Nuclear medicine radioactivity purity on-line detection system based on energy spectrum analysis

Abstract

The present application relates to the technical field of nuclear radiation, and discloses a nuclear medicine radioactivity purity online detection system based on energy spectrum analysis, which comprises a synchronous signal acquisition module, a real-time background suppression module for carrying out correlation screening on the gamma energy spectrum based on the environmental background radiation to obtain a background suppression energy spectrum of the nuclear medicine sample, a target peak shape extraction module, a dynamic response spectrum construction module for driving simulation of the target nuclide of the gamma energy spectrum based on the pure target peak spectrum and a preset detector response function to obtain a dynamic full response spectrum of the target nuclide, a dynamic spectrum contribution stripping module for carrying out contribution spectrum accurate stripping on the spectrum contribution of the target nuclide based on the background suppression energy spectrum and the dynamic full response spectrum to obtain a limit residual spectrum of the nuclear medicine sample, an impurity characteristic identification module and a radioactivity purity calculation module. The present application can improve the online detection efficiency of the nuclear medicine radioactivity purity based on energy spectrum analysis.

Description

Technical Field

[0001] This invention relates to the field of nuclear radiation technology, and in particular to an online detection system for radiopharmaceutical radioactivity purity based on energy spectrum analysis. Background Technology

[0002] Existing methods mainly rely on offline energy dispersive spectroscopy measurements and manual analysis, which suffer from low efficiency, poor real-time performance, and susceptibility to environmental background interference. Especially in complex energy dispersive spectral backgrounds, the identification and removal of target nuclides and impurity peaks depend on manual experience, lacking automated and standardized processing procedures. This results in highly subjective and poor repeatability of the detection results, making it difficult to meet the urgent needs of radiopharmaceutical production lines for rapid and high-precision purity monitoring.

[0003] Existing technologies for background subtraction often employ static or empirical models, which cannot adapt to real-time changes in background radiation environments. In peak extraction and impurity identification, they frequently rely on fixed thresholds or simple fitting methods, making it difficult to accurately separate overlapping peaks from weak peak signals. The lack of dynamic simulation of the detector response function and quantitative stripping of spectral contributions results in the inability to completely separate residual background from the target nuclide's contribution, thus affecting the sensitivity of impurity identification and the accuracy of purity calculation. These problems limit the reliability and automation level of online detection of radiopharmaceutical radiopurity. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an online detection system for radiopharmaceutical radiopurity based on energy spectrum analysis to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy analysis, the system comprising:

[0006] The synchronous signal acquisition module is used to acquire the gamma spectrum of radiopharmaceutical samples and the background radiation of the environment;

[0007] A real-time background suppression module is used to perform correlation screening on the gamma spectrum based on the environmental background radiation to obtain the background suppression spectrum of the radiopharmaceutical sample.

[0008] The target peak extraction module is used to extract the net peak of the target nuclide full-energy peak from the background suppression energy spectrum to obtain the pure target peak spectrum of the radiopharmaceutical sample.

[0009] The dynamic response spectrum construction module is used to drive the simulation of the target nuclide in the gamma energy spectrum based on the pure target peak spectrum and the preset detector response function to obtain the dynamic full response spectrum of the target nuclide.

[0010] The dynamic spectral contribution stripping module is used to accurately strip the spectral contribution of the target nuclide based on the background suppression energy spectrum and the dynamic full response spectrum, so as to obtain the limiting residual spectrum of the radiopharmaceutical sample.

[0011] An impurity feature identification module is used to identify the impurity peak positions in the limiting residual spectrum to obtain the impurity feature peaks of the radiopharmaceutical sample.

[0012] The radionuclide purity calculation module is used to calculate the radionuclide purity of the radiopharmaceutical sample based on the impurity characteristic peaks and the pure target peak spectrum.

[0013] Optionally, obtaining the gamma spectrum and background radiation of the radiopharmaceutical sample includes:

[0014] The gamma photons of the radiopharmaceutical sample are converted into electrical pulse signals, and the electrical pulse signals are amplified to obtain the gamma energy spectrum of the radiopharmaceutical sample.

[0015] The ambient background radiation of the radiopharmaceutical sample was collected simultaneously.

[0016] Optionally, the step of correlating and screening the gamma spectrum based on the ambient background radiation to obtain the background suppression spectrum of the radiopharmaceutical sample includes:

[0017] The synchronous anti-coincidence signal event stream of the ambient background radiation is timestamped with the original pulse event stream of the gamma spectrum to obtain the time-domain correlated pulse pair of the radiopharmaceutical sample.

[0018] The gamma spectrum is identified in real time based on the background events of the time-domain correlated pulse pairs to obtain the effective pulse event set of the gamma spectrum;

[0019] The energy spectrum of the effective pulse event set is reconstructed to obtain the background suppression energy spectrum of the radiopharmaceutical sample.

[0020] Optionally, the step of extracting the net peak from the target nuclide's full-energy peak in the background suppression spectrum to obtain the pure target peak spectrum of the radiopharmaceutical sample includes:

[0021] Determine the energy range of the full-energy peak of the target nuclide;

[0022] Linear background fitting is performed on the background suppression energy spectrum and the boundary of the energy range to obtain the continuous spectrum of the background contribution of the full-energy peak of the target nuclide;

[0023] By performing stepwise difference calculations on the energy range and the background contribution continuous spectrum, the pure target peak spectrum of the radiopharmaceutical sample is obtained.

