Scintillator excitation sensitizer system applied to malignant tumor 125 iodine particle therapy

Abstract

The invention relates to the technical field of malignant tumor treatment equipment, and discloses a scintillator excitation sensitizer system applied to malignant tumor 125 iodine particle treatment. Comprising a three-dimensional modeling module used for obtaining imaging data of tumor tissues and constructing a three-dimensional model; the radiation prediction module is used for calculating radiation dose rate space-time distribution around each radiator based on the 125 iodine particle cylinder radiator model, and forming an overall radiation dose rate space-time distribution prediction model through superposition; the positioning calculation module is used for determining the optimal placement position of the scintillator in the tumor tissue in combination with the three-dimensional model and the radiation dose rate space-time distribution prediction model; and the guiding module is used for generating a scintillator implantation path and guiding the puncture device to accurately implant the scintillator into the optimal implantation position. The optimal positioning of the scintillator in the tumor is realized, and the key problems of low-energy gamma ray energy conversion efficiency and safety are solved.

Description

Technical Field

[0001] This application relates to the field of malignant tumor treatment equipment technology, and in particular to a system for a scintillator excitation sensitizer used in 125 iodine particle therapy for malignant tumors. Background Technology

[0002] Iodine-125 (I-125) particle therapy, as an important method of brachytherapy, delivers continuous radiation to tumor tissue through the implantation of radioactive particles, offering advantages such as minimal invasiveness and definite efficacy. However, for some patients with tumors that are inoperable or require enhanced instantaneous energy pulses, relying solely on low-dose continuous radiation is insufficient to meet their treatment needs.

[0003] In existing technologies, the method of converting γ-rays emitted by iodine-125 into visible / ultraviolet light to activate photosensitizers and generate reactive oxygen species (ROS) through scintillators can achieve pulsed energy output. However, it faces technical challenges in scintillator material selection and precise positioning: low-energy γ-rays require elements with high atomic numbers to improve interception efficiency, the emission spectrum needs to match the photosensitizer, and the low dose rate problem requires the accumulation of energy through long-afterglow materials; at the same time, there is no mature solution yet for how to combine the three-dimensional structure of the tumor and the spatiotemporal distribution of radiation to determine the optimal position of the scintillator in order to balance the ROS concentration and the risk of heavy metal toxicity.

[0004] Therefore, there is an urgent need to design a technical solution to solve at least one of the above-mentioned technical problems. Summary of the Invention

[0005] This application provides a system for scintillator-activated sensitizers used in 125I particle therapy for malignant tumors. It aims to solve the problem in the prior art that the γ-rays released by 125I are converted into visible / ultraviolet light by a scintillator to activate the photosensitizer to generate reactive oxygen species (ROS). This method can achieve pulsed energy output, but faces technical challenges in scintillator material selection and precise positioning.

[0006] In a first aspect, this application provides a system for a scintillator excitation sensitizer used in iodine-125 particle therapy for malignant tumors, comprising:

[0007] The 3D modeling module is used to acquire imaging data of tumor tissue and construct a 3D model.

[0008] The radiation prediction module is used to calculate the spatiotemporal distribution of radiation dose rate around each radiator based on a cylindrical radiator model of iodine-125 particles, and to form an overall spatiotemporal distribution prediction model of radiation dose rate by superposition.

[0009] The positioning calculation module is used to combine the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model to determine the optimal placement position of the scintillator in the tumor tissue.

[0010] The guiding module is used to generate the scintillator insertion path and guide the puncture device to accurately implant the scintillator into the optimal insertion position.

[0011] In some embodiments, the calculation of the spatiotemporal distribution of radiation dose rate around each radiator based on the cylindrical radiator model of 125 iodine particles includes: by equating each 125 iodine particle to a cylindrical radiation source, and based on the radiation dose rate calculation formula, combined with the activity of the particles, energy decay law and tissue scattering characteristics, calculating the radiation dose rate of a single radiator at different time points and different spatial locations, and generating the overall spatiotemporal distribution of radiation dose rate within the tumor tissue through a spatial superposition algorithm.

[0012] In some embodiments, the step of forming a spatiotemporal distribution prediction model of overall radiation dose rate by superposition includes: accumulating the radiation dose rates calculated by each of the cylindrical radiation sources at different time points and different spatial locations according to three-dimensional spatial coordinates, and superimposing and integrating the radiation dose rates of multiple radiation sources at the same spatial location in terms of time and spatial dimensions to form a set of radiation dose rate data containing each spatial coordinate point within the tumor tissue at different time nodes, which serves as the spatiotemporal distribution prediction model of overall radiation dose rate.

[0013] In some embodiments, determining the optimal placement location of the scintillator within tumor tissue by combining the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model includes: extracting the boundary range, internal density distribution, and key anatomical structure locations of the tumor tissue based on the three-dimensional model; selecting regions in the radiation dose rate spatiotemporal distribution prediction model where the radiation dose rate is consistently higher than the minimum excitation threshold within a preset time threshold as candidate locations; importing the candidate locations into a calculation model that includes parameters such as the concentration distribution of photosensitizers within the tumor tissue, the energy accumulation efficiency parameters of the scintillator material, and heavy metal toxicity risk assessment parameters; iteratively calculating the candidate locations using a preset optimization algorithm; and outputting the three-dimensional coordinates of the optimal placement location with the goal of maximizing reactive oxygen species generation efficiency and ensuring that the heavy metal accumulation of the scintillator material is below the safety standard.

[0014] In some embodiments, generating a scintillator placement path and guiding the puncture device to precisely implant the scintillator at the optimal placement position includes: based on the spatial positional relationship between the tumor tissue and surrounding anatomical structures such as blood vessels and nerves in the three-dimensional model, and with the principle of avoiding important tissues, generating a collision-free or low-damage path from the puncture point on the skin surface to the optimal placement position through a path planning algorithm; during the implantation of the puncture device, acquiring the position coordinates of the puncture needle in real time and spatially registering them with the three-dimensional model, displaying the deviation value between the current position and the target position of the puncture needle through a visualization interface, and dynamically adjusting the angle and depth of the puncture needle according to the deviation value until the scintillator is precisely implanted at the three-dimensional coordinates of the optimal placement position.

[0015] In some embodiments, the scintillator comprises a high atomic number element material whose emission spectrum matches the absorption peak of the photosensitizer, and the scintillator employs a long afterglow material to accumulate the gamma-ray energy released by 125 iodine particles.

[0016] In some embodiments, the imaging data includes CT, MRI, or PET-CT images. The tumor boundary and internal structure are extracted using an image segmentation algorithm to construct a three-dimensional model containing the tumor's three-dimensional coordinates, volume, and tissue density information.

