weight of each coincidence event:

[0063] ,

[0064] wherein, is the weight of coincidence event i before correction, is the weight after correction, denotes the decay correction factor of coincidence event i at time t.

[0065] According to a third aspect of the present application, a computer storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of the multi-nuclide data processing method as described above.

[0066] According to a fourth aspect of the present application, a computer program product is provided, which includes a computer program or instructions, and the computer program or instructions are executed by a processor to implement the steps of the multi-nuclide data processing method as described above.

[0067] According to a fifth aspect of the present application, a digital PET system is provided, which uses the multi-nuclide data processing device as described above to perform decay correction in image reconstruction.

[0068] The multi-nuclide data processing method, device, computer storage medium and digital PET system provided by the present application can separate and correct the mixed decay signals of multiple nuclides through multi-exponential fitting, separate the contributions of different nuclides, and solve the quantization error problem of the traditional single exponential model in the multi-nuclide scene, and are suitable for complex scenes where multiple nuclides coexist, such as proton therapy or heavy ion therapy or multiple tracer PET. BRIEF DESCRIPTION OF DRAWINGS

[0069] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0070] Figure 1A flowchart of a multi-nuclide data processing method according to an embodiment of the present application;

[0071] Figure 2 A diagram of an event according to an embodiment of the present application;

[0072] Figure 3 A diagram of a random coincidence according to an embodiment of the present application;

[0073] Figure 4 A histogram of a partial first bin according to an embodiment of the present application;

[0074] Figure 5 A histogram of a full first bin according to an embodiment of the present application;

[0075] Figure 6 A diagram of a multi-nuclide data processing device according to an embodiment of the present application;

[0076] Figure 7 A diagram of a multi-nuclide data processing system according to an embodiment of the present application;

[0077] Figure 8 An internal structure diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned objects, features and advantages of the present application more clear, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the present application can be practiced without using some or all of these specific details. In other instances, well-known process steps have not been described in detail in order to avoid unnecessarily obscuring the present application.

[0079] It should be noted that when an element is referred to as being "on" another element, it can be directly on the other element or intervening elements can also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements can also be present. As used herein, the term "substantially" means that the difference between the two values is within an error range that is considered to be equivalent by those skilled in the art. As used herein, the terms "vertical", "horizontal", "left", "right", and similar expressions are used for illustrative purposes only.

[0080] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0081] To solve the technical problems existing in the prior art, the present application provides a multi-nuclide data processing method, device and supporting application which can at least realize multi-nuclide data decay correction.

[0082] In some embodiments, the multi-nuclide data processing method can be executed by a multi-nuclide data processing device. For example, the multi-nuclide data processing method can be partially or entirely stored in the form of programs or instructions in a storage device (such as a self-contained storage module of a detection device or an external storage device), which, when executed, can implement the multi-nuclide data processing method. The multi-nuclide data processing device disclosed in the present application for implementing the above multi-nuclide data processing method can be a device with a large amount of computing resources (for example, a computer, a server, cloud computing, etc.), or a device with limited computing resources (for example, an FPGA (Field Programmable Gate Array, programmable gate array) chip board, an ASIC (Application-Specific Integrated Circuit, application-specific integrated circuit) chip board, etc. hardware circuit).

[0083] Some preferred embodiments of the present application are described below with reference to the accompanying drawings. It should be noted that the following description is for the purpose of illustration and is not intended to limit the scope of protection of the present application.

[0084] Figure 1 For the flowchart of the multi-nuclide data processing method in one embodiment of the present application, in one embodiment, the multi-nuclide data processing method can include the following steps S100 to S400.

[0085] S100: obtaining a to-be-processed coincidence event, and correcting the to-be-processed coincidence event to obtain the coincidence event.

[0086] In different embodiments of the present application, the step S100 is an optional step. In some embodiments, the coincidence events can have been obtained through previous acquisition process and stored in a data storage system, in which case the step S100 can not be performed and the existing coincidence events can be directly read from the data storage system for subsequent processing and analysis. In other embodiments, especially when the method of the embodiments of the present application is applied to real-time monitoring application scenarios, the acquisition and correction of the to-be-processed coincidence events described in the step S100 can be performed synchronously with the subsequent steps to achieve real-time acquisition and processing of data.

[0087] The to-be-processed coincidence events can be acquired by a PET system, come from a variety of coexisting nuclides such as proton therapy or heavy ion therapy or multi-tracer PET, and can be event data acquired in a beam-on state or event data acquired after therapy in a beam-off state.

[0088] It should be particularly noted that a full-digital PET system can accurately extract the energy, position and time information of a single event, and a coincidence system can use energy window discrimination, position exclusion and time window rejection to determine coincidence, obtain coincidence events and obtain projection data. Generally, in a time window, there are two pulses with energy in the energy window range, and the line connecting the crystals on the detectors capturing the two pulses passes through the imaging field of view (FOV), then the two pulses are recorded as a coincidence event. An example of a true coincidence event is shown in Figure 2 In the figure, the outer circular ring is a detection ring, the inner elliptical shape is a test body, the straight line with a bidirectional arrow is a response line, and the arrow indicates the direction of photon flight. The solid point in the figure indicates the position of the nuclide annihilation event. However, due to the limited time window width for detecting true coincidence, two unrelated single events can be mistaken for a pair of coincidence events when they are close in time, which results in random coincidence. An example of a random coincidence event is shown in Figure 3The dashed line corresponds to the event that causes the random coincidence. In addition to the random coincidence, there are background noises in the actual measurement process, such as environmental radiation, electronic noise, etc. These noises form background events, and these events also need to be corrected. Therefore, in an example of the present application, step S100 includes: obtaining to-be-processed coincidence events, and performing random correction, background correction, or both random correction and background correction on the to-be-processed coincidence events to obtain coincidence events. In the embodiment of the present application, the step S100, that is, the random correction and background correction processing, can be performed or at least partially performed before the subsequent decay correction processing, and the time distribution-based count data after random correction and background correction can be used for subsequent decay correction processing, as described further below. However, it is also conceivable that the step S100 is an optional step, for example, in different cases, one or both of random correction and background correction can not be performed as needed.

[0089] The following shows an example of performing both random correction and background correction on to-be-processed coincidence events. In an example of the present application, step S100 includes steps S110-S120.

[0090] Step S110: Obtain to-be-processed coincidence events and process to form a first list set, determine the start time of beam-off, delay the to-be-processed coincidence events and process to form a second list set.

[0091] In some embodiments, the collected coincidence events can be preprocessed. In some embodiments, after the collection is completed, the coincidence events of each channel are arranged in time increasing order within the detector channel, and at this time, sorting can be performed to obtain a coincidence event set arranged in global time increasing order. In other embodiments, the original data that is not arranged in time order can also be sorted, for example, sorting operation according to the timestamp field, etc. It can be understood that the above sorting operation can include merge sorting or other sorting algorithms, which will not be described here.

[0092] Wherein, with respect to beam-off, taking proton therapy as an example, in the treatment process, the target body is bombarded by a proton beam to produce a positron nuclide, and the "beam" in the beam-off is the proton beam, and the beam-off indicates the end of treatment, and there is no proton beam bombarding the target body.

[0093] In some examples of the present application, the i-th to-be-processed coincidence event in the first list set is For example:

[0094] ,

[0095] Wherein, , are the positions of the two single events in the i-th to-be-processed coincidence event, respectively, , are the energies of the two single events, respectively, , are the arrival times of the two single events, respectively.

[0096] The i-th coincidence event in the second list set is, for example,

[0097] ,

[0098] wherein, , are the positions of the two single events in the i-th coincidence event to be processed, , are the energies of the two single events, respectively, , are the delayed arrival times of the two single events, respectively.

[0099] According to an example of the present application, when the coincidence events to be processed are from a proton therapy or heavy ion therapy scenario, the step S110 comprises the following steps S111 and S113.

[0100] Step S111: obtaining a first average time of each of the coincidence events to be processed, taking the first average time as a first reference time of the corresponding coincidence event to be processed, and dividing all the coincidence events to be processed into a plurality of first bins at equal intervals after arranging them according to the first reference time.