[0024] Optionally, the step of driving the simulation of the target nuclide based on the pure target peak spectrum and a preset detector response function to obtain the dynamic full response spectrum of the target nuclide includes:

[0025] The peak energy of the target nuclide is determined based on the pure target peak spectrum;

[0026] The dynamic full response spectrum of the target nuclide is obtained by performing full-spectrum construction calculations on the pulse energy of the peak energy and the preset detector response function.

[0027] Optionally, the construction of the preset detector response function includes:

[0028] The detector structure for acquiring the radiopharmaceutical sample;

[0029] Monte Carlo particle transport simulation was performed on the particle transport process of the radiopharmaceutical sample based on the detector structure and the material composition of the radiopharmaceutical sample to obtain a set of energy deposition values ​​of the radiopharmaceutical sample.

[0030] The detector response function of the radiopharmaceutical sample is constructed based on the set of energy deposition values.

[0031] Optionally, the step of precisely stripping the contribution spectrum of the target nuclide based on the background suppression energy spectrum and the dynamic full response spectrum to obtain the limiting residual spectrum of the radiopharmaceutical sample includes:

[0032] The intensity of the main full-energy peak region of the target nuclide is calibrated based on the background suppression energy spectrum and the dynamic full-response spectrum to obtain the intensity factor of the target nuclide;

[0033] The total spectral contribution of the target nuclide is obtained by scaling the dynamic full response spectrum and the intensity factor.

[0034] The difference between the background suppression energy spectrum and the total spectrum contribution is compared to obtain the limiting residual spectrum of the radiopharmaceutical sample.

[0035] Optionally, the formula for calculating the limiting residual spectrum is:

[0036]

[0037] in, It is the limiting residual spectrum, This is the background suppression energy spectrum. It is the dynamic full response spectrum, It is the peak area of ​​the background suppression energy spectrum. It is the specific energy of the target nuclide. It is the peak area of ​​the dynamic full response spectrum. It is the energy range of the target nuclide. It is an energy variable. It is an integral variable.

[0038] Optionally, the step of identifying impurity peak positions in the limiting residual spectrum to obtain the impurity characteristic peaks of the radiopharmaceutical sample includes:

[0039] Regional statistical analysis is performed on the limiting residual spectrum to obtain the dynamic identification threshold of the limiting residual spectrum;

[0040] Peak position identification is performed on the limiting residual spectrum and the dynamic identification threshold to obtain the candidate peak region of the radiopharmaceutical sample;

[0041] Peak position is located in the candidate peak region to obtain the impurity characteristic peaks of the radiopharmaceutical sample.

[0042] Optionally, calculating the radionuclide purity of the radiopharmaceutical sample based on the impurity characteristic peaks and the pure target peak spectrum includes:

[0043] The net peak area of ​​the impurity is obtained by integrating the area of ​​the impurity characteristic peak.

[0044] Extract the target net peak area of ​​the target nuclide based on the pure target peak spectrum;

[0045] The total net peak area of ​​the radiopharmaceutical sample is calculated based on the target net peak area and the impurity net peak area.

[0046] The purity of the radiopharmaceutical sample is evaluated based on the total net peak area and the target net peak area to obtain the radionuclide purity of the radiopharmaceutical sample.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] This invention effectively overcomes the interference of environmental radiation fluctuations on energy spectrum analysis by introducing a synchronous signal acquisition and real-time background suppression module. Utilizing timestamp alignment and anti-coincidence discrimination techniques, background events are correlated and subtracted in real time, significantly improving the signal-to-noise ratio of the gamma spectrum. This method avoids the problems of poor adaptability and high residual background in dynamic environments associated with traditional static background subtraction methods, providing a clean data foundation for the subsequent accurate extraction of target nuclide characteristic peaks and enhancing the reliability and stability of radiopharmaceutical purity detection under complex radiation environments.

[0049] This invention achieves highly sensitive identification of the full-energy peaks and impurity characteristics of target nuclides through dynamic response spectrum construction and precise spectral contribution stripping technology. The detector response function is constructed using Monte Carlo simulation, and the contribution of the target nuclide is precisely scaled and subtracted based on the intensity factor to obtain the limiting residual spectrum, effectively separating weak impurity peaks that are difficult to identify using traditional methods. Finally, the purity of the radionuclide is calculated through automatic peak region identification and net peak area integration, improving the automation and quantitative accuracy of detection and meeting the dual requirements of high precision and high efficiency for online quality control. Attached Figure Description

[0050] Figure 1 This is a functional block diagram of an online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy, provided in an embodiment of the present invention. Detailed Implementation

[0051] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0052] This application provides an online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy (EDS). The executing entity of this system includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the system provided in this application: a server, a terminal, etc. In other words, the online detection system for radiopharmaceutical radiopurity based on EDS can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms.

[0053] like Figure 1 The diagram shown is a functional block diagram of an online detection system for radiopharmaceutical radiopurity based on energy spectrum analysis, provided in an embodiment of the present invention.

[0054] The radiopharmaceutical radiopurity online detection system 100 based on energy dispersive spectroscopy (EDS) analysis described in this invention can be installed in an electronic device. Depending on the functions implemented, the system 100 may include a synchronous signal acquisition module 101, a real-time background suppression module 102, a target peak shape extraction module 103, a dynamic response spectrum construction module 104, a dynamic spectrum contribution stripping module 105, an impurity feature identification module 106, and a radiopurity calculation module 107. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0055] In this embodiment, the functions of each module / unit are as follows:

[0056] The synchronous signal acquisition module 101 is used to acquire the gamma spectrum and background radiation of the radiopharmaceutical sample.