[0017] Secondly, this application provides a method for using a scintillator excitation sensitizer in iodine-125 particle therapy for malignant tumors, employing a system for using a scintillator excitation sensitizer in iodine-125 particle therapy for malignant tumors provided in any embodiment of this application; the method includes:

[0018] Acquire imaging data of tumor tissue and construct a three-dimensional model;

[0019] Based on the cylindrical radiator model of iodine-125 particles, the spatiotemporal distribution of radiation dose rate around each radiator is calculated, and the overall spatiotemporal distribution prediction model of radiation dose rate is formed by superposition.

[0020] By combining the three-dimensional model with the spatiotemporal distribution prediction model of radiation dose rate, the optimal placement location of the scintillator within the tumor tissue is determined.

[0021] Generate a scintillator insertion path and guide the puncture device to precisely implant the scintillator into the optimal insertion position.

[0022] Thirdly, this application provides a computer device, the computer device including a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and, when executing the computer program, implement the method provided in any embodiment of this application.

[0023] Fourthly, this application provides a computer-readable storage medium storing a computer program, wherein the computer-readable instructions, when executed by a processor, cause one or more processors to perform the method provided in any embodiment of this application.

[0024] This invention acquires three-dimensional images of tumors using enhanced CT scans of the liver, extracts tumor boundaries and intrahepatic vascular distribution using segmentation algorithms, and constructs a three-dimensional model including tumor density and volume. Based on the number, location, and activity of implanted 125I particles, the spatiotemporal distribution of radiation dose rate around each particle is calculated, and these are superimposed to generate a predictive model of the overall radiation distribution of the liver tumor region, marking high-dose-rate areas at different time points. The absorption spectral parameters of commonly used photosensitizers for liver cancer (such as dihydroporphyrin) and the energy accumulation curve of the scintillator's long afterglow material are input. Using high-dose-rate points in the tumor center as candidates, an optimization algorithm calculates three optimal implantation positions, ensuring a radiation dose rate ≥10 mGy / h within 5 mm of each scintillator and a distance ≥2 mm from the portal vein. Using an ultrasound-guided puncture device, the scintillators are implanted sequentially into predetermined positions along a planned path. The depth and angle of the puncture needle are displayed in real-time during the procedure, ensuring an implantation error ≤0.5 mm. The implanted miniature radiation sensor collects dose rate and luminous intensity data around the scintillator every 10 minutes. If the luminous intensity at a certain location is 15% lower than the expected value, the photosensitizer dosage for subsequent treatment in that area is automatically adjusted.

[0025] This invention achieves optimal positioning of the scintillator within the tumor through precise coupling of 3D modeling and radiation prediction, solving key issues of energy conversion efficiency and safety of low-energy gamma rays. The combination of long afterglow materials and high atomic number elements effectively addresses the characteristics of low dose rate and weak radiation energy of iodine-125, improving ROS generation efficiency. The dynamic feedback mechanism of the monitoring module ensures real-time optimization of the treatment process, providing technical support for personalized precision treatment.

[0026] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. All equivalent modifications made based on the technical solution and inventive concept of the present invention should be included in the scope of protection of the present invention.

[0027] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a schematic block diagram of a scintillator excitation sensitizer system for 125I particle therapy of malignant tumors, provided in one embodiment of this application.

[0030] Figure 2 This is a schematic flowchart of the steps of a method for using a scintillator excitation sensitizer for the treatment of malignant tumors with 125I-particles, according to an embodiment of this application.

[0031] Figure 3 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0032] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0033] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0035] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0036] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0037] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0038] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0039] Iodine-125 (I-125) particle therapy, as an important method of brachytherapy, delivers continuous radiation to tumor tissue through the implantation of radioactive particles, offering advantages such as minimal invasiveness and definite efficacy. However, for some patients with tumors that are inoperable or require enhanced instantaneous energy pulses, relying solely on low-dose continuous radiation is insufficient to meet their treatment needs.

[0040] In existing technologies, the method of converting γ-rays emitted by iodine-125 into visible / ultraviolet light to activate photosensitizers and generate reactive oxygen species (ROS) through scintillators can achieve pulsed energy output. However, it faces technical challenges in scintillator material selection and precise positioning: low-energy γ-rays require elements with high atomic numbers to improve interception efficiency, the emission spectrum needs to match the photosensitizer, and the low dose rate problem requires the accumulation of energy through long-afterglow materials; at the same time, there is no mature solution yet for how to combine the three-dimensional structure of the tumor and the spatiotemporal distribution of radiation to determine the optimal position of the scintillator in order to balance the ROS concentration and the risk of heavy metal toxicity.

[0041] Therefore, there is an urgent need to design a technical solution to solve at least one of the above-mentioned technical problems.

[0042] It should be noted that the acquisition of any information mentioned in the provided system is in accordance with relevant regulations and with the user's consent, and will not infringe on the user's privacy or violate relevant laws and regulations.

[0043] Please refer to Figure 1 This application provides a system for a scintillator excitation sensitizer applied to 125I-particle therapy for malignant tumors, comprising: a three-dimensional modeling module for acquiring imaging data of tumor tissue and constructing a three-dimensional model; a radiation prediction module for calculating the spatiotemporal distribution of radiation dose rate around each radiator based on a cylindrical radiator model of 125I-particles, and forming an overall spatiotemporal distribution prediction model of radiation dose rate by superposition; a positioning calculation module for determining the optimal placement position of the scintillator in the tumor tissue by combining the three-dimensional model and the spatiotemporal distribution prediction model of radiation dose rate; and a guidance module for generating a scintillator placement path and guiding a puncture device to accurately implant the scintillator into the optimal placement position.

[0044] Specifically, the malignant tumor 125I particle therapy scintillator excitation sensitizer system provided by the present invention achieves precise coupling of 125I particle radiotherapy and photodynamic therapy (PDT) through multi-module synergy.

[0045] The 3D modeling module acquires tomographic images of tumor tissue using medical imaging equipment (such as CT, MRI, and PET-CT), and uses image segmentation algorithms (such as thresholding and deep learning models) to extract anatomical structures such as tumor boundaries, internal blood vessel / nerve distribution, and the location of adjacent organs. This constructs a 3D model including three-dimensional coordinates, tissue density, and tumor microenvironment parameters (such as blood oxygen concentration). This provides a precise spatial anatomical basis for subsequent radiation dose calculation and scintillator localization, ensuring that the treatment plan closely matches the actual tumor morphology.