[0101] Regarding the first bins, for example, N coincidence events to be processed span 0-100 seconds, and each second is divided into a first bin, so there are 100 first bins.

[0102] Specifically, the step S111 comprises drawing a statistical histogram of the first list mode set L of all N coincidence events to be processed according to the average time of each coincidence event, and the average time of the i-th coincidence event is defined as follows:

[0103] ,

[0104] wherein, is the arrival time of the first single event in the i-th coincidence event, is the arrival time of the second single event in the i-th coincidence event.

[0105] In some embodiments, the generation of the statistical histogram optionally comprises a step of time binning. By way of illustration and not limitation, the statistical histogram can be generated by binning the time axis, i.e. dividing the acquisition period into a series of consecutive and non-overlapping sub-intervals according to a preset or adaptive time parameter (e.g. time length), and the count value of each sub-interval (represented as the height of the column in the statistical histogram) can be obtained by counting the number of coincidence events / single events in each sub-interval. In this embodiment, the time parameter can be selected according to experience, which is not limited in the present application.

[0106] Step S112: Count the number of single events in each first bin, select the first bin whose number of single events exceeds a predetermined threshold, and regard the consecutive first bins as one interval, and take the last interval formed by the selected first bins or the end time of the last selected first bin as the start time of the beam-off.

[0107] Specifically, the size of the first bin can be selected as needed, and assuming that there are n first bins, the number of single events in each first bin is counted respectively, and is represented as follows:

[0108]

[0109] wherein, is the start time of the jth first bin, is the number of single events in the jth first bin, j = 1, …, n.

[0110] It should be particularly noted that the times of all first bins are consecutive, is both the start time of the jth first bin and the end time of the (j-1)th first bin, and the single event falling on the time division node of the two bins is counted as the single event in the previous bin.

[0111] Referring to Figure 4 , a first bin diagram of part of the to-be-processed coincidence events is shown, wherein the red dashed line represents the predetermined threshold, the blue solid line represents the number of single events, and the light green box represents the interval formed by the consecutive first bins. If the last selected first bin does not form an interval with the previous first bin, the end time of the first bin is taken as the start time of the beam-off , otherwise, the end time of the last selected interval is taken as the start time of the beam-off .

[0112] ​The predetermined threshold value can be customized, for example, any value selected from the range of 20% to 50% of the peak of the number of single events in all the first bins, or a value greater than the average of the number of single events in all the first bins, and the specific value can be adjusted according to experience or actual signal characteristics and noise level, and the present application does not limit this.

[0113] Step S113: Obtain the second average time of each of the coincidence events, take the second average time as the second reference time of the corresponding to-be-processed coincidence event, and divide all the to-be-processed coincidence events into a plurality of second bins at equal intervals according to the second reference time, the number of the first bins is the same as the number of the second bins, and one-to-one correspondence exists between the first bins and the second bins.

[0114] In the above example of n first bins, there are n corresponding second bins, and the number of single events in each first bin is counted respectively and represented as follows:

[0115] ,

[0116] wherein, is the start time of the jth second bin, is the number of single events in the jth second bin, j = 1…, n, and all the single events forming the coincidence events can be considered as random coincidence events.

[0117] Step S120: Randomly correcting, background correcting, or simultaneously randomly correcting and background correcting the to-be-processed coincidence events according to the start time of the beam, the first list set and the second list set to obtain coincidence events.

[0118] Specifically, in one example of the present application, step S120 includes the following steps S121-S122.

[0119] Step S121: Subtracting the number of single events in the corresponding second bin from the number of single events in the first bin to achieve random correction.

[0120] In the above example, the number of single events in the first bin formed by the first list set L is subtracted from the number of single events in the second bin formed by the second list set D to obtain the actual coincidence bin , which is represented as follows:

[0121] .

[0122] It can be understood that the random coincidence is corrected in the actual coincidence bin.

[0123] Step S122: Obtain the minimum value of the number of single events in all first bins as the background level, and subtract the background level from the number of single events in the first bins to achieve background correction.

[0124] Exemplarily, the minimum value in the actual coincidence bin results can be selected as the background level based on the actual coincidence bin as the calculation basis :

[0125] ,

[0126] Then, the binning of coincidence events is as follows:

[0127] .

[0128] When applied in the scene of non-proton therapy and heavy ion therapy, the step of multi-exponential fitting of the coincidence events obtained in step S122 is the target observation value of multi-exponential fitting, which reflects the true radioactive decay signal strength.

[0129] According to an example of the present application, when applied in the scene of proton therapy or heavy ion therapy, step S120 further includes the following step S123.

[0130] Step S123: Eliminate coincidence events before the start time of the beam to perform multi-exponential fitting on the remaining coincidence events.

[0131] Specifically, in succession to the above example, the remaining data is represented as follows:

[0132] ,

[0133] Where t off represents the start time of the beam, T k represents the start time of the kth single event, T k-1 represents the start time of the single event before T k .

[0134] Referring to Figure 5 , an actual coincidence binning statistical histogram in a proton therapy scene is shown, and the actual coincidence events therein are the remaining data, and multi-exponential fitting is performed on this part of data.

[0135] S200: Obtain the nuclide category corresponding to the coincidence events according to prior information.

[0136] The prior information is summarized from previous experiments or practical applications. According to the application scene or the particles acting on the object, the induced target generating nuclide category can be summarized. For example, the prior information can be a table including the application scene, the particle category and / or the nuclide category, or the prior information can be a database obtained by simulating the application scene, the particle category and / or the nuclide category in a simulation system.

[0137] For example, in the proton therapy scene, the main components of the positron nuclide induced by protons are 15 O, 11 C and 10 C. For another example, in the heavy ion therapy scene, the main components of the positron nuclide induced by heavy ion beams (such as carbon ions 12 C) are 12 C, 16 O and 14 N. For another example, in the oil exploration scene, the main components of the positron nuclide induced by neutrons are 16 O, 12 C, 28 Si, 27 Al.

[0138] S300: Perform multi-index fitting according to the coincidence event and the nuclide category to obtain the sum of the activities of all nuclides.

[0139] For example, it is assumed that the sum of the activities of all kinds of nuclides represents the total radioactivity in the image, which changes over time, that is:

[0140] ,

[0141] ,

[0142] wherein m=1……M, M is the total number of nuclides, is the time difference relative to the Beam off start time, e is the natural constant, is the decay constant, which is given by the known half-life of each nuclide, is the initial activity coefficient of the mth nuclide at the reference time, in one example, the reference time can be , and the reference time can also be the 0 time.

[0143] Further, taking the Beam off binned data as input, it is assumed that there are first bins in the Beam off interval, from the kth first bin to the nth first bin, the time vector and the observation value vector of the first bin can be obtained, the time vector is , the observation vector is , the time point of each first bin is converted to , a time vector is formed, aiming to unify the time origin to facilitate the fitting of the exponential decay model.

[0144] Specifically, in one example of the present application, step S300 includes:

[0145] Fixed half-life fitting or variable half-life fitting is adopted.

[0146] More specifically, taking the above multi-exponential fitting function as an example: when fixed half-life fitting is adopted, the model function is:

[0147] ,

[0148] where the unknown quantity is only The original nonlinear fitting problem is converted into a linear fitting problem, and the calculation efficiency is higher.

[0149] where the parameter vector A can be represented as: Taking the least squares method as an example, the objective function is:

[0150] ,

[0151] where χ² represents the sum of squared residuals between the observation values and the model predicted values, minimizing this function can find the parameter value that makes the model best fit the observation data, represents the number of single events in the jth interval, represents the time span of the coincidence events in the jth interval after background correction.

[0152] For the linearized problem, it can be written in matrix form:

[0153] ,

[0154] This is the matrix representation of linear regression, where is the design matrix, which contains the exponential decay values of each time point, is the parameter vector to be solved, is the observation vector, this form is convenient for solving using matrix operations.

[0155] where the design matrix is:

[0156] .

[0157] The design matrix each row corresponds to a time point, and each column corresponds to the decay function value of a nuclide, the matrix element Xjm represents the decay factor of the mth nuclide at the jth time point, the matrix constructed in this way makes the linear combination can represent the superimposed decay signal of multiple nuclides.