[0057] In this embodiment of the invention, obtaining the gamma spectrum and background radiation of the radiopharmaceutical sample includes:

[0058] The gamma photons of the radiopharmaceutical sample are converted into electrical pulse signals, and the electrical pulse signals are amplified to obtain the gamma energy spectrum of the radiopharmaceutical sample.

[0059] The ambient background radiation of the radiopharmaceutical sample was collected simultaneously.

[0060] In detail, in the physical environment, the radiopharmaceutical sample is placed within the measurement area of ​​the radiation detector. The sample spontaneously decays and releases gamma photons. These photons enter the detector crystal and, through physical processes such as the photoelectric effect and Compton scattering, transfer their energy to the crystal material, causing ionization and the formation of charge carriers or the generation of light in the scintillator.

[0061] Furthermore, the electric field inside the detector collects these charges, effectively converting the energy information of the photons into a very small electrical pulse signal. This process achieves a direct, linear conversion of nuclear radiation energy into an electrical signal, providing an accurate and reliable raw data foundation for subsequent energy spectrum analysis and ensuring the fidelity of the energy information.

[0062] Specifically, the raw electrical pulse signal output by the detector is extremely weak, typically in the millivolt range, and cannot be processed directly. Therefore, the signal must pass through electronic components such as preamplifiers and main amplifiers. These components perform amplification, proportionally increasing the pulse amplitude to the volt level, while simultaneously shaping the pulse to make it easily measurable by an analog-to-digital converter.

[0063] Furthermore, amplification and shaping significantly improve the signal-to-noise ratio and measurability of the signal, converting weak pulse signals into standard, stable electrical signals, creating the necessary conditions for subsequent digital processing, and ensuring the accuracy and stability of the measurement.

[0064] In detail, the gamma spectrum is generated by a multichannel analyzer. The multichannel analyzer measures the amplitude of each amplified pulse and assigns it to the corresponding channel address. The resulting spectrum has its horizontal axis representing energy, derived from the pulse amplitude, and its vertical axis representing the count at that energy, i.e., the number of photons detected. The multichannel analyzer achieves the digitization and visualization of energy through amplitude analysis, rapidly generating an intuitive energy spectrum distribution map. This provides a core basis for qualitative identification of nuclides and quantitative calculation of activity, improving detection efficiency.

[0065] Furthermore, in physical environments, such as nuclear medicine production lines, sample detectors are inevitably subject to interference from background radiation, such as cosmic rays and naturally occurring radioactive nuclides.

[0066] Specifically, synchronous acquisition is a crucial step, typically achieved using an auxiliary detector, such as an anti-coincidence detector, that surrounds the main detector. When radiation from the environment, such as high-energy cosmic rays, passes through the anti-coincidence detector and simultaneously strikes the main detector, the two detectors generate signals synchronously.

[0067] The real-time background suppression module 102 is used to perform correlation screening on the gamma spectrum based on the environmental background radiation to obtain the background suppression spectrum of the radiopharmaceutical sample.

[0068] In this embodiment of the invention, the step of correlating and screening the gamma spectrum based on the environmental background radiation to obtain the background suppression spectrum of the radiopharmaceutical sample includes:

[0069] The synchronous anti-coincidence signal event stream of the ambient background radiation is timestamped with the original pulse event stream of the gamma spectrum to obtain the time-domain correlated pulse pair of the radiopharmaceutical sample.

[0070] The gamma spectrum is identified in real time based on the background events of the time-domain correlated pulse pairs to obtain the effective pulse event set of the gamma spectrum;

[0071] The energy spectrum of the effective pulse event set is reconstructed to obtain the background suppression energy spectrum of the radiopharmaceutical sample.

[0072] Specifically, in the physical environment, background radiation, such as cosmic rays and natural radioactivity, is dynamic. Traditional static background subtraction methods, which measure the background first and then subtract it from the sample spectrum, fail in this situation. When radiation from the external environment, such as a high-energy cosmic ray particle, passes simultaneously through both the sentinel detector and the main detector, the two detectors will generate signals synchronously.

[0073] In detail, two event streams are continuously compared with extremely high temporal resolution, such as nanoseconds. When it is determined that the event timestamps in the two event streams are aligned, that is, the time difference is less than a preset matching time window, such as 20 nanoseconds, the two events are identified as originating from the same penetrating background event, and they are paired to generate a temporally correlated pulse pair.

[0074] Furthermore, each raw pulse event is provided with a real-time label indicating whether it is background radiation. This process is performed event by event, ensuring accurate capture even if background radiation levels fluctuate in real time. This effectively overcomes the interference of environmental radiation fluctuations on energy spectrum analysis and provides accurate input data for subsequent real-time identification.

[0075] In detail, the raw pulse event stream acquired by the main detector is a mixture, containing both valuable signals from the radiopharmaceutical sample and a large amount of background from the environment. Temporally correlated pulse pairs are used as filters. Each event in the raw pulse event stream is adjudicated in real time: when a pulse event is detected, the list of temporally correlated pulse pairs is immediately queried.

[0076] Specifically, if a negative match signal is found within the matching time window, the event is classified as a background event and immediately discarded. Conversely, if another pulse event arrives and no matching negative match signal is found after querying, it is classified as a valid event. This valid event is allowed to pass and stored in the valid pulse event set.

[0077] Furthermore, the screening process is real-time, occurring synchronously with data acquisition, ensuring that only clean events are accumulated. This improves the signal-to-noise ratio of the gamma spectrum because a large amount of background, especially the Compton continuum, is removed in real time, making subsequent analyses, such as the identification of weak impurity peaks, possible.