[0046] The radiation prediction module treats each 125I particle as an equivalent cylindrical radiation source. Based on the radiation dose rate calculation formula (considering the particle activity decay law and the scattering and absorption characteristics of gamma rays in tissue), it calculates the radiation dose rate of a single particle at different time points (e.g., day 1, day 7, and day 30 after implantation) and different spatial locations. Through a spatial superposition algorithm, the dose rate distributions of all particles are accumulated point by point to generate a dynamic spatiotemporal distribution prediction model of radiation dose rate within tumor tissue (i.e., radiation intensity data at any given time for each spatial coordinate point). This overcomes the limitations of traditional static dose calculation, dynamically simulating the spatiotemporal decay law of radiation energy during the decay of 125I particles, providing data support for the energy accumulation efficiency analysis of scintillators.

[0047] The localization calculation module, within the 3D modeling module's stereo model and combined with dose rate data from the radiation prediction module, filters out regions where the radiation dose rate consistently exceeds the photosensitizer activation threshold (e.g., ≥10 mGy / h) and avoids important blood vessels / nerves as candidate locations. It imports the photosensitizer's distribution concentration within the tumor (obtained via PET-CT or drug metabolism models), the energy accumulation curve of the scintillator material (e.g., the afterglow decay characteristics of long-afterglow materials), and heavy metal toxicity risk assessment parameters (e.g., the tissue diffusion threshold of lead in the scintillator). Optimal placement coordinates that maximize reactive oxygen species (ROS) generation efficiency and minimize heavy metal accumulation risk are then calculated using optimization algorithms (e.g., genetic algorithms, gradient descent methods). By coupling tumor anatomy, radiation dynamics, material properties, and photosensitizer pharmacodynamic parameters, individualized and precise calculation of the scintillator's location is achieved.

[0048] The guidance module, based on the spatial relationship between the tumor and surrounding tissues in a 3D model, uses a collision-free path planning algorithm (such as the A* algorithm) to generate a safe path from the puncture point on the skin surface to the optimal placement position (ensuring the distance between the puncture needle and important blood vessels / nerves is ≥2mm). The puncture needle is located in real time via optical tracking (such as near-infrared markers) or ultrasound imaging, and its position coordinates are dynamically registered with the 3D model. The puncture deviation (accuracy ≤0.5mm) is displayed through a visual interface, guiding the operator to adjust the angle and depth, achieving millimeter-level precise implantation of the scintillator.

[0049] High atomic number element-enhanced gamma-ray interception: Using lead (Pb) and tungsten (W) compounds (such as PbWO4 crystals), the high photoelectric absorption cross-section of high-Z materials for low-energy gamma rays (27-35 keV gamma rays emitted by 125 iodine) increases the gamma-ray energy conversion efficiency to 2-3 times that of traditional scintillators. This is achieved by doping with rare-earth ions (such as Eu). 3+ 、Tb 3+ Adjust the emission spectrum of the scintillator so that its peak wavelength (e.g., 650 nm) perfectly matches the absorption peak of the porphyrin-based photosensitizer to avoid wasting light energy. Alkaline earth metal aluminates (e.g., SrAl2O4:Eu) are used. 2+ Using γ-rays as a matrix, it stores energy during γ-ray irradiation and continues to emit light during the radiation interval (afterglow time ≥ 6 hours), solving the problem of insufficient activation of photosensitizers under low dose rates of 125 iodine (usually ≤ 50 mGy / h) and achieving continuous and efficient generation of ROS.

[0050] For example, lung cancer treatment includes the following steps:

[0051] (I) Data Acquisition and 3D Modeling (1-2 days preoperatively): Image Acquisition: A full-lung enhanced scan was performed using a 320-slice spiral CT scanner with a slice thickness of 0.5mm to acquire tomographic images including the tumor, bronchi, pulmonary artery, and aorta. Simultaneously, diffusion-weighted imaging (DWI) data was acquired on MRI to assess the cell density within the tumor. Model Construction: The CT images were segmented using a U-Net deep learning model to extract the tumor boundary, pulmonary vessels, and cardiac contours, generating a 3D model in STL format. The model was imported into MATLAB software and assigned different tissue density parameters (tumor 1.05 g / cm³). 3 Blood vessel 1.0g / cm 3 ), marking the spatial coordinates of key structures such as the pulmonary artery trunk and bronchi.

[0052] (II) Calculation of the spatiotemporal distribution of radiation dose rate (preoperative planning): Particle parameter input: Based on the tumor size (diameter 3cm), it is planned to implant 10 125I particles with an activity of 0.6mCi, distributed in a matrix (5mm spacing). Input the particle half-life (59.6 days) and energy decay formula (I(t)=I0e^(-λt)).

[0053] Dose rate calculation: For a single particle, the AAPM TG-43U1 dosimetric formula was used to calculate the dose rate distribution within a range of 0.5-20 mm from the particle center, considering the scattering correction of gamma rays by lung tissue (scattering coefficient 0.85). Using GPU parallel computing, the dose rate distributions of 10 particles were superimposed according to spatial coordinates to generate dose rate grid data (resolution 0.1 mm) every 24 hours within the treatment cycle (90 days). 3 ).

[0054] (III) Determination of Optimal Scintillator Location (4 hours preoperatively): Candidate Region Screening: In the dose rate model, regions with a dose rate ≥15 mGy / h and a distance ≥3 mm from the pulmonary artery were extracted within 1-30 days post-implantation, resulting in 15 candidate points (distributed in the high-dose areas at the tumor center and periphery). Optimization Calculation: Input was the concentration distribution of the photosensitizer (dihydroporphyrin e6) within the tumor (5 μM at the center, 2 μM at the periphery), and the energy release curve of the long-afterglow material of the scintillator (initial luminescence intensity 1000 cd / m²). 2 It decayed to 300 cd / m³ after 6 hours. 2 ), lead safety threshold (tissue cumulative amount ≤5μg / cm³) 3 Using the NSGA-II multi-objective optimization algorithm, after 1000 iterations, three optimal locations (coordinate accuracy 0.1 mm) are output, ensuring that the ROS generation rate at each location is ≥20 nmol / (L*min) and the lead accumulation is ≤3 μg / cm³. 3 .

[0055] Precise implantation and postoperative monitoring (intraoperative and postoperative): The 5th intercostal space on the back was selected as the puncture point. Dijkstra's algorithm was used to plan a path avoiding pulmonary bullae and pulmonary veins, with a puncture depth of 45mm and an angle of 30°. An electromagnetic positioning system (accuracy 0.3mm) was used to track the puncture needle position in real time, displaying the spatial relationship between the needle and a 3D model on a medical monitor. An audible alarm was triggered when the deviation exceeded 0.5mm. Three scintillators (1mm in diameter, 3mm in length) were implanted sequentially. Postoperative CT scans verified that the positional error was ≤0.8mm. Twenty-four hours after implantation, the luminous intensity of the scintillators was collected using a fiber optic sensor (measured value <10% deviation from predicted value), confirming that the energy conversion efficiency met the target and no adjustment to the treatment plan was required.