[0158] Then the solution of the least squares is:

[0159] ,

[0160] This is the analytical solution of the linear least squares problem, and the optimal parameter estimate value is obtained directly through matrix operation. is called Moore-Penrose pseudo-inverse, when is reversible, this solution is the unique optimal solution.

[0161] More specifically, taking the above multi-exponential fitting function as an example: when using variable half-life fitting, the model function is:

[0162] .

[0163] In variable half-life fitting, the decay constant is also taken as a fitting parameter, which can consider the slight change of half-life under actual measurement conditions, but will increase the complexity and instability of fitting.

[0164] The parameter vector is:

[0165] ,

[0166] The constraint condition is:

[0167] .

[0168] In order to prevent the half-life from deviating from the physically reasonable range during fitting, the constraint condition is set to limit the decay constant within the maximum and minimum range of the theoretical value. In an example, the constraint condition is 95%~105% of the initial value. This allows a certain flexibility to adapt to actual conditions, and avoids non-physical fitting results.

[0169] Taking the nonlinear least squares method as an example, the Levenberg-Marquardt algorithm is used to solve:

[0170] ,

[0171] This is a nonlinear optimization problem, which needs to be solved using an iterative algorithm. The Levenberg-Marquardt algorithm combines the advantages of gradient descent and Gauss-Newton methods, and has fast convergence speed when approaching the optimal solution and good stability when far from the optimal solution.

[0172] The Jacobian matrix is:

[0173] ,

[0174] The Jacobian matrix contains the partial derivatives of the model function with respect to each parameter and is the key to calculating the search direction and step size in nonlinear optimization algorithms. The matrix elements J jm represent the sensitivity of the model value at the jth observation point to the mth parameter.

[0175] For the activity coefficient A

[0176] ,

[0177] The partial derivative of the model function with respect to the activity coefficient A m is the corresponding exponential decay factor, which reflects the linear influence of the change in the activity coefficient on the model output.

[0178] For the decay constant λ

[0179] ,

[0180] The partial derivative of the model function with respect to the decay constant λ m contains the time factor Δt, indicating that the influence of the decay constant increases with time, which is consistent with the mathematical properties of the exponential function. The negative sign indicates that the larger the decay constant, the faster the activity decays.

[0181] Take the three-exponential fitting function as an example. In the proton therapy scenario, the main components of the positron nuclide generated by the proton-induced target are 15 O, 11 C, and 10 C. The following three-exponential model is constructed, assuming that the total radioactivity is the superposition of the independent decay of the three nuclides, then:

[0182] ,

[0183] where, 15 the half-life of O is 2.04 minutes, 11 the half-life of C is 20.334 minutes, 10 the half-life of C is 19.29 seconds, and the decay constants are:

[0184] ,

[0185] ,

[0186] ,

[0187] Therefore, in this scenario, , the variables are only , and .

[0188] In the above embodiments of the present application, the coefficients of the exponential terms of the multi-exponential model are in the form of not explicitly multiplying the decay constant, i.e. the coefficient of the exponential term contains the meaning of the decay constant of the corresponding nuclide. Thus, after completing the multi-exponential fitting, if the contribution ratio of each nuclide needs to be calculated, each initial activity coefficient can be divided by its corresponding decay constant, for example, the initial activity coefficient of nuclide m can be divided by its decay constant to obtain the contribution ratio of nuclide m. In subsequent examples, the form of not explicitly multiplying the decay constant is used.

[0189] In other embodiments of the present application, the coefficients of the exponential terms of the multi-exponential model can also be in the form of explicitly multiplying the decay constant, at this time the decay constant is shown as an explicit factor in the model, and the decay constant and another fitting coefficient together act as the coefficient of the exponential term. Thus, after completing the multi-exponential fitting, the aforementioned another fitting coefficient can be directly used to calculate the contribution ratio of the nuclide, which falls within the protection scope of the present application.

[0190] S400: calculating a decay correction factor according to the sum of the decay activities of all the nuclides and the sum of the initial activities of all the nuclides, the decay correction factor being used to perform decay correction on each coincidence event in image reconstruction.

[0191] Specifically, in one example of the present application, the decay correction factor is the ratio of the sum of the initial activities of all the nuclides to the sum of the decay activities of all the nuclides, and in image reconstruction, the decay correction factor is used to perform decay correction on the weight of each coincidence event.

[0192] It needs to be specially pointed out that after completing the multi-exponential fitting, the initial activity coefficient off and the corresponding decay constant of each nuclide Beam off starting time t are obtained. Based on these parameters, the total theoretical activity at any time t, i.e. the total decay activity, can be calculated:

[0193] .

[0194] The core idea of decay correction is to compensate for the decrease in count rate due to radioactive decay. In an ideal case (no decay), the activity at all times should be equal to the total activity at the reference time. Therefore, the decay correction factor (DCF) is defined as:

[0195] ,

[0196] wherein, in the proton therapy or heavy ion therapy scenario, ​Total initial activity at the start time of Beam off.

[0197] Exemplarily, taking the three-exponential fitting model of the above proton therapy scenario as an example, the decay correction factor DCF can be calculated using the following formula:

[0198] .

[0199] The weight of each event is corrected during reconstruction, for example, the decay correction can be completed by multiplying the weight of each event by DCF. Before the decay correction, the weight of each event can include geometric efficiency correction, scattering correction, attenuation correction, and normalization correction, etc. After applying the decay correction of the present application, the corrected weight of the i-th coincidence event is expressed as follows:

[0200] ,

[0201] wherein, wi is the weight of coincidence event i before correction, wi' is the weight after correction.

[0202] It should be particularly pointed out that in the non-proton / heavy ion therapy scenario, the t used in all the above formulas should be replaced by t. .

[0203] The above multi-nuclide data processing method provided by the present application separates and corrects the mixed decay signals of multiple nuclides through multi-exponential fitting, separates the contributions of different nuclides, and can solve the quantization error problem of the traditional single-exponential model in the multi-nuclide scenario, and process the complex scenario of coexistence of multiple nuclides such as proton therapy or heavy ion therapy or multiple tracer PET.

[0204] In addition, compared with the prior art, the present application has the following advantages:

[0205] 1. Support fixed half-life fitting (only fit initial activity coefficient) and variable half-life fitting (allow half-life to adjust within a controllable range), balance calculation efficiency and flexibility.

[0206] 2. Through the decay correction factor (DCF) calculation for the weight of each coincidence event, directly correct the single event in the reconstruction stage, avoid the rough processing of the whole image in the traditional method.

[0207] 3. For the short half-life nuclide (such as 10 C) of “Beam on PET”, the present application can correct the decay effect in real time at the list mode level, avoid the biological clearance effect and signal loss caused by the traditional long acquisition time.

[0208] 4. Through statistical histogram and threshold analysis, the "Beam off" time point in proton / heavy ion therapy is automatically identified, the data during therapy and after therapy are distinguished, and a time reference is provided for decay correction.

[0209] 5. Not only suitable for proton / heavy ion therapy monitoring, but also applicable to multi-tracer dynamic PET, nuclear medicine multi-nuclide imaging, etc., and has cross-field universality.

[0210] Based on the description of the above multi-nuclide data processing method embodiments, the application further provides a multi-nuclide data processing device. The device can include devices (including distributed systems), software (applications), modules, components, servers, clients, etc. that use the methods described in the embodiments of the present application, and devices combined with necessary implementation hardware. Based on the same innovative concept, the device in one or more embodiments provided by the embodiments of the present application is described in the following embodiments. Since the implementation scheme of the device to solve the problem is similar to the method, the implementation of the specific device in the embodiments of the present application can be referred to the implementation of the foregoing method, and the repeated parts will not be described herein. The term "module" or "module" used below can be a combination of software and / or hardware that can implement a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware can also be implemented.

[0211] Figure 6 For the structure diagram of the multi-nuclide data processing device in one of the embodiments of the present application, in one of the embodiments, the multi-nuclide data processing device 600 can include a first correction module 610, a nuclide category acquisition module 620, a fitting module 630, and a second correction module 640.