[0078] In detail, energy spectrum reconstruction is the final step of the real-time background suppression module, which reassembles the processed, discrete event data stream into a statistically meaningful energy spectrum. The original energy spectrum is the statistical distribution of the pulse amplitudes of all events in the original pulse event stream.

[0079] Specifically, energy spectrum reconstruction is a selective reconstruction process. Its input is a set of effective impulse events, which is a dataset that has been filtered in real time and theoretically has eliminated all background events that are synchronous with the anti-coincidence detector.

[0080] Furthermore, the multichannel analyzer processes each event in the set of valid pulse events one by one. For each event, the multichannel analyzer reads its pulse amplitude and converts it into a channel address number according to the energy scale relationship. The multichannel analyzer increments the count value of the channel address number by one. This action is repeated millions of times within a set measurement time.

[0081] In detail, the pulse amplitude distributions of all valid events are accumulated to form a background-suppressed energy spectrum. This spectrum has a significantly improved signal-to-noise ratio compared to the original spectrum. For example, the high-energy Compton continuum caused by cosmic rays will be greatly suppressed in the background-suppressed energy spectrum.

[0082] Specifically, it provides an extremely pure data foundation for subsequent target peak extraction and impurity feature identification, enabling the identification of weak impurity peaks that are submerged by the background in the original energy spectrum, thereby improving the reliability and stability of radiopharmaceutical purity detection.

[0083] The target peak extraction module 103 is used to extract the net peak of the target nuclide full-energy peak from the background suppression energy spectrum to obtain the pure target peak spectrum of the radiopharmaceutical sample.

[0084] In this embodiment of the invention, the step of extracting the net peak of the target nuclide from the background suppression energy spectrum to obtain the pure target peak spectrum of the radiopharmaceutical sample includes:

[0085] Determine the energy range of the full-energy peak of the target nuclide;

[0086] Linear background fitting is performed on the background suppression energy spectrum and the boundary of the energy range to obtain the continuous spectrum of the background contribution of the full-energy peak of the target nuclide;

[0087] By performing stepwise difference calculations on the energy range and the background contribution continuous spectrum, the pure target peak spectrum of the radiopharmaceutical sample is obtained.

[0088] Specifically, the background suppression spectrum has eliminated background events, but it is still a complex distribution, containing the full-energy peak of the target nuclide—the peak formed when the detector absorbs all the energy of the gamma photon—as well as the continuous spectrum formed by Compton scattering of the same photon in the detector or sample. To accurately calculate the area of ​​the full-energy peak, the range of this peak must first be defined.

[0089] In detail, the peak apex and maximum count value are found near a preset location, and then an energy range is defined by expanding outwards to both sides based on the detector's energy resolution. For example, for a peak of 140.5 keV, the energy range might be determined as [135 keV, 145 keV]. This parameter energy range is crucial; it not only defines the range for subsequent calculations, but its boundaries also serve as input for the next step of linear background fitting, directly affecting the accuracy of background subtraction.

[0090] Specifically, the energy range is intelligently defined based on the physical characteristics of the detector to ensure the scientific validity and consistency of the analysis range. This means that the target peak signal can be fully covered while effectively eliminating interference from irrelevant regions, thus ensuring the accuracy and repeatability of subsequent quantitative calculations.

[0091] Furthermore, after determining the energy range, linear background fitting is performed. Even for a single-energy gamma source, its energy spectrum will not consist of just a single spike. Due to the Compton scattering effect, photons interact with the detector material or the sample itself, depositing only a portion of the energy; the full-energy peak is always superimposed on a smooth continuous spectrum. To obtain the net peak area, this continuous spectrum must be subtracted. Linear background fitting is a computationally fastest and most suitable approximation method.

[0092] In detail, the count values ​​corresponding to the two boundary energy points are read from the background suppression energy spectrum using the energy interval boundaries. A linear function is constructed between these two points. Geometrically, this is equivalent to drawing a straight line between the two legs of the full-energy peak; this line is considered the continuous spectrum of the background contribution below the peak. This provides the data basis for the next step of point-by-point subtraction. Although more complex nonlinear fitting methods exist, linear fitting is computationally highly efficient, perfectly meeting the requirements of rapid processing in online detection.

[0093] Specifically, stepwise interpolation is the process of extracting the net signal from the total signal, representing a linear estimate of the continuous background within that interval. It is the standard background subtraction operation in energy spectrum analysis. The stepwise aspect means performing the operation address by address.

[0094] Furthermore, starting from the beginning of the energy range and ending at the end, a subtraction operation is performed for each channel address. Through precise subtraction address by address, the Compton scattering background beneath the peak can be completely eliminated, ultimately yielding a pure peak signal of the target nuclide. This directly eliminates the interference of the background on the calculation of the net peak area, providing the most accurate and crucial data foundation for subsequent activity calculations and purity analysis, significantly improving the accuracy of quantitative results.

[0095] In detail, the portion of the total energy spectrum belonging to the Compton scattering plateau is subtracted, leaving only the counts contributed by the photoelectric effect. The final output parameter of this operation is the pure target peak spectrum. In this pure spectrum, the continuous background below the peak has been removed, and the baseline returns to near zero. This spectrum is the cornerstone of subsequent analyses and will be used in the dynamic response spectrum construction module and the radioactivity purity calculation module.

[0096] The dynamic response spectrum construction module 104 is used to drive the simulation of the target nuclide in the gamma energy spectrum based on the pure target peak spectrum and the preset detector response function, so as to obtain the dynamic full response spectrum of the target nuclide.