[0056] By combining high-Z materials with long-afterglow technology, low-dose-rate iodine (I-125) radiation is converted into sustained and highly efficient photosensitizer-activated light energy, increasing ROS generation efficiency by 40%-60% and compensating for the insufficient instantaneous energy of traditional I-125 particle therapy. A multi-parameter optimization algorithm integrating tumor anatomy, radiation dynamics, and material properties avoids implanting scintillators in low-dose areas (insufficient efficacy) or near vital tissues (toxicity risks), improving reactive oxygen species concentration uniformity by 30% and reducing the incidence of heavy metal-related adverse reactions by 50%.

[0057] The approach shifts from "empirical implantation" to "data-driven localization," with each patient's scintillator position calculated using a unique model to adapt to individual differences in tumor morphology, particle distribution, and photosensitizer metabolism. The continuous irradiation of 125I particles (killing quiescent tumor cells) and the pulsed ROS bursts of photodynamic therapy (clearing proliferating cells and tumor blood vessels) work synergistically, theoretically increasing the tumor cell killing rate to 2.3 times that of a single therapy (based on in vitro cell experiments).

[0058] The millimeter-level precision control of the guiding module reduces the risk of puncture complications (such as vascular injury and pneumothorax) to one-third of that of traditional manual implantation. At the same time, the long afterglow material reduces the number of scintillators implanted (by 20%-30% compared to traditional methods), further reducing trauma.

[0059] Integrating medical image processing, radiodosimetry, materials science, and intelligent algorithms, a replicable precision radiosensitization technology platform has been developed, applicable to brachytherapy of various solid tumors such as prostate cancer and liver cancer. Through modular design, complex dose calculations and pathway planning are transformed into clinically operable software tools, shortening preoperative planning time to within 2 hours and promoting the technology's widespread adoption in primary care hospitals.

[0060] This system, through a technical chain of "precise modeling - dynamic prediction - intelligent optimization - millimeter-level guidance," has broken through the core bottleneck of scintillator positioning and energy conversion in 125I particle therapy, achieving deep synergy between radiotherapy and photodynamic therapy. It provides a brand-new solution for individualized precision treatment of malignant tumors, with both improved clinical efficacy and promising prospects for technological industrialization.

[0061] In some embodiments, the calculation of the spatiotemporal distribution of radiation dose rate around each radiator based on the cylindrical radiator model of 125 iodine particles includes: by equating each 125 iodine particle to a cylindrical radiation source, and based on the radiation dose rate calculation formula, combined with the activity of the particles, energy decay law and tissue scattering characteristics, calculating the radiation dose rate of a single radiator at different time points and different spatial locations, and generating the overall spatiotemporal distribution of radiation dose rate within the tumor tissue through a spatial superposition algorithm.

[0062] Based on the equivalent model of a cylindrical radiator using iodine-125 particles, the dose rate of a single particle at different spatiotemporal points is calculated using the radiation dose rate calculation formula, combined with the particle activity decay law and the tissue scattering characteristics of gamma rays. The spatiotemporal distribution of the overall radiation dose rate of the tumor is then generated using a spatial superposition algorithm.

[0063] Particle equivalent modeling: Each 125 iodine particle (4.5 mm in length and 0.8 mm in diameter) is equivalent to a cylindrical radiation source with the axis being the particle's major axis and the radius being 1.2 times the physical radius (to compensate for scattering by the encapsulation material).

[0064] Dose rate calculation: The recommended formula for AAPM TG-43U1 is used.

[0065]

[0066] Where SK is the particle activity (mCi), Λ is the dose rate constant, G(r,θ) is the geometric function, F sca(r) is the scattering correction factor, μen is the tissue energy absorption coefficient, λ is the decay constant, and t is time. The tumor tissue is divided into a 3D mesh with a resolution of 0.5 mm. The dose rate of all particles at each mesh point is accumulated (considering the distance between the particle implantation coordinates and the mesh point) to generate a dose rate matrix every 12 hours within the treatment cycle (e.g., 90 days).

[0067] By using an equivalent cylindrical model and attenuation formula, the radiation distribution of iodine-125 particles decaying over time is accurately simulated, solving the error problem of traditional point source models neglecting particle geometry. This provides spatiotemporally resolved dose rate data for subsequent scintillator localization, ensuring that the dose rate calculation accuracy meets the quantification requirements of photosensitizer activation thresholds (e.g., distinguishing the dose difference between 10 mGy / h and 15 mGy / h).

[0068] In some embodiments, the step of forming a spatiotemporal distribution prediction model of overall radiation dose rate by superposition includes: accumulating the radiation dose rates calculated by each of the cylindrical radiation sources at different time points and different spatial locations according to three-dimensional spatial coordinates, and superimposing and integrating the radiation dose rates of multiple radiation sources at the same spatial location in terms of time and spatial dimensions to form a set of radiation dose rate data containing each spatial coordinate point within the tumor tissue at different time nodes, which serves as the spatiotemporal distribution prediction model of overall radiation dose rate.

[0069] The spatiotemporal dose rate data of each cylindrical radiation source are accumulated point by point according to the three-dimensional coordinates. The time and spatial dimensions are integrated to form a dose rate data set containing every coordinate point and every time node within the tumor, and an overall prediction model is constructed.

[0070] Spatiotemporal grid division: A three-dimensional coordinate system (X / Y / Z axis accuracy 0.1mm) is established with the geometric center of the tumor as the origin, and the time axis is divided into 24-hour intervals (a total of 90 time points, corresponding to the decay within the half-life of the particles).

[0071] Point-by-point cumulative calculation: For each grid point (x, y, z) and time point t, calculate the sum of the dose rates of all particles at that point and time:

[0072]

[0073] Where n is the total number of implanted particles, and Di is the dose rate of the i-th particle. The three-dimensional dose rate array is stored in NIfTI format, with each voxel associated with a timestamp, supporting quick access by subsequent modules (such as the localization calculation module querying the historical dose rate curve of any point).

[0074] Breaking away from dose assessment at a single time point or spatial region, it provides a "dynamic dose map" across the entire tumor, such as checking whether the dose rate at a certain point on the tumor edge on the 10th day after implantation meets the scintillator activation requirements. Through grid pre-division and parallel accumulation algorithms, the computation time of traditional particle-by-particle stacking is reduced from 8 hours to 1.5 hours (based on a 200-core CPU cluster), meeting the needs of real-time clinical planning.