[0212] Specifically, the first correction module 610 is configured to acquire a to-be-processed coincidence event, and correct the to-be-processed coincidence event to obtain a coincidence event.

[0213] In different embodiments of the present application, the first correction module 610 is an optional module. In some embodiments, the coincidence event can have been obtained through a previous acquisition process and stored in a data storage system, at which time the first correction module 610 is not needed, and the existing coincidence event can be directly read from the data storage system for subsequent processing and analysis.

[0214] The to-be-processed coincidence event can be acquired by a PET system, from a proton therapy or heavy ion therapy or a multi-tracer PET scenario where multiple nuclides coexist, and can be event data acquired during beam (Beam on) or event data acquired after therapy.

[0215] It should be noted that the all-digital PET system can accurately extract the energy, position and time information of single events, and the coincidence system can determine the coincidence by energy window discrimination, position exclusion and time window rejection to obtain the coincidence events and projection data. Generally, in a time window, there are two pulses with energy in the energy window range, and the line connecting the crystals on the detectors capturing the two pulses passes through the imaging field of view (FOV), then the two pulses are recorded as a coincidence event. An example of a true coincidence event is shown in Figure 2 where the outer circle is the detection ring, the inner ellipse is the test body, the straight line with double-headed arrow is the response line, and the arrow indicates the direction of photon flight. The solid dot in the figure indicates the position of the nuclide annihilation event. However, due to the limited width of the time window for detecting true coincidence, two unrelated single events may be mistaken for a pair of coincidence events when they are close in time, resulting in random coincidence. An example of a random coincidence event is shown in Figure 3 where the dashed line corresponds to the event that causes random coincidence. In addition to random coincidence, there are also background noises such as environmental radiation, electronic noise, etc. in actual measurement, which form background events. These events also need to be corrected. Therefore, in an example of the present application, the first correction module 610 is configured to obtain a to-be-processed coincidence event, and perform random correction, background correction, or both random correction and background correction on the to-be-processed coincidence event to obtain a coincidence event.

[0216] The following shows an example of both random correction and background correction on the to-be-processed coincidence event. Specifically, the first correction module 610 is configured to obtain a to-be-processed coincidence event and process it to form a first list set, determine the start time of beam-off, and delay the to-be-processed coincidence event and process it to form a second list set.

[0217] In some embodiments, the acquired coincidence events can be pre-processed. In some embodiments, after acquisition is completed, the coincidence events of each channel are already arranged in time increasing order within the detector channel, and at this time, sorting can be performed to obtain a coincidence event set arranged in global time increasing order. In other embodiments, the original data that is not arranged in time order can also be sorted, for example, sorted according to the timestamp field. It can be understood that the above sorting operation can include merge sorting or other sorting algorithms, which will not be described here.

[0218] wherein, with respect to beam-off, taking proton therapy as an example, during the treatment, the target body is bombarded by a proton beam to produce a positron nuclide, the "beam" in beam-off is the proton beam, and beam-off indicates the end of treatment, i.e. no proton beam bombarding the target body.

[0219] wherein the i-th to-be-processed coincidence event in the first list set For example, are:

[0220] ,

[0221] wherein, , are positions of two single events in the i-th to-be-processed coincidence event, respectively, , are energies of the two single events, respectively, , are arrival times of the two single events, respectively.

[0222] The i-th coincidence event in the second list set is, for example:

[0223] ,

[0224] wherein, , are positions of two single events in the i-th to-be-processed coincidence event, respectively, , are energies of the two single events, respectively, , are delayed arrival times of the two single events, respectively.

[0225] Specifically, when the to-be-processed coincidence events come from a proton therapy or heavy ion therapy scenario, the first correction module 610 is configured to: obtain a first average time of each of the to-be-processed coincidence events, take the first average time as a first reference time of the corresponding to-be-processed coincidence event, and arrange all the to-be-processed coincidence events according to the first reference time and then divide them into a plurality of first bins at equal intervals; count the number of single events in each first bin, select a first bin in which the number of single events exceeds a predetermined threshold, and regard the continuous first bins as one interval, taking the last interval formed by the selected first bins or the end time of the last one of the selected first bins as the start time of beam-off; obtain a second average time of each of the coincidence events, take the second average time as a second reference time of the corresponding to-be-processed coincidence event, and arrange all the to-be-processed coincidence events according to the second reference time and then divide them into a plurality of second bins at equal intervals, the number of the first bins being the same as that of the second bins and corresponding one by one.

[0226] Regarding the first bins, for example, N to-be-processed coincidence events span 1-100 seconds, and each second is divided into a first bin, so there are 100 first bins.

[0227] In particular, the first correction module 610 is configured to draw a statistical histogram of the average time of each coincidence event for the first list mode set L of all N to-be-processed coincidence events, and the average time of the i-th coincidence event is denoted as The definitions are as follows:

[0228] ,

[0229] Wherein, is the arrival time of the first single event in the i-th coincidence event, is the arrival time of the second single event in the i-th coincidence event.

[0230] In particular, the size of the first bin can be selected as needed, assuming that there are n first bins, and the number of single events in each first bin is counted respectively, denoted as follows:

[0231] ,

[0232] Wherein, is the start time of the j-th first bin, is the number of single events in the j-th bin, j = 1…, n.

[0233] It needs to be particularly pointed out that the times of all first bins are continuous, is the start time of the j-th bin, and is also the end time of the j-1-th bin, and the single event falling on the time division node of the two bins is counted as the single event in the previous bin.

[0234] Referring to Figure 4 , a first bin diagram of part of the to-be-processed coincidence events is shown, wherein the red dashed line represents the predetermined threshold, the blue solid line represents the number of single events, and the light green box represents the interval formed by the continuous first bins. If the last first bin does not form an interval with the previous first bin, the end time of the first bin is taken as the start time of the Beam off , otherwise, the end time of the last interval is taken as the start time of the Beam off .

[0235] Regarding the predetermined threshold, it can be customized, for example, any value selected in the range of 20% to 50% of the peak value of the number of single events in all first bins, or a value greater than the average value of the number of single events in all first bins.

[0236] Following the above example of n first bins, there are n second bins corresponding to the first bins, and the number of single events in each first bin is counted respectively, denoted as follows:

[0237] ,

[0238] in, Let j be the start time of the second sub-box. Let be the number of single events in the j-th bin, where j = 1, ..., n. The composite events formed by all the single events can be considered as random composite events.

[0239] Specifically, the first correction module 610 is further configured to: perform random correction, background correction, or simultaneous random correction and background correction on the matching event to be processed based on the start time of the beam separation, the first list set, and the second list set to obtain the matching event.

[0240] More specifically, the first correction module 610 is configured to: subtract the corresponding number of single events in the second sub-bin from the number of single events in the first sub-bin to achieve random correction; obtain the minimum value of the number of single events in all the first sub-bins as the background level; and subtract the corresponding number of single events in the second sub-bin and the background level from the number of single events in the first sub-bin to achieve random correction and background correction to obtain the conforming event.

[0241] Continuing with the example above, subtract the number of single events in the first bin formed by the first list set L from the number of single events in the second bin formed by the second list set D to obtain the actual binning results. , means as follows:

[0242] .

[0243] Understandably, random coincidences were corrected in the actual coincidence binning.

[0244] For example, the actual binning conformance can be used as the basis for calculation, and the minimum value among the actual binning conformance results can be selected as the background level. :

[0245] ,

[0246] Then, binning that matches the event as follows:

[0247] .

[0248] When applied to nonproton therapy and heavy ion therapy scenarios, multi-exponential fitting is performed using the coincidence events obtained here.

[0249] According to an example of this application, when applied in proton therapy or heavy ion therapy scenarios, the first correction module 610 is configured to: remove coincidence events prior to the start time of the beam separation, and perform multi-exponential fitting with the remaining coincidence events.

[0250] Specifically, continuing the above example, the residual data is represented as follows:

[0251] ,

[0252] where t off represents the start time of the beam, T k represents the start time of the kth single event, T k-1 represents the start time of the kth single event, T k represents the start time of the single event before T

[0253] Referring to Figure 5 , a real coincidence binning statistical histogram in a proton therapy scenario is shown, where the data below the yellow dashed line is the coincidence event or residual data, and a multi-exponential fitting is performed on this part of data.