[0097] In this embodiment of the invention, the step of driving the target nuclide of the gamma energy spectrum based on the pure target peak spectrum and a preset detector response function to obtain the dynamic full response spectrum of the target nuclide includes:

[0098] The peak energy of the target nuclide is determined based on the pure target peak spectrum;

[0099] The dynamic full response spectrum of the target nuclide is obtained by performing full-spectrum construction calculations on the pulse energy of the peak energy and the preset detector response function.

[0100] In detail, the pure target peak spectrum is the net peak after double background subtraction: the environmental background and the Compton continuous background. In the physical environment of nuclear radiation measurements, due to detector or amplifier temperature drift or high count rate effects, the peak positions of the energy spectrum may shift slightly, causing the measured peak energy to be not exactly equal to its theoretical value. Ensuring the purity of the input data and considering the real-time state of the detector provides a true and reliable input for subsequent high-precision simulations, avoiding errors caused by equipment drift at the source.

[0101] Specifically, the most accurate peak energy is calculated from the current pure target peak spectrum. This high-precision value is the first key output parameter of this module, ensuring that subsequent simulation calculations are driven by the instrument's real-time status, laying the foundation for constructing a dynamic full-response spectrum that perfectly matches the current measurement data.

[0102] In detail, real-time calculation of peak energy enables dynamic calibration of the detector state, ensuring precise alignment of the simulated spectrum with the measured spectrum on the energy axis. This improves the accuracy of subsequent spectrum matching and stripping, overcoming the deviation caused by using fixed theoretical values ​​in traditional methods.

[0103] Furthermore, a pre-defined detector response function is invoked. This function is pre-constructed through a complex Monte Carlo particle transport simulation. This simulation considers all the physical details of the detector, such as its construction, crystal size, shape, shell, collimator, and material composition (germanium crystal, aluminum shell, copper cold finger, etc.), calculating all interactions of gamma photons, photoelectric effect, Compton scattering, electron pair generation, and energy deposition values.

[0104] Specifically, the detector response function constructed through Monte Carlo simulation highly replicates the physical response process of the real detector, including all features such as the full-energy peak, Compton plateau, and escape peak, providing the possibility for accurately stripping the full spectrum contribution of the target nuclide.

[0105] In detail, the peak energy, or pulse energy, is used to drive the response function, causing it to output a complete and idealized dynamic total response spectrum. This dynamic total response spectrum is the standard fingerprint of the target nuclide, depicting the shape and relative proportion of all features of the nuclide, including its full-energy peaks, Compton continuum, and escape peaks.

[0106] Specifically, the response function is driven by the peak energy measured in real time, generating a dynamic full-response spectrum that perfectly matches the current measurement conditions. This method can not only simulate full-energy peaks but also accurately reproduce the shape and intensity of the Compton scattering background, which is a key innovation for achieving subsequent high-precision full-spectrum stripping and effective identification of weak impurity peaks.

[0107] In this embodiment of the invention, the construction of the preset detector response function includes:

[0108] The detector structure for acquiring the radiopharmaceutical sample;

[0109] Monte Carlo particle transport simulation was performed on the particle transport process of the radiopharmaceutical sample based on the detector structure and the material composition of the radiopharmaceutical sample to obtain a set of energy deposition values ​​of the radiopharmaceutical sample.

[0110] The detector response function of the radiopharmaceutical sample is constructed based on the set of energy deposition values.

[0111] In detail, in the physical environment of nuclear radiation measurement, each detector is unique. Even minute manufacturing differences, such as crystal size and dead-layer thickness, can lead to variations in its response characteristics, i.e., the shape of its energy spectrum. This provides a theoretical basis for subsequent personalized, high-precision Monte Carlo simulations, ensuring that the constructed detector response function accurately reflects the response characteristics of a specific device.

[0112] Specifically, the parametric detector fabrication is a complex geometry and materials database that defines in detail the size, shape, and relative position of every component in the simulation environment, from the detector crystal itself to its aluminum shell, internal support structure, and external lead shield and sample holder.

[0113] In detail, any omission or error in the detector construction parameters, such as not accurately measuring the distance from the sample vial to the detector end window, will lead to distortion in the subsequent Monte Carlo simulation, ultimately causing the detector response function to deviate from reality, and thus affecting the accuracy of dynamic spectral contribution stripping.

[0114] Furthermore, after obtaining precise detector structure and material composition, a Monte Carlo particle transport simulation was performed. This simulation was conducted on a dedicated simulation platform, where a model perfectly replicating the physical environment was first constructed in virtual space based on the input parameters. The simulation then commenced: a massive number of gamma photons were emitted from the virtual location of the ribonucleoside sample.

[0115] In detail, the simulation program uses the Monte Carlo algorithm to calculate the particle transport process for each photon and its generated secondary particles, such as photons and electrons after Compton scattering. This includes calculating the probability, direction, and energy loss of every interaction between the particles and the detector structure and materials, including the photoelectric effect, Compton scattering, electron pair formation, etc.

[0116] Specifically, the simulation program precisely tracks particles until they escape the simulation region or their energy is depleted. For each initial gamma photon in the simulation, the total energy deposited within the detector's active volume is accumulated, and this total energy value is recorded as an event. This set of hundreds of millions of energy values ​​is the raw fingerprint of the detector's response to photons of that energy.

[0117] Furthermore, the energy deposition value set is a list containing hundreds of millions of discrete energy values, representing all energy events that the detector may record. This energy deposition value set is then histogramized. An energy grid is defined, and each energy value in the set is iterated through, assigned to its corresponding energy address, and its count is incremented by one.