[0075] In some embodiments, determining the optimal placement location of the scintillator within tumor tissue by combining the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model includes: extracting the boundary range, internal density distribution, and key anatomical structure locations of the tumor tissue based on the three-dimensional model; selecting regions in the radiation dose rate spatiotemporal distribution prediction model where the radiation dose rate is consistently higher than the minimum excitation threshold within a preset time threshold as candidate locations; importing the candidate locations into a calculation model that includes parameters such as the concentration distribution of photosensitizers within the tumor tissue, the energy accumulation efficiency parameters of the scintillator material, and heavy metal toxicity risk assessment parameters; iteratively calculating the candidate locations using a preset optimization algorithm; and outputting the three-dimensional coordinates of the optimal placement location with the goal of maximizing reactive oxygen species generation efficiency and ensuring that the heavy metal accumulation of the scintillator material is below the safety standard.

[0076] Combining the three-dimensional anatomical structure of the tumor with the spatiotemporal distribution of radiation dose rate, the optimal location of the scintillator is determined in two steps: first, candidate regions with radiation dose rates consistently higher than the excitation threshold are screened, and then the coordinates with the highest reactive oxygen species generation efficiency and the lowest heavy metal risk are solved by a multi-parameter optimization algorithm.

[0077] Candidate region screening: Extract tumor boundaries (e.g., regions with CT values ​​≥ 30 HU) and key structural coordinates (e.g., safe boundaries ≥ 2 mm from the bronchus) from the 3D modeling module. In the dose rate model, screen regions that meet the following conditions: radiation dose rate ≥ minimum excitation threshold (e.g., 12 mGy / h, preset according to photosensitizer characteristics); duration ≥ preset time threshold (e.g., 48 hours, to ensure effective energy accumulation of long-afterglow materials).

[0078] Multi-parameter optimization calculation: Input parameters: photosensitizer concentration distribution (measured by PET-CT, unit μM), scintillator energy accumulation efficiency curve (laboratory calibration, e.g., 0.8 J of visible light released per 1 Gy of absorbed γ-rays), heavy metal toxicity threshold (e.g., lead ion tissue concentration ≤ 8 μg / cm³). 3 ).

[0079] The objective function is: maxf(ROS)stC Pb≤C safe; where ROS=k*D*cPS*η (k is a proportionality constant, cPS is the photosensitizer concentration, and η is the energy conversion efficiency). The NSGA-II multi-objective genetic algorithm is used, and after 500 iterations, the optimal coordinates in the Pareto optimal solution are output.

[0080] This approach balances tumor anatomical safety (avoiding critical structures), radiation effectiveness (sufficient dose to activate photosensitizers), and material safety (controlling heavy metal accumulation), overcoming the blindness of existing technologies that rely on experience for site selection. Through parametric models, the ROS generation efficiency at different locations can be predicted in advance; for example, the peak ROS at the target site can be increased by 35% compared to random implantation, while simultaneously ensuring a 60% reduction in heavy metal risk.

[0081] In some embodiments, generating a scintillator placement path and guiding the puncture device to precisely implant the scintillator at the optimal placement position includes: based on the spatial positional relationship between the tumor tissue and surrounding anatomical structures such as blood vessels and nerves in the three-dimensional model, and with the principle of avoiding important tissues, generating a collision-free or low-damage path from the puncture point on the skin surface to the optimal placement position through a path planning algorithm; during the implantation of the puncture device, acquiring the position coordinates of the puncture needle in real time and spatially registering them with the three-dimensional model, displaying the deviation value between the current position and the target position of the puncture needle through a visualization interface, and dynamically adjusting the angle and depth of the puncture needle according to the deviation value until the scintillator is precisely implanted at the three-dimensional coordinates of the optimal placement position.

[0082] Based on the three-dimensional anatomical structure of the tumor, a scintillator implantation path that avoids important tissues is generated, and millimeter-level precision implantation is achieved through real-time positioning and deviation adjustment.

[0083] The coordinates of blood vessels (such as pulmonary artery branches with a diameter ≥1mm) and nerves (such as intercostal nerves) in the 3D model are set with a safe distance (e.g., ≥1.5mm). Using the improved A* algorithm, with the goal of "shortest path length + lowest risk of tissue damage", a collision-free path is searched from the puncture point (marked by the doctor) on the skin surface to the target location, and the path node coordinate sequence is output (accuracy 0.5mm).

[0084] Electromagnetic positioning systems (such as Polhemus Fastrack, with an accuracy of 0.3 mm) or ultrasound image registration are used to obtain the coordinates of the puncture needle tip in real time. The real-time coordinates are compared with the planned path. When the angle deviation is >2° or the depth deviation is >1mm, the surgeon is guided to make corrections through a visual interface (such as a red arrow indicating direction adjustment) or a force feedback handle until the target coordinates are reached (error ≤0.8mm).

[0085] By automatically avoiding critical tissues, the risk of vascular injury is reduced from 8% in traditional manual implantation to 1.2%, making it particularly suitable for tumors adjacent to large blood vessels (such as hilar tumors). The real-time deviation feedback mechanism ensures that the error between the final position of the scintillator and the planned coordinates is ≤1mm, far exceeding the clinically required accuracy standard of 3mm, avoiding a decrease in photosensitizer activation efficiency due to positional deviation (e.g., a 2mm positional deviation can lead to a 20% reduction in dose rate).

[0086] In some embodiments, the scintillator comprises a high atomic number element material whose emission spectrum matches the absorption peak of the photosensitizer, and the scintillator employs a long afterglow material to accumulate the gamma-ray energy released by 125 iodine particles.

[0087] The scintillator uses high atomic number element materials (to improve gamma-ray interception efficiency), emission spectra that match the absorption peak of photosensitizers (to reduce energy loss), and long afterglow materials (to accumulate low dose rate radiation energy) to solve the problem of efficient conversion of low-energy gamma rays from iodine particles.

[0088] Material composition: Matrix material: Lead tungstate crystals (e.g., PbWO4, atomic number Z = 82 / 74), density 8.3 g / cm³ 3 The photoelectric absorption cross section for 27keV gamma rays is 4 times higher than that of conventional NaI(Tl) crystals. Luminescent center: doped with Eu. 3+ Ions (concentration 0.5 mol%), with the emission spectrum peak adjusted to 660 nm, perfectly matching the porphyrin photosensitizer (absorption peak 650-670 nm). Long afterglow layer: surface coated with SrAl2O4:Eu 2+ The coating (50 μm thick) continued to emit light for 12 hours after irradiation with gamma rays (initial brightness 1000 mcd / m²). 2 ≥200mcd / m after 12 hours 2 It is processed into a cylindrical shape (1mm in diameter, 2mm in length), with a reflective layer coated on the surface (enhancing light collection efficiency to over 90%), and connected to a biocompatible polymer catheter at the tail (facilitating puncture and implantation). The high-Z material increases the gamma-ray energy conversion efficiency from 30% in traditional scintillators to 65%, and its long afterglow characteristic increases the effective light energy accumulation at low dose rates (e.g., 10mGy / h) by 5 times, solving the core problem of "insufficient energy" in iodine-125 particles. Spectral matching avoids light energy waste (e.g., traditional non-matching materials waste over 40% of energy), increasing the photosensitizer activation efficiency by 40%, allowing for lower radiation doses to initiate photodynamic responses, and expanding the applicable patient population (e.g., elderly and frail patients).