[0254] Specifically, the nuclide category obtaining module 620 is configured to obtain the nuclide category corresponding to the coincidence event according to prior information. The prior information is obtained from past experiments or practical applications, and according to the application scenario or the particles acting on the object, the nuclide category generated by the target body can be summarized. For example, the prior information can be a table including the application scenario, the particle category, and the nuclide category, or the prior information can be a database obtained by simulating the application scenario, the particle category, and / or the nuclide category in a simulation system.

[0255] For example, taking the proton therapy scenario as an example, the main components of the positron nuclide generated by the target body induced by protons are 15 O, 11 C and 10 C. For another example, taking the heavy ion therapy scenario as an example, the main components of the positron nuclide generated by the target body induced by heavy ion beams (such as carbon ions 12 C) are 12 C, 16 O and 14 N. For another example, taking the oil exploration scenario as an example, the main components of the positron nuclide generated by neutrons are 16 O, 12 C, 28 Si, 27 Al.

[0256] Specifically, the fitting module 630 is configured to perform a multi-exponential fitting on the coincidence event and the nuclide category to obtain the sum of the decay activities of all nuclides.

[0257] For example, the sum of the decay activities of all kinds of nuclides is set to represent the total radioactivity in the image, which changes with time, i.e.

[0258] ,

[0259] ,

[0260] where the exponential decay constant is given by the known half-life of each nuclide , is the initial activity coefficient of each nuclide at the reference time.

[0261] Take the three-exponential fitting function as an example. In the proton therapy scenario, the main components of the positron nuclides generated by the proton-induced target are 15 O, 11 C and 10 C. The following three-exponential model is constructed, assuming that the total radioactivity is the superposition of the independent decay of the three nuclides, that is:

[0262] ,

[0263] where, 15 the half-life of O is 2.04 minutes, 11 the half-life of C is 20.334 minutes, and the half-life of 10C is 19.29 seconds.

[0264] More specifically, the fitting module 630 is configured to use fixed half-life fitting or variable half-life fitting.

[0265] Taking the above multi-exponential fitting function as an example: when using fixed half-life fitting, the unknown quantity is only , and the model function in the fitting formula is only related to . In the three-exponential fitting scenario, there are 3 variables to be fitted; when using variable half-life fitting, the unknown quantity of the model function has and , and the model function in the fitting formula is related to and . In the three-exponential fitting scenario, there are 6 variables to be fitted. In actual engineering applications, fixed half-life fitting or variable half-life fitting with constraints is preferably used, for example, the change range of the half-life is reduced to a controllable range (e.g. 95%~105% of the initial value).

[0266] Specifically, in one example of the present application, the fitting module 630 is configured to use a non-linear least squares method to perform multi-exponential fitting with the coincidence events as input, and estimate the best square approximation to obtain the sum of the decay activities of all nuclides. For example, taking as input, the best square approximation of the multi-exponential function is estimated.

[0267] In one example of the present application, the fitting module 630 is configured to set the sum of the activity of all kinds of nuclides to represent the total radioactivity in the image, which changes over time, i.e.:

[0268] ,

[0269] ,

[0270] wherein m = 1 … M, M is the total number of nuclides, is the time difference relative to the Beam off start time, is the decay constant, which is derived from the known half-life of each nuclide is given, is the initial activity coefficient of the m-th nuclide at the reference time, in one example, the reference time is may be , the reference time can also be the 0 time.

[0271] Further, taking the Beam off binned data as input, assuming that there are first bins in the Beam off interval, from the k-th first bin to the n-th first bin, the time vector and the observation value vector of the first bin can be obtained, the time vector is , and the observation value vector is , the time point of each first bin is converted into to form a time vector, the purpose is to unify the time starting point in order to fit the exponential decay model.

[0272] Further, the fitting module 630 is configured to use fixed half-life fitting or variable half-life fitting.

[0273] More specifically, taking the above multi-exponential fitting function as an example: when using fixed half-life fitting, the model function is:

[0274] ,

[0275] wherein the unknown quantity is only , which converts the original nonlinear fitting problem into a linear fitting problem, and the calculation efficiency is higher.

[0276] wherein the parameter vector A can be represented as: , taking the least squares method as an example, the objective function is:

[0277] ,

[0278] where χ2represents the sum of squared residuals between the observed values and the model predicted values, minimizing this function can find the parameter values that make the model best fit the observed data, represents the number of single events in the jth interval, represents the time span of coincidence events in the jth interval after background correction.

[0279] For the linearized problem, it can be written in matrix form:

[0280] ,

[0281] This is the matrix representation of linear regression, where is the design matrix, containing the exponential decay values at each time point, is the parameter vector to be solved, is the observed value vector, this form is convenient for solving using matrix operations.

[0282] where the design matrix is:

[0283] .

[0284] The design matrix each row corresponds to a time point, and each column corresponds to the decay function value of a nuclide, the matrix element X jm represents the decay factor of the mth nuclide at the jth time point, the matrix constructed in this way makes the linear combination can represent the superimposed decay signal of multiple nuclides.

[0285] Then the least squares solution is:

[0286] ,

[0287] This is the analytical solution of the linear least squares problem, and the optimal parameter estimate value is obtained directly through matrix operations. is called the Moore-Penrose pseudo-inverse, when is invertible, this solution is the unique optimal solution.

[0288] More specifically, taking the above multi-exponential fitting function as an example: when using variable half-life fitting, the model function is:

[0289] ,

[0290] In variable half-life fitting, the decay constant is also considered as a fitting parameter, which can consider the small changes of half-life under actual measurement conditions, but will increase the complexity and instability of fitting.

[0291] The parameter vector is:

[0292] ,

[0293] The constraint is:

[0294] .

[0295] To prevent the half-life from deviating from the physically reasonable range during fitting, a constraint is set to limit the decay constant within the maximum and minimum range of the theoretical value. In one example, the constraint is 95% ~ 105% of the initial value. This allows a certain flexibility to adapt to actual conditions, while avoiding non-physical fitting results.

[0296] Taking the nonlinear least squares method as an example, the Levenberg-Marquardt algorithm is used to solve:

[0297] ,

[0298] This is a nonlinear optimization problem, which needs to be solved using an iterative algorithm. The Levenberg-Marquardt algorithm combines the advantages of gradient descent and Gauss-Newton methods, with fast convergence near the optimal solution and good stability far from the optimal solution.

[0299] The Jacobian matrix is:

[0300] ,

[0301] The Jacobian matrix contains the partial derivatives of the model function with respect to each parameter, which is the key to calculating the search direction and step size in nonlinear optimization algorithms. The matrix element J jm represents the sensitivity of the model value at the jth observation point to the mth parameter.

[0302] For the activity coefficient:

[0303] ,

[0304] The partial derivative of the model function with respect to the activity coefficient A m is the corresponding exponential decay factor, which reflects the linear influence of the activity coefficient change on the model output.

[0305] For the decay constant:

[0306] ,

[0307] The partial derivative of the model function with respect to the decay constant λ m contains the time factor Δt, indicating that the influence of the decay constant increases with time, which is consistent with the mathematical properties of the exponential function. The negative sign indicates that the larger the decay constant, the faster the activity decays.

[0308] Take the three-index fitting function as an example. In the proton therapy scenario, the main components of the positron nuclide generated by the proton-induced target body are 15 O, 11 C and 10 C. The following three-index model is constructed, assuming that the total radioactivity is the superposition of the independent decay of the three nuclides, that is:

[0309] ,

[0310] Among them, 15 The half-life of O is 2.04 minutes, 11 The half-life of C is 20.334 minutes, 10 The half-life of C is 19.29 seconds, and the decay constants are respectively:

[0311] ,

[0312] ,

[0313] ,

[0314] Therefore, in this scenario, The variable is only , and .