[0118] For example, when simulating 140.5 keV photons, the energy deposition values ​​of many events will fall around 140.5 keV, many events will fall between 0 and about 50 keV, and others will fall at specific backscattering peak positions. The resulting histogram is the detector response function.

[0119] The dynamic spectral contribution stripping module 105 is used to accurately strip the spectral contribution of the target nuclide based on the background suppression energy spectrum and the dynamic full response spectrum, so as to obtain the limiting residual spectrum of the radiopharmaceutical sample.

[0120] In this embodiment of the invention, the precise stripping of the contribution spectrum of the target nuclide based on the background suppression energy spectrum and the dynamic full response spectrum to obtain the limiting residual spectrum of the radiopharmaceutical sample includes:

[0121] The intensity of the main full-energy peak region of the target nuclide is calibrated based on the background suppression energy spectrum and the dynamic full-response spectrum to obtain the intensity factor of the target nuclide;

[0122] The total spectral contribution of the target nuclide is obtained by scaling the dynamic full response spectrum and the intensity factor.

[0123] The difference between the background suppression energy spectrum and the total spectrum contribution is compared to obtain the limiting residual spectrum of the radiopharmaceutical sample.

[0124] In detail, the intensity of the background suppression spectrum is determined by the sample's true activity and measurement duration, while the dynamic full response spectrum is constructed based on Monte Carlo simulations, and its absolute intensity is arbitrary, representing only the shape. Before stripping, the intensity of the simulated spectrum must be calibrated to match the true spectrum. The intensity calibration process is for calculating this calibration coefficient.

[0125] Specifically, the main full-energy peak region of the target nuclide is identified, and the net peak area in the background suppression spectrum and the net peak area in the dynamic full-response spectrum are calculated by integration. A division operation is then performed to obtain a dimensionless intensity factor. This intensity factor precisely quantifies how many times the target nuclide signal in the real spectrum is greater than that in the simulated spectrum; it is a prerequisite for subsequent amplitude scaling and precise stripping.

[0126] Furthermore, after obtaining the intensity factor, amplitude scaling is immediately performed. This single calibration value of the intensity factor is applied to the entire dynamic full response spectrum. The target nuclide's contribution to the energy spectrum is full-spectrum. Traditional methods struggle to remove this Compton background because it overlaps with impurity peaks. The amplitude scaling operation is a channel-by-channel multiplication: the count value of each energy channel in the dynamic full response spectrum is multiplied by the same intensity factor.

[0127] In detail, all features in the ideal spectrum, such as the full-energy peaks, Compton plateau, and backscattering peaks, are proportionally stretched to an intensity completely consistent with the contribution of the target nuclide in the background suppression spectrum. The output of this action is the total spectral contribution. This total spectral contribution plot is the best estimate of the contribution of the pure target nuclide in the background suppression spectrum, including all relevant physical effects.

[0128] Specifically, the difference comparison is the final action performed by the dynamic spectral contribution stripping module, which essentially involves address-by-address subtraction to achieve precise contribution spectrum stripping. It possesses both the background suppression energy spectrum and the total spectral contribution. For each address of both energy spectra, the subtraction operation is performed one by one.

[0129] In detail, because the total spectral contribution not only perfectly matches the full-energy peak of the target nuclide but also its complete Compton continuum, this subtraction operation will strip away all the contribution of the target nuclide from the background suppression spectrum. Its output parameter is the limiting residual spectrum, which is an almost flat spectrum that fluctuates only around zero. Unless impurities are present in the original sample, the characteristic peaks of the impurities will be clearly highlighted in the limiting residual spectrum.

[0130] In this embodiment of the invention, the formula for calculating the limiting residual spectrum is:

[0131]

[0132] in, It is the limiting residual spectrum, This is the background suppression energy spectrum. It is the dynamic full response spectrum, It is the peak area of ​​the background suppression energy spectrum. It is the specific energy of the target nuclide. It is the peak area of ​​the dynamic full response spectrum. It is the energy range of the target nuclide. It is an energy variable. It is an integral variable.

[0133] In detail, The limiting residual spectrum represents the signal remaining after removing all known target nuclide contributions from the actual measurement data. Theoretically, if the radiopharmaceutical sample is pure, this spectrum will only contain statistical noise; if impurities are present, the characteristic peaks of the impurities will be clearly visible in the spectrum.

[0134] Specifically, This refers to the background suppression energy spectrum, which represents the actual measurement data. It is the energy spectrum acquired on the detector and processed by the real-time background suppression module.

[0135] Furthermore, The dynamic full response spectrum is a pre-constructed, normalized energy spectrum generated through Monte Carlo particle transport simulation. It represents the complete fingerprint that a pure nuclide should produce in this specific detector, including all features such as the full-energy peak, Compton plateau, and escape peak.

[0136] Specifically, It is the peak area of ​​the background suppression spectrum, which is related to the background suppression spectrum. ,exist Integrating within the energy range yields the net peak area of ​​the background suppression spectrum.

[0137] In detail, It is the specific energy of the target nuclide, the peak energy actually measured from the pure target peak spectrum. (Using...) To drive generation This ensures that the simulated spectrum and the real spectrum are perfectly aligned on the energy axis.

[0138] Furthermore, It is the peak area of ​​the dynamic full response spectrum. In the same Integrating over the interval yields the net peak area of ​​the dynamic full response spectrum.

[0139] In detail, It is the energy range of the target nuclide, which is an energy range, for example... The area surrounding the peak.

[0140] Specifically, It is an energy variable, representing It is a matter of energy A changing function.

[0141] Furthermore, It is an integration variable, a dummy variable used only for integration calculations, used in... The peak area is obtained by summing the values ​​within the interval.