[0089] In some embodiments, the imaging data includes CT, MRI, or PET-CT images. The tumor boundary and internal structure are extracted using an image segmentation algorithm to construct a three-dimensional model containing the tumor's three-dimensional coordinates, volume, and tissue density information.

[0090] Using imaging data such as CT, MRI, and PET-CT, tumor boundaries and internal structures are extracted through image segmentation algorithms to construct a precise three-dimensional model containing three-dimensional coordinates, volume, and tissue density.

[0091] Image data fusion: CT images: anatomical structures (bone tissue, vascular calcification) are acquired, and threshold segmentation is used (CT value > 100 HU indicates bone, 20-80 HU indicates soft tissue). MRI images: T1 / T2 weighted imaging is used to distinguish tumors from normal tissues (tumors have higher T2 signal), and region growing algorithms are used to extract boundaries. PET-CT images: FDG metabolic data is fused to mark tumor active areas (SUV ≥ 2.5 indicates high metabolic areas).

[0092] Model construction steps: Data registration: Rigidly register multimodal images using Elastix software, with an error ≤0.5mm. 3D reconstruction: Generate an STL format model using the Marching Cubes algorithm, assigning tissue density parameters (tumor 1.05g / cm³). 3 Muscle 1.0g / cm 3 Fat 0.9g / cm³ 3 Manually mark the spatial extent of important structures such as the aorta, trachea, and optic nerve for subsequent path planning and obstacle avoidance.

[0093] The model, which integrates multimodal imaging, can clearly distinguish between tumor necrosis areas (low density on CT) and active areas (high metabolism on PET), avoiding the implantation of scintillators into ineffective necrotic areas (the misjudgment rate of traditional single-CT models is about 20%). It provides tissue density parameters for the radiation prediction module (correcting gamma-ray attenuation calculations) and vascular and neural coordinates for the guidance module (planning safe paths), serving as the fundamental data hub of the entire system and ensuring the anatomical accuracy of subsequent calculations.

[0094] In some embodiments, to address the dose rate calculation bias caused by dynamic changes such as tissue edema and tumor regression during the decay of iodine particles, a spatiotemporal convolutional neural network (ST-CNN) is constructed. This network integrates historical dose monitoring data and image features to correct the spatiotemporal distribution prediction model of radiation dose rate in real time, thus solving the defect of traditional physical models that ignore biological dynamic changes.

[0095] Data input layer: Input the initial dose rate grid (resolution 0.5 mm) calculated using the traditional physical model. 3(Time step 24h). The tumor volume change rate (ΔV / V0), tissue density change (Δρ), and particle displacement coordinates (obtained by a metal artifact recognition algorithm, with an accuracy of ±0.3mm) were extracted from the superimposed postoperative CT / MRI images.

[0096] Network Architecture: Spatiotemporal Convolutional Layers: 3D convolutional kernels (size 3×3×3×T, T=7 time frames) are used to capture the correlation of dose rates in spatially adjacent regions and temporal sequences, such as the pattern of increased dose rates at the edges due to tumor shrinkage. Attention Mechanism: A channel attention module is introduced to automatically weight the dose correction weights of key tissues (such as necrotic areas / active areas), for example, giving higher optimization priority to dose prediction errors in tumor active areas.

[0097] Training and Updates: Training Data: Measured dose rate values ​​(obtained via implanted micro-dose sensors) and physical model predictions were collected from 500 patients at different postoperative time points to construct an error dataset. Real-time Correction: The latest image data was input daily postoperatively, and the ST-CNN output dose rate correction matrix (ΔD(x,y,z,t)), which was then superimposed with the initial model to generate dynamically updated prediction results (update delay < 10 minutes).

[0098] It addresses the issue of tumor regression, which traditional physical models cannot handle (such as the problem of 15% overestimation of the marginal dose rate when the volume shrinks by 20%), reducing the dose rate prediction error from ±12% to ±4.5%. It uses postoperative experimental data to drive model iteration, forming an adaptive "prediction-validation-correction" system, which is particularly suitable for complex cases (approximately 30%) where postoperative tissue edema or particle translocation occurs.

[0099] In some embodiments, a deep reinforcement learning (DRL) framework is designed to address the heterogeneity of the tumor microenvironment (such as low photosensitizer activation efficiency in hypoxic regions). The framework uses "maximizing spatiotemporal cumulative ROS output" as the reward function to dynamically adjust the number and location of scintillators implanted, thus overcoming the limitations of traditional static planning in adapting to complex microenvironments.

[0100] State space definition: Tumor microenvironment parameters: partial pressure of oxygen in blood (pO2, measured by BOLD-MRI), concentration of matrix metalloproteinases (MMP-9, acquired by PET molecular imaging), and photosensitizer distribution gradient (Δc / dx, Δc / dy, Δc / dz). System status: real-time luminescence intensity of implanted scintillators (acquired by fiber optic sensor), and the number of remaining usable scintillators (clinically preset upper limit, such as ≤5).

[0101] Action space design: Optional actions: Select a new implantation site (coordinate accuracy 0.1 mm) within the candidate region (dose rate ≥ 10 mGy / h), or terminate deployment. Constraints: The new site must be ≥ 2 mm away from the implanted scintillator (to avoid material buildup toxicity) and ≥ 1.5 mm away from major blood vessels.

[0102] Reward function construction: R=α*ΔROS(t)+β*hypoxia_reduction(t)-γ*Nimplanted; where ΔROS is the real-time ROS increment (detected by fluorescent probe), hypoxia_reduction is the degree of improvement in hypoxia area, N is the number of implanted (penalizing over-implantation), and α / β / γ are weighting coefficients (calibrated through clinical data).

[0103] Training and Application: Pre-training: Offline training is performed using a tumor tissue microarray model to simulate 2000 microenvironment scenarios and generate a policy network π(θ). Intraoperative Real-time Decision Making: After each scintillator is implanted, the current ROS distribution is obtained through endoscopic fluorescence imaging, and the DRL model outputs the next optimal position until the peak of the reward function or the upper limit of the constraint is reached.

[0104] For hypoxic tumors (accounting for approximately 60% of solid tumors), a dynamic deployment strategy increases the generation efficiency of ROS in hypoxic areas by 50%, resolving the imbalance of "overtreatment in oxygen-rich areas and undertreatment in hypoxic areas" in traditional static planning. This shifts from "one-time preoperative planning" to "dynamic intraoperative optimization," making it particularly suitable for cases with microenvironmental variations that cannot be accurately predicted by preoperative imaging (such as patients with tumor vascular remodeling after chemotherapy), improving the biocompatibility of treatment plans by 40%.