[0315] In the above specific embodiments of the present application, the coefficients of the index terms of the multi-index model are in the form of not explicitly multiplying the decay constant, that is, the coefficient of the index term Itself contains the meaning of the decay constant of the corresponding nuclide. Therefore, after completing the multi-index fitting, if the contribution ratio of each nuclide needs to be calculated, each initial activity coefficient can be divided by its corresponding decay constant, for example, the contribution ratio of nuclide m can be obtained by dividing By In subsequent examples, this form of not explicitly multiplying the decay constant is reused.

[0316] In some embodiments of the present application, the coefficients of the index terms of the multi-index model can also be in the form of explicitly multiplying the decay constant, at this time the decay constant is displayed as an explicit factor in the model, and the decay constant and another fitting coefficient are used as the coefficient of the index term. Therefore, after completing the multi-index fitting, the aforementioned another fitting coefficient can be directly used to calculate the contribution ratio of the nuclide, which falls within the protection scope of the present application.

[0317] Specifically, the second correction module 640 is configured to calculate a decay correction factor based on the sum of the decay activities of all nuclides and the sum of the initial activities of all nuclides. This decay correction factor is used to perform decay correction on each coincidence event during image reconstruction. More specifically, the decay correction factor is the ratio of the sum of the initial activities of all nuclides to the sum of the decay activities of all nuclides. During image reconstruction, the second correction module 640 uses this decay correction factor to perform decay correction on the weight of each coincidence event.

[0318] Specifically, in one example of this application, the decay correction factor is the ratio of the sum of the initial activities of all nuclides to the sum of the decay activities of all nuclides. During image reconstruction, the decay correction factor is used to perform decay correction on the weight of each coincident event.

[0319] It should be noted that after completing the multi-exponential fitting, the beamoff start time t for each nuclide was obtained. off initial activity coefficient and the corresponding decay constant Based on these parameters, the total theoretical activity, or total decay activity, at any time t can be calculated:

[0320] .

[0321] The core idea of ​​decay correction is to compensate for the decrease in count rate caused by radioactive decay. Ideally (without decay), the activity at all times should equal the total activity at the reference time. Therefore, the DecayCorrection Factor (DCF) is defined as follows:

[0322] ,

[0323] In proton therapy or heavy ion therapy scenarios, , where is the total initial activity at the start of Beam off.

[0324] For example, taking the triple-exponential fitting model of the proton therapy scenario described above as an example, the decay correction factor DCF can be calculated using the following formula:

[0325] .

[0326] During reconstruction, the weights of each event are corrected. For example, decay correction can be achieved by multiplying the weight of each event by the DCF. Before decay correction, the weights of each event may include geometric efficiency correction, scattering correction, attenuation correction, and normalization correction. After applying the decay correction method of this application, the corrected weight of the i-th coincident event is expressed by the following formula:

[0327] ,

[0328] wherein, is the weight of coincidence event i before correction, is the weight after correction.

[0329] The above multi-nuclide data processing device provided by the present application separates and corrects the mixed decay signals of multiple nuclides through multi-exponential fitting, separates the contributions of different nuclides, and can solve the quantization error problem of the traditional single-exponential model in the multi-nuclide scene, and process the complex scene of multiple nuclides coexisting such as proton therapy or heavy ion therapy or multi-tracer PET.

[0330] In addition, compared with the prior art, the present application has the following advantages:

[0331] 1. Support fixed half-life fitting (fit only initial activity coefficient) and variable half-life fitting (allow half-life to adjust within controllable range), balance calculation efficiency and flexibility.

[0332] 2. Through the decay correction factor (DCF) calculation of the weight of each coincidence event, directly correct the single event in the reconstruction stage, avoid the rough processing of the whole image in the traditional method.

[0333] 3. For the short half-life nuclide (such as 10 C) of “Beam on PET”, the present application can correct the decay effect in real time at the list mode level, avoid the biological clearance effect and signal loss caused by the traditional long acquisition time.

[0334] 4. Through statistical histogram and threshold analysis, automatically identify the “Beam off” time point in proton / heavy ion therapy, distinguish the data during treatment and after treatment, and provide time reference for decay correction.

[0335] 5. Not only suitable for proton / heavy ion therapy monitoring, but also can be extended to multi-tracer dynamic PET, nuclear medicine multi-nuclide imaging and other scenes, with cross-field universality.

[0336] It should be understood that, Figure 6The apparatuses and modules thereof shown can be implemented in various ways. For example, in some embodiments, the apparatuses and modules thereof can be implemented in hardware, software, or a combination of software and hardware. The hardware portion can be implemented with special logic, while the software portion can be stored in memory and executed by suitable instruction execution apparatuses, such as a microprocessor or specially designed hardware. Those skilled in the art can understand that the above-described methods and apparatuses can be implemented using computer-executable instructions and / or included in processor control codes, such as provided on a carrier medium, such as a diskette, CD or DVD-ROM, programmable memory (firmware), or data carrier such as an optical or electrical signal carrier. The apparatuses and modules thereof of the present application can be implemented not only in hardware circuitry, such as very large scale integrated circuits or gate arrays, such as logic chips, transistors, etc., or programmable hardware devices, such as field programmable gate arrays, programmable logic devices, etc., but also in software, for example, executed by various types of processors, and also by a combination of the above-mentioned hardware circuitry and software (e.g., firmware).

[0337] It should be noted that the above description of the modules is for convenience of description only and does not limit the present application to the scope of the embodiments. It can be understood that, for those skilled in the art, after understanding the principles of the apparatus, the modules can be combined arbitrarily or connected to other modules to form subsystems without departing from the principles. For example, the modules can share a storage module, and each module can have its own storage module. Variations such as these are within the scope of the present application.

[0338] Figure 7 A decay correction system for implementing the multi-nuclide data processing method in one embodiment of the present application is shown schematically. Referring to Figure 7 , the decay correction system S00 can include a processing component S20, which further includes one or more processors, and a memory resource represented by a memory S22, for storing instructions, such as application programs, executable by the processors of the processing component S20. The application programs stored in the memory S22 can include one or more instructions, each module corresponding to a set of instructions. In addition, the processing component S20 is configured to execute the instructions to perform the above-described multi-nuclide data processing method.

[0339] The operations and / or methods described in the embodiments of the specification implemented by one processor can also be implemented by multiple processors collectively or independently. For example, if the processor of the processing device performs steps S100-S400 in the specification of the present application, it should be understood that steps S100-S400 can also be performed collectively or independently by two different processors of the processing device (for example, a first processor performs steps S100-S200, a second processor performs steps S300-S400, or the first and second processors collectively perform steps S100-S400).

[0340] The decay correction system S00 can also include a power component S24 configured to perform power management of the decay correction system S00, a wired or wireless network interface S26 configured to connect the decay correction system S00 to a network, and an input / output (I / O) interface S28. The decay correction system S00 can operate based on an operating system stored in the memory S22, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD or the like.

[0341] In an exemplary embodiment, a computer readable storage medium including instructions, such as the memory S22 including instructions, is also provided, which can be executed by the processor of the multi-nucleus data processing system S00 to complete the above method. The storage medium can be a computer readable storage medium, for example, the computer readable storage medium can be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk and optical data storage device, etc.

[0342] In an exemplary embodiment, a computer program product is also provided, which includes instructions in the computer program product, which can be executed by the processor of the decay correction system S00 to complete the above method.

[0343] In one embodiment, a computer device, which can be a server, is provided, and its internal structure diagram can be as shown in Figure 8 Figure 8 ​The internal structure diagram of the computer device in one of the embodiments of the present application is shown. The computer device comprises a processor, a memory and a network interface connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store data related to users and tasks used in the above multi-nucleus data processing method. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a multi-nucleus data processing method.

[0344] Those skilled in the art can understand that, Figure 8 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can comprise more or fewer components than those shown in the figure, or some components can be combined, or have a different arrangement of components.

[0345] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in each embodiment provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in each embodiment provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in each embodiment provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0346] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for hardware+program type embodiments, since they are basically similar to method embodiments, they are described more simply, and the relevant parts can be referred to the part of the method embodiment.

[0347] It should be noted that the apparatus, electronic device, server and the like described above according to the description of the method embodiments can also include other implementations, and the specific implementation can be referred to the description of the related method embodiments. Meanwhile, the mutual combination of the features of each method and the device, apparatus, server embodiments constitutes a new embodiment, which still falls within the scope of the embodiments covered by the present application, and is not described one by one here.