[0142] In detail, It is the intensity factor of the target nuclide, which calculates the ratio of the actual peak area to the ideal peak area. It serves as a bridge connecting the actual measurement and the ideal model, quantifying how many times the actual signal is greater than the ideal template.

[0143] Furthermore, This is the total spectral contribution of the target nuclide, calculated by multiplying the intensity factor by the entire dynamic full response spectrum. The total spectral contribution has the same shape as the dynamic full response spectrum, but its intensity perfectly matches the target nuclide contribution in the background suppression spectrum.

[0144] The impurity feature identification module 106 is used to identify the impurity peak positions in the limiting residual spectrum to obtain the impurity feature peaks of the radiopharmaceutical sample.

[0145] In this embodiment of the invention, the step of identifying impurity peak positions in the limiting residual spectrum to obtain the impurity characteristic peaks of the radiopharmaceutical sample includes:

[0146] Regional statistical analysis is performed on the limiting residual spectrum to obtain the dynamic identification threshold of the limiting residual spectrum;

[0147] Peak position identification is performed on the limiting residual spectrum and the dynamic identification threshold to obtain the candidate peak region of the radiopharmaceutical sample;

[0148] Peak position is located in the candidate peak region to obtain the impurity characteristic peaks of the radiopharmaceutical sample.

[0149] Specifically, the limiting residual spectrum is calculated by the dynamic spectral contribution stripping module. Ideally, this spectrum should fluctuate around the zero baseline, with the fluctuation amplitude representing the statistical noise. Due to differences in the original counts, such as the Compton plateau count in the main peak region being much higher than in the high-energy region, the statistical noise in the limiting residual spectrum is not uniform across different energy regions.

[0150] In detail, the local standard deviation of the limiting residual spectrum is calculated region by region using a sliding window technique to obtain the dynamic identification threshold. This output dynamic threshold automatically adjusts with energy levels, raising the threshold in regions with high noise levels and lowering it in regions with low noise levels.

[0151] Furthermore, the limiting residual spectrum is used as the signal, and the dynamic identification threshold is used as the baseline. The limiting residual spectrum is traversed address by address, and a comparison is performed at each address. In a physical environment, a real impurity peak will manifest as the signal value limiting residual spectrum of multiple consecutive addresses continuously exceeding the dynamic identification threshold, while statistical noise may only be a spike at a single address.

[0152] In detail, peak identification involves finding regions that continuously exceed a threshold. Once such regions are identified, they are marked and output as candidate peak regions. This candidate peak region is an energy range that may contain impurity peaks. Eliminating most statistical noise fluctuations and narrowing the calculation range for subsequent precise positioning is an important step in achieving automated and standardized processing.

[0153] Specifically, the process focuses on multiple candidate peak regions and performs peak location analysis. Typically, a Gaussian function is used to fit the limiting residual spectrum to the data points within that candidate region. The fitting process calculates the optimal parameters for the peak shape, the most crucial of which is the peak's center position, i.e., the precise energy of the impurity characteristic peak. Another important output of this process is the fitted peak area, i.e., the net impurity peak area.

[0154] The radionuclide purity calculation module 107 is used to calculate the radionuclide purity of the radiopharmaceutical sample based on the impurity characteristic peaks and the pure target peak spectrum.

[0155] In this embodiment of the invention, the calculation of the radionuclide purity of the radiopharmaceutical sample based on the impurity characteristic peaks and the pure target peak spectrum includes:

[0156] The net peak area of ​​the impurity is obtained by integrating the area of ​​the impurity characteristic peak.

[0157] Extract the target net peak area of ​​the target nuclide based on the pure target peak spectrum;

[0158] The total net peak area of ​​the radiopharmaceutical sample is calculated based on the target net peak area and the impurity net peak area.

[0159] The purity of the radiopharmaceutical sample is evaluated based on the total net peak area and the target net peak area to obtain the radionuclide purity of the radiopharmaceutical sample.

[0160] In detail, within the physical environment of nuclear radiation measurements, the list of impurity characteristic peaks is traversed. For each peak in the list, an integration operation is performed on its corresponding candidate peak region data in the limiting residual spectrum, which is typically a byproduct of Gaussian fitting. If multiple impurity peaks are found, the areas of all impurity peaks are summed. Ultimately, this operation outputs the net impurity peak area, a parameter representing the total activity contribution of all radioactive impurities in the sample.

[0161] Specifically, the pure target peak spectrum is retrieved. This spectrum is the product of linear background fitting and stepwise interpolation, containing only the net peak signal of the target nuclide. An area integration operation is performed on this pure target peak spectrum to obtain the target net peak area.

[0162] Furthermore, in energy dispersive spectroscopy (EDS), under ideal calibration, activity is proportional to net peak area. The target net peak area represents the activity of the target nuclide, and the impurity net peak area represents the total activity of all impurity nuclides. This output parameter, total net peak area, represents the sum of all radioactive signals from the radiopharmaceutical sample measured by the detector in this online detection, and it will serve as the denominator for the final purity assessment.

[0163] In detail, based on the standard definition of radionuclide purity in nuclear radiation measurements, a division operation is performed. The target net peak area and total net peak area are retrieved. The purity assessment process involves calculating the total net peak area and the target net peak area. The output of this process is the radionuclide purity, expressed as a percentage, which precisely quantifies the purity of the radiopharmaceutical sample.