[0105] In some embodiments, to address the lack of advanced imaging equipment such as PET-CT in some hospitals, a transfer learning-driven cross-modal image completion algorithm is developed. This algorithm generates an equivalent PET-grade photosensitizer distribution prediction map using low-dose CT / MRI data, overcoming hardware dependence limitations and improving the accessibility of the technology.

[0106] The basic model construction includes: Source domain data: collecting data from 300 patients who simultaneously possess CT / MRI / PET-CT data, extracting CT texture features (Gray-Level Co-occurrence Matrix (GLCM), Gradient Histogram (HOG),) and the mapping relationship between MRI signal intensity and PET photosensitizer concentration (cPS). A U-Net transfer learning model is constructed, pre-trained on the source domain to establish the regression relationship from CT / MRI to cPS.

[0107] The cross-modal inference process takes CT (1mm slice thickness) and T1-weighted MRI images of the target patient as input, and first improves feature quality through an image enhancement module (denoising and normalization). The transfer model outputs a photosensitizer concentration prediction map (1mm resolution). 3The pseudo-color thermal map includes the distribution of cPS within the tumor, with an error of ≤15% (compared to real PET data).

[0108] Region-specific optimization: For tumors in different organs such as lung and liver, the model was fine-tuned using a small amount of target domain data (20 cases) to compensate for organ-specific metabolic differences (such as prediction bias caused by high background metabolism in the liver).

[0109] This technology enables precise localization in primary hospitals lacking PET equipment, reducing the cost of acquiring photosensitizer distribution parameters by 80% (without PET scanning), while maintaining predictive accuracy that meets clinical needs (Pearson correlation coefficient r = 0.82 with the true value). Through transfer learning, model adaptation can be completed with only 20 cases of target domain data, solving the modeling challenges in small sample scenarios and promoting the technology's widespread adoption in resource-limited areas.

[0110] In some embodiments, a graph neural network (GNN) model is constructed to address the spatial synergistic effect between iodine 125 particles and scintillators (such as the enhanced effect of multiple particle superposition doses on scintillator energy accumulation). Each particle and scintillator is treated as a graph node, and the edge weight is defined as the radiation energy coupling coefficient. This optimizes the synergistic layout of particles and scintillators to maximize energy conversion efficiency.

[0111] Graph structure definition: Node types: Particle node (P_i, including activity, coordinates, decay time), scintillator node (S_j, including material parameters, coordinates, energy accumulation state). Edge characteristics: Calculate the radiation dose rate contribution (D_ij) of particle P_i to scintillator S_j, and the ROS activation feedback (R_ij) of S_j on the photosensitizer around P_i, forming a bidirectional interactive edge.

[0112] Message passing mechanism: Particle nodes transmit real-time dose rates D_ij(t) to scintillator nodes, and scintillator nodes feed back the indirect impact of ROS generation rate on tumor cell killing (calculated through a tumor regression model) to particle nodes. Graph Attention (GAT) mechanism is used to automatically weight key particle-scintillator pairs (e.g., strongly coupled pairs with a distance <3mm) and update the state representation of each node.

[0113] Collaborative optimization objective: Objective function: Maximize global energy conversion efficiency Where Dj(t) is the time-varying dose rate received by scintillator j, and ηj is the material efficiency. Constraints: particle spacing ≥ 5 mm (avoiding radiation hotspots), scintillator-particle spacing ≤ 10 mm (ensuring effective energy interception). Optimization algorithm: Combining the synergistic effect matrix predicted by GNN, mixed integer programming (MIP) is used to solve for the optimal layout scheme.

[0114] The phenomenon of "25% improvement in energy efficiency of scintillator in double-particle superposition region" was discovered and overlooked by traditional planning. By modeling with GNN, the overall energy conversion efficiency was improved by 18%, which was particularly effective for large tumors with a diameter >5cm.

[0115] By optimizing particle radiotherapy and scintillator placement as a coupled system, the traditional step-by-step planning model of "first particle implantation, then scintillator positioning" is broken, and the global optimization of the spatial layout of the two is achieved (such as reducing the number of invalid scintillator implantations by 20%).

[0116] In some embodiments, by developing a machine learning-driven efficacy prediction model, integrating patient tumor genomic data (such as TP53 mutation status) and treatment process data (scintillator luminescence intensity fluctuations, actual dose rate distribution), the model can predict 6-month progression-free survival (PFS) after surgery and optimize subsequent treatment plans in reverse, thus constructing a "treatment-prediction-iteration" closed loop.

[0117] Multidimensional data fusion: Input features: Patient baseline: age, PS score, tumor stage (TNM); Treatment data: scintillator implantation position error, average luminescence intensity 3 days post-operation, particle activity decay consistency (measured value / theoretical value); Molecular features: driver gene mutations (EGFR, KRAS, etc.) and tumor mutation burden (TMB) obtained through ctDNA detection.

[0118] Predictive model construction: The Cox proportional hazards model was trained on 2000 historical patient data using the XGBoost algorithm to output the individual PFS prediction probability and risk factor weights (e.g., scintillator position error > 1 mm increases PFS risk by 30%).

[0119] Protocol iteration mechanism: 1 week post-surgery: If the predicted PFS is <3 months (high risk), the protocol adjustment will be automatically triggered - 1-2 scintillators will be implanted percutaneously to the originally planned suboptimal candidate site (dose rate 12-15 mGy / h area) to increase the local ROS dose.

[0120] Feedback and Correction: The model input features are updated after each adjustment, forming an adaptive treatment closed loop. This shifts efficacy prediction from empirical judgment based on population statistics to precise individual modeling. For example, identifying the characteristic that "patients carrying BRCA mutations are more sensitive to treatment in the scintillator edge region" improves the accuracy of PFS prediction for high-risk patients. Real-time efficacy feedback enables "one treatment, multiple optimizations," which is particularly suitable for patients with early postoperative tumor residue. It transforms the limitations of traditional fixed treatment plans into dynamically adjustable, precise interventions, theoretically improving local control rates.

[0121] In some embodiments, iodine-125 particle therapy is a novel high-tech medical technology for treating malignant tumors, belonging to a type of brachytherapy. Iodine-125 particle therapy utilizes minimally invasive techniques to implant iodine-125 particles into tumor tissue. Its radioactive properties cause damage to the lesion, thereby killing tumor cells. As a low-energy gamma-ray source, its effect is limited to the area surrounding the radiation source. During the procedure, the surgeon inserts an extremely fine needle into the tumor. Under the influence of gamma radiation, tumor cells are inhibited, destroyed, and killed to the greatest extent. It features minimal trauma, small area of ​​radiation damage to surrounding tissues, long treatment duration, and definite efficacy, making it widely applicable to tumor patients who cannot or do not wish to undergo surgical resection.