[0348] In the description of the present specification, the description of the terms "one embodiment", "an embodiment", and / or "some embodiments", "some embodiments", "other embodiments", "ideal embodiment" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example, and some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0349] The technical features of the above-described embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above-described embodiments are not described, however, as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present specification.

[0350] The above-described embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the present application. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

[0351] The basic concepts have been described herein, and it is obvious that the above detailed disclosure is only used as an example and does not constitute a limitation on the present specification. Although it is not explicitly stated herein, those skilled in the art can make various modifications, improvements and corrections to the present specification. Such modifications, improvements and corrections are suggested in the present specification, so such modifications, improvements and corrections still fall within the spirit and scope of the exemplary embodiments of the present specification.

[0352] Moreover, those skilled in the art will appreciate that the various aspects of the disclosure can be illustrated and described in connection with a number of various kinds of systems or circumstances, including any new and useful processes, machines, products, or compositions of matter, or any new and useful improvements thereof, as defined by the plain language of the claims. Accordingly, the various aspects of the disclosure can be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by combinations of hardware and software. The above described hardware or software can be referred to as a "data block", "module", "engine", "component", "element", or "system". Furthermore, the various aspects of the disclosure can be embodied as computer products including computer readable program coding.

[0353] Computer storage media can include a propagated data signal with computer program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal can take any of a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. Computer storage media can be any media that can be accessed by a computer. As used herein, "computer storage media" includes one or more of volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program code, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, solid state drives (SSDs) that are based on RAM, flash memory or other solid state memory

[0354] Computer program code for carrying out operations of the aspects of the disclosure can be written in any one or more of a variety of programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, or a conventional procedural programming language such as C language, Visual Basic, Fortran 3003, Perl, COBOL 3002, PHP, ABAP, a dynamic programming language such as Python, Ruby and Groovy, or other programming languages. The program code can execute entirely on the user's computer, or partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider) or within cloud computing

[0355] Furthermore, the order of the processing elements and sequences described in this specification are not intended to be construed as a limitation, unless specifically stated, but are included to provide a complete description of one or more embodiments of the present specification. Regardless of the particular sequence of processing elements and sequences, however, the description herein of a process should be understood to include any and all combinations of one or more elements, and sequences that can be perceived as either open-ended or specific.

[0356] Similarly, it is to be noticed that the term "comprising", used in the description, should not be interpreted as being restricted only to the means listed thereafter. It is to be understood that the term "comprising" means that any additional element, which is not specifically mentioned, is optionally present or can be added. In some embodiments, the description of an embodiment using the term "comprising" can also be interpreted as using the term "including" or "consisting of". Furthermore, the description herein of any particular embodiment of the present specification is intended to be illustrative only and is not intended to be limiting unless specifically stated. Thus, while the present specification has been described in terms of one or more embodiments, it is to be understood that the embodiments disclosed are not to be limited to the specific forms or methods described unless otherwise specifically stated.

[0357] Some embodiments use numerical designations to describe components, quantities of attributes. It is to be understood that such numerical designations used in the description of embodiments are, in some examples, modified by the adjectives "about", "approximately", or "generally". Unless otherwise stated, "about", "approximately", or "generally" indicates that the stated numerical value is permitted to vary by ±20%. Accordingly, in some embodiments, numerical parameters in the description and claims are approximations that can vary depending on the requirements of the particular embodiments. In some embodiments, numerical parameters should be considered in the context of the number of significant digits used for measurement and the general accuracy associated with such measurements. Although the numerical ranges and parameters setting forth the broadest scope of the embodiments described herein are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values set forth in the specific examples are provided to give a general understanding of the embodiments.

[0358] Each patent, patent application, publication, and other material cited in this specification is incorporated herein by reference in its entirety for the teachings relevant to the sentence and / or paragraph in which the reference is presented. Document histories, to the extent not inconsistent with this specification, are excluded to the extent that they are inconsistent or contradictory with the content of this specification, to the extent that they limit the broadest scope of the claims of this specification, and to the extent that they are added to this specification after the date of this specification. It is to be noted that, where descriptions, definitions, and / or terminology used in the attached material are inconsistent or in conflict with this specification, the descriptions, definitions, and / or terminology used in this specification prevail.

[0359] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims that follow.

Claims

1. A multinuclei data processing method, characterized by, The method comprises the following steps: obtaining a nuclide category corresponding to the coincidence event according to prior information; performing multi-index fitting according to the coincidence event and the nuclide category to obtain a sum of decay activities of all nuclides; calculating a decay correction factor according to the sum of decay activities of all nuclides and a sum of initial activities of all nuclides, the decay correction factor being used for decay correction of each coincidence event during image reconstruction.

2. The multinuclei data processing method of claim 1, wherein, Before obtaining the nuclide category corresponding to the coincidence event according to prior information, the multi-nuclide data processing method further comprises the following steps: obtaining a to-be-processed coincidence event, and correcting the to-be-processed coincidence event to obtain the coincidence event.

3. The multi-nucleus data processing method according to claim 2, wherein, The correction of the to-be-processed coincidence event comprises random correction, background correction, or random correction and background correction simultaneously.

4. The multi-nucleus data processing method according to claim 2, wherein, The method comprises the following steps: obtaining a to-be-processed coincidence event, and processing the to-be-processed coincidence event to form a first list set, determining a beam-off start time, delaying the to-be-processed coincidence event and processing the to-be-processed coincidence event to form a second list set; performing random correction, background correction, or random correction and background correction simultaneously on the to-be-processed coincidence event according to the beam-off start time, the first list set and the second list set to obtain the coincidence event.

5. The multinuclei data processing method of claim 4, wherein, The method comprises the following steps: obtaining a to-be-processed coincidence event, and processing the to-be-processed coincidence event to form a first list set, determining a beam-off start time, delaying the to-be-processed coincidence event and processing the to-be-processed coincidence event to form a second list set; obtaining a first average time of each to-be-processed coincidence event, taking the first average time as a first reference time of the corresponding to-be-processed coincidence event, and equally dividing a plurality of first bins formed by arranging all the to-be-processed coincidence events according to the first reference time; 6. The multinuclei data processing method of claim 5, wherein, obtaining a second average time of each coincidence event, taking the second average time as a second reference time of the corresponding to-be-processed coincidence event, and equally dividing a plurality of second bins formed by arranging all the to-be-processed coincidence events according to the second reference time, the number of the first bins being the same as the number of the second bins, and the first bins and the second bins corresponding to each other. The method comprises the following steps:

7. The multi-nucleus data processing method according to claim 5, wherein, counting the number of single events in each first bin, selecting a first bin with a number of single events exceeding a predetermined threshold, and regarding continuous first bins as a same interval, and taking the end time of the last interval or the first bin as the beam-off start time. The method comprises the following steps: subtracting the number of single events in the corresponding second bin from the number of single events in the first bin to realize random correction; obtaining a minimum value of the number of single events in all first bins as a background level, and subtracting the background level from the number of single events in the first bin to realize background correction.

8. The polyatomic data processing method of claim 7, wherein, Randomly correcting, background correcting, or both randomly correcting and background correcting the to-be-processed coincidence events according to the start time of the off-beam, the first list set and the second list set to obtain the coincidence events, comprising: Rejecting the coincidence events before the start time of the off-beam, and performing multi-exponential fitting on the remaining coincidence events.

9. The multi-nucleus data processing method according to claim 1, wherein, Performing multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities, comprising: Using a non-linear least square method to perform multi-exponential fitting with the coincidence events as input, and estimating the best square approximation to obtain the sum of all nuclide decay activities.

10. The polyatomic data processing method of claim 9, wherein, Performing multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities, comprising: Setting the sum of decay activities of all kinds of nuclides to represent the total radioactivity in the image, and using a fixed half-life fitting or a variable half-life fitting.

11. The polyatomic data processing method of claim 10, wherein, Performing multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities, comprising: The model function of the multi-exponential fitting is: ， wherein, is a parameter vector, , is the initial activity coefficient of the mth nuclide at the reference time, M is the total number of nuclides, is the decay constant of the mth nuclide, is the time span of the coincidence events after background correction.