[0164] In the several embodiments provided by this invention, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0165] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0166] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0167] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0168] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy (EDS), characterized in that, The system includes: The synchronous signal acquisition module is used to acquire the gamma spectrum of radiopharmaceutical samples and the background radiation of the environment; A real-time background suppression module is used to perform correlation screening on the gamma spectrum based on the environmental background radiation to obtain the background suppression spectrum of the radiopharmaceutical sample, including: The synchronous anti-coincidence signal event stream of the ambient background radiation is timestamped with the original pulse event stream of the gamma spectrum to obtain the time-domain correlated pulse pair of the radiopharmaceutical sample. The gamma spectrum is identified in real time based on the background events of the time-domain correlated pulse pairs to obtain the effective pulse event set of the gamma spectrum; The energy spectrum of the pulse amplitude of the effective pulse event set is reconstructed to obtain the background suppression energy spectrum of the radiopharmaceutical sample; The target peak extraction module is used to extract the net peak of the target nuclide full-energy peak from the background suppression energy spectrum to obtain the pure target peak spectrum of the radiopharmaceutical sample. The dynamic response spectrum construction module includes: The peak energy of the target nuclide is determined based on the pure target peak spectrum; Based on the peak energy driving the preset detector response function, a dynamic full response spectrum that perfectly matches the conditions is generated. The peak energy is used to drive the generation of the dynamic full response spectrum, ensuring that the simulated spectrum and the real spectrum are aligned on the energy axis; A dynamic spectral contribution stripping module is used to accurately strip the spectral contribution of the target nuclide based on the background suppression energy spectrum and the dynamic full response spectrum, to obtain the limiting residual spectrum of the radiopharmaceutical sample, including: By performing integral calculations on the main full-energy peak region of the target nuclide, the net peak area of ​​the main full-energy peak region in the background suppression energy spectrum and the net peak area in the dynamic full response spectrum are obtained. The ratio of the net peak area in the background suppression spectrum to the net peak area in the dynamic full response spectrum is used as the intensity factor of the target nuclide. Multiply the count value of the energy address in the dynamic full response spectrum by the intensity factor to obtain the total spectral contribution of the target nuclide; The difference between the background suppression energy spectrum and the total spectrum contribution is compared to obtain the limiting residual spectrum of the radiopharmaceutical sample; An impurity feature identification module is used to identify the impurity peak positions in the limiting residual spectrum to obtain the impurity feature peaks of the radiopharmaceutical sample. The radionuclide purity calculation module is used to calculate the radionuclide purity of the radiopharmaceutical sample based on the impurity characteristic peaks and the pure target peak spectrum.

2. The online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy as described in claim 1, characterized in that, The acquisition of the gamma spectrum and background radiation of the radiopharmaceutical sample includes: The gamma photons of the radiopharmaceutical sample are converted into electrical pulse signals, and the electrical pulse signals are amplified to obtain the gamma energy spectrum of the radiopharmaceutical sample. The ambient background radiation of the radiopharmaceutical sample was collected simultaneously.

3. The online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy as described in claim 1, characterized in that, The step of extracting the net peak of the target nuclide from the background suppression spectrum to obtain the pure target peak spectrum of the radiopharmaceutical sample includes: Determine the energy range of the full-energy peak of the target nuclide; Linear background fitting is performed on the background suppression energy spectrum and the boundary of the energy range to obtain the continuous spectrum of the background contribution of the full-energy peak of the target nuclide; By performing stepwise difference calculations on the energy range and the background contribution continuous spectrum, the pure target peak spectrum of the radiopharmaceutical sample is obtained.

4. The online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy as described in claim 1, characterized in that, The construction of the preset detector response function includes: The detector structure for acquiring the radiopharmaceutical sample; Monte Carlo particle transport simulation was performed on the secondary particle transport process generated by gamma photons based on the detector structure and the material composition of the radiopharmaceutical sample to obtain a set of energy deposition values ​​for the radiopharmaceutical sample. The detector response function of the radiopharmaceutical sample is constructed based on the set of energy deposition values.

5. The online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy as described in claim 1, characterized in that, The formula for calculating the limiting residual spectrum is: in, It is the limiting residual spectrum, This is the background suppression energy spectrum. It is the dynamic full response spectrum, It is the peak area of ​​the background suppression energy spectrum. It is the peak energy. It is the peak area of ​​the dynamic full response spectrum. It is the energy range of the target nuclide. It is an energy variable. It is an integral variable.

6. The online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy as described in claim 1, characterized in that, The process of identifying impurity peak positions in the limiting residual spectrum to obtain the characteristic impurity peaks of the radiopharmaceutical sample includes: Regional statistical analysis is performed on the limiting residual spectrum to obtain the dynamic identification threshold of the limiting residual spectrum; Peak position identification is performed on the limiting residual spectrum and the dynamic identification threshold to obtain the candidate peak region of the radiopharmaceutical sample; Peak position is located in the candidate peak region to obtain the impurity characteristic peaks of the radiopharmaceutical sample.

7. The online detection system for radiopharmaceutical radiopurity based on energy dispersive spectroscopy as described in claim 1, characterized in that, The calculation of the radionuclide purity of the radiopharmaceutical sample based on the characteristic peaks of the impurities and the pure target peak spectrum includes: The net peak area of ​​the impurity is obtained by integrating the area of ​​the impurity characteristic peak. Extract the target net peak area of ​​the target nuclide based on the pure target peak spectrum; The total net peak area of ​​the radiopharmaceutical sample is calculated based on the target net peak area and the impurity net peak area. The purity of the radiopharmaceutical sample is evaluated based on the total net peak area and the target net peak area to obtain the radionuclide purity of the radiopharmaceutical sample.

Images