12. The polyatomic data processing method of claim 11, wherein, Performing multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities, comprising: Using a fixed half-life and a least square method to perform fitting, and the objective function is: ， where χ2represents the residual sum of squares between the observed values and the model predicted values, represents the number of bins that meet the event, represents the number of single events in the jth interval, represents the time span of the events that meet the event in the jth interval after background correction.

13. The multi-nucleus data processing method according to claim 10, wherein, Performing multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities, comprising: Using a variable half-life and a least square method to perform fitting, and the model function is: ， The objective function is: ， wherein, denotes a parameter vector, , denotes the number of bins that fit the events, denotes the number of single events in the jth interval, denotes the initial activity coefficient of the mth nuclide at the reference time, M denotes the total number of nuclides, denotes the decay constant of the mth nuclide, denotes the time span of the coincidence events after background correction, denotes the time span of the coincidence events in the jth interval after background correction.

14. The polyatomic data processing method of claim 13, wherein, The decay constant is limited in the range of the maximum value and the minimum value of the theoretical value.

15. The multi-energic data processing method of claim 1, wherein, Calculating a decay correction factor according to the sum of all nuclide decay activities and the sum of all nuclide initial activities, the decay correction factor being used to decay correct each coincidence event in image reconstruction, comprising: The decay correction factor is the ratio of the sum of all nuclide initial activities to the sum of all nuclide decay activities, and the decay correction factor is used to decay correct the weight of each coincidence event in image reconstruction.

16. The polyatomic data processing method of claim 15, wherein, The formula of the decay correction factor is: ， wherein, denotes a decay correction factor at time t, denotes an initial activity coefficient of the mth nuclide at a reference time, M denotes the total number of nuclides, denotes a decay constant of the mth nuclide.

17. The polyatomic data processing method of claim 15, wherein, Decay correcting the weight of each coincidence event in image reconstruction using the decay correction factor, comprising: decay correcting the weight of each coincidence event using the following formula: ， wherein, represents the weight of coincidence event i before correction, represents the weight after correction, represents the decay correction factor of coincidence event i at time t.

18. A multi- nucleic data processing device, characterized by Comprising: A nuclide category acquisition module configured to obtain the nuclide category corresponding to the coincidence events according to prior information; A fitting module configured to perform multi-exponential fitting according to the coincidence events and the nuclide categories to obtain the sum of all nuclide decay activities; A second correction module configured to calculate a decay correction factor according to the sum of all nuclide decay activities and the sum of all nuclide initial activities, the decay correction factor being used to decay correct each coincidence event in image reconstruction.

19. The polyatomic data processing apparatus of claim 18, wherein, Further comprising: A first correction module configured to obtain to-be-processed coincidence events, and correct the to-be-processed coincidence events to obtain coincidence events.

20. The polyatomic data processing apparatus of claim 19, wherein, The first correction module is configured to obtain to-be-processed coincidence events, and randomly correct, background correct, or both randomly correct and background correct the to-be-processed coincidence events to obtain coincidence events.

21. The multi-nucleus data processing apparatus according to claim 19, wherein, The first correction module is configured to: The method comprises the following steps: obtaining to-be-processed coincidence events and processing to form a first list set, determining a start time of beam-off, delaying the to-be-processed coincidence events and processing to form a second list set; According to the start time of beam-off, the first list set and the second list set, the to-be-processed coincidence events are randomly corrected, background corrected or simultaneously randomly corrected and background corrected to obtain coincidence events.

22. The polyatomic data processing device of claim 21, wherein, The first correction module is configured to: Obtain a first average time of each to-be-processed coincidence event, take the first average time as a first reference time of the corresponding to-be-processed coincidence event, and divide all the to-be-processed coincidence events into a plurality of first bins at equal intervals according to the first reference time; Obtain a second average time of each coincidence event, take the second average time as a second reference time of the corresponding to-be-processed coincidence event, and divide all the to-be-processed coincidence events into a plurality of second bins at equal intervals according to the second reference time, the number of the first bins is the same as the number of the second bins, and one-to-one correspondence.

23. The polyatomic data processing apparatus of claim 22, wherein, The first correction module is configured to: Count the number of single events in each first bin, select a first bin whose number of single events exceeds a predetermined threshold, and regard the continuous first bins as a same interval, and take the end time of the last interval or the first bin as the start time of beam-off.

24. The polyatomic data processing apparatus of claim 22, wherein, The first correction module is configured to: Subtract the number of single events in the corresponding second bin from the number of single events in the first bin to realize random correction; Obtain the minimum value of the number of single events in all first bins as a background level, and subtract the background level from the number of single events in the first bin to realize background correction.

25. The polyatomic data processing device of claim 24, wherein, The first correction module is configured to: Eliminate coincidence events before the start time of beam-off, and perform multi-exponential fitting on the remaining coincidence events.

26. The polyatomic data processing apparatus of claim 18, wherein, The fitting module is configured to: Take the coincidence events as input, perform multi-exponential fitting by using a non-linear least square method, and estimate the sum of all nuclide decay activities by using a best square approximation.

27. The polyatomic data processing device of claim 26, wherein, The fitting module is configured to: Set the sum of all nuclide decay activities of all kinds to represent the total radioactivity in the image, and perform fitting by using a fixed half-life or by using a variable half-life.

28. The polyatomic data processing apparatus of claim 27, wherein, The model function of multi-exponential fitting is: ， wherein, is a parameter vector, , is the initial activity coefficient of the mth nuclide at the reference time, M is the total number of nuclides, is the decay constant of the mth nuclide, is the time span of the coincidence events after background correction.

29. The polyatomic data processing apparatus of claim 28, wherein, The fitting module is configured to perform fitting by using a fixed half-life and a least square method, and the objective function is: ， where χ2represents the residual sum of squares between the observed values and the model predicted values, represents the number of bins that meet the event, represents the number of single events in the jth interval, represents the time span of the events that meet the event in the jth interval after background correction.

30. The polyatomic data processing apparatus of claim 27, wherein, The fitting module is configured to perform fitting by using a variable half-life and a least square method, and the model function is: ， The objective function is: ， wherein, denotes a parameter vector, , denotes the number of bins that meet the event, denotes the number of single events in the jth interval, denotes the initial activity coefficient of the mth nuclide at the reference time, M denotes the total number of nuclides, denotes the decay constant of the mth nuclide, denotes the time span of the coincidence event after background correction, denotes the time span of the coincidence event in the jth interval after background correction.

31. The multi-energic data processing device according to claim 30, wherein, The decay constant is limited in a range of a maximum value and a minimum value of a theoretical value.

32. The polyatomic data processing device of claim 18, wherein, The decay correction factor is a ratio of the sum of all nuclide initial activities to the sum of all nuclide decay activities, and the second correction module is configured to perform decay correction on the weight of each coincidence event by using the decay correction factor during image reconstruction.

33. The polyatomic data processing device of claim 32, wherein, The formula of the decay correction factor is: ， wherein, denotes a decay correction factor at time t, denotes an initial activity coefficient of the mth nuclide at a reference time, M denotes the total number of nuclides, denotes a decay constant of the mth nuclide.

34. The polyatomic data processing apparatus of claim 32, wherein, During image reconstruction, the weight of each coincidence event is decay-corrected by using the decay correction factor, which comprises: decay-correcting the weight of each coincidence event by using the following formula: ， wherein, represents the weight of coincidence event i before correction, represents the weight after correction, represents the decay correction factor of coincidence event i at time t.

35. A computer storage medium having stored thereon a computer program, characterized in that The computer program, when executed by a processor, implements the steps of the multi-nuclide data processing method as claimed in any one of claims 1 to 17.

36. A computer program product, characterised in that, A computer program or instructions, characterized in that the computer program or instructions, when executed by a processor, implement the steps of the multi-nuclide data processing method as claimed in any one of claims 1 to 17.

37. A digital PET system characterized by, The multi-nuclide data processing apparatus as claimed in any one of claims 18 to 34 is used for decay correction at image reconstruction.

Images