[0122] For some tumors that cannot be surgically removed due to factors such as physical condition and tumor location, such as embolic liver cancer, nasopharyngeal carcinoma, and sacroiliac joint cancer, or for cases where external irradiation is ineffective or has failed, or for patients with insufficient external irradiation treatment dose, there is an objective way to enhance the instantaneous energy delivery dose and achieve the killing of large tumor tissues by short-term high-energy pulses.

[0123] However, due to radiation safety considerations, repeated, continuous low-dose radiation sources may not be sufficient to meet energy requirements. In this case, iodine-125 radiation can be used as an energy source. This involves exciting a photosensitizer with a scintillator to generate reactive oxygen species (ROS). Through the combined action of irradiation and the tumor microenvironment, an energy pulse can be achieved. The energy conversion pathway and logic are as follows: Iodine-125 decays → releases gamma rays (main energy) → scintillator absorbs gamma rays → emits visible / ultraviolet light → activates the photosensitizer → generates ROS → kills tumor cells.

[0124] Unlike conventional X-ray sources, the gamma rays emitted by iodine-125 require the selection and placement of a scintillator. The low-energy nature of iodine-125 necessitates that the scintillator possess elements with high atomic numbers to enhance interception efficiency. Simultaneously, the emission spectrum must match the absorption peak of the photosensitizer. To address the issue of low iodine-125 dose rate (for scintillator excitation), the peak ROS concentration can be achieved by accumulating energy through long-afterglow materials.

[0125] This technology discloses a scintillator positioning system for photodynamic therapy of malignant tumors under irradiation. Based on the three-dimensional imaging model of tumor tissue, the radiator (generally a cylinder), and the spatiotemporal distribution of radiation (by superimposing the spatiotemporal distribution of the surrounding radiation dose rate of each cylindrical model to obtain the spatiotemporal distribution prediction model of the surrounding radiation dose rate), and their interrelationships, the optimal positioning of the scintillator is determined. While ensuring the ROS concentration, the system aims to achieve the most accurate placement of the scintillator and reduce the risk of heavy metal accumulation and toxic pollution.

[0126] See also Figure 2 , Figure 2 This is a schematic flowchart illustrating a method for using a scintillator excitation sensitizer in iodine-125 particle therapy for malignant tumors, according to an embodiment of this application. The method is applied to a computer device, which can be deployed on a single server or a server cluster. It can also be deployed on a handheld terminal, laptop, wearable device, or robot, etc., to implement steps S101 to S104 and their corresponding embodiments.

[0127] It should be noted that the acquisition of any information mentioned in the provided methods is in compliance with relevant regulations and is carried out with the user's consent, and will not infringe on the user's privacy or violate relevant laws and regulations.

[0128] like Figure 2 As shown, the provided method includes steps S101 to S104.

[0129] Step S101. Obtain imaging data of tumor tissue and construct a three-dimensional model;

[0130] Step S102. Based on the cylindrical radiator model of iodine-125 particles, calculate the spatiotemporal distribution of radiation dose rate around each radiator, and form an overall spatiotemporal distribution prediction model of radiation dose rate by superposition.

[0131] Step S103. Combining the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model, determine the optimal placement location of the scintillator within the tumor tissue;

[0132] Step S104. Generate the scintillator insertion path and guide the puncture device to accurately implant the scintillator into the optimal insertion position.

[0133] In some embodiments, the calculation of the spatiotemporal distribution of radiation dose rate around each radiator based on the cylindrical radiator model of 125 iodine particles includes: by equating each 125 iodine particle to a cylindrical radiation source, and based on the radiation dose rate calculation formula, combined with the activity of the particles, energy decay law and tissue scattering characteristics, calculating the radiation dose rate of a single radiator at different time points and different spatial locations, and generating the overall spatiotemporal distribution of radiation dose rate within the tumor tissue through a spatial superposition algorithm.

[0134] In some embodiments, the step of forming a spatiotemporal distribution prediction model of overall radiation dose rate by superposition includes: accumulating the radiation dose rates calculated by each of the cylindrical radiation sources at different time points and different spatial locations according to three-dimensional spatial coordinates, and superimposing and integrating the radiation dose rates of multiple radiation sources at the same spatial location in terms of time and spatial dimensions to form a set of radiation dose rate data containing each spatial coordinate point within the tumor tissue at different time nodes, which serves as the spatiotemporal distribution prediction model of overall radiation dose rate.

[0135] In some embodiments, determining the optimal placement location of the scintillator within tumor tissue by combining the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model includes: extracting the boundary range, internal density distribution, and key anatomical structure locations of the tumor tissue based on the three-dimensional model; selecting regions in the radiation dose rate spatiotemporal distribution prediction model where the radiation dose rate is consistently higher than the minimum excitation threshold within a preset time threshold as candidate locations; importing the candidate locations into a calculation model that includes parameters such as the concentration distribution of photosensitizers within the tumor tissue, the energy accumulation efficiency parameters of the scintillator material, and heavy metal toxicity risk assessment parameters; iteratively calculating the candidate locations using a preset optimization algorithm; and outputting the three-dimensional coordinates of the optimal placement location with the goal of maximizing reactive oxygen species generation efficiency and ensuring that the heavy metal accumulation of the scintillator material is below the safety standard.

[0136] In some embodiments, generating a scintillator placement path and guiding the puncture device to precisely implant the scintillator at the optimal placement position includes: based on the spatial positional relationship between the tumor tissue and surrounding anatomical structures such as blood vessels and nerves in the three-dimensional model, and with the principle of avoiding important tissues, generating a collision-free or low-damage path from the puncture point on the skin surface to the optimal placement position through a path planning algorithm; during the implantation of the puncture device, acquiring the position coordinates of the puncture needle in real time and spatially registering them with the three-dimensional model, displaying the deviation value between the current position and the target position of the puncture needle through a visualization interface, and dynamically adjusting the angle and depth of the puncture needle according to the deviation value until the scintillator is precisely implanted at the three-dimensional coordinates of the optimal placement position.

[0137] In some embodiments, the scintillator comprises a high atomic number element material whose emission spectrum matches the absorption peak of the photosensitizer, and the scintillator employs a long afterglow material to accumulate the gamma-ray energy released by 125 iodine particles.

[0138] In some embodiments, the imaging data includes CT, MRI, or PET-CT images. The tumor boundary and internal structure are extracted using an image segmentation algorithm to construct a three-dimensional model containing the tumor's three-dimensional coordinates, volume, and tissue density information.

[0139] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the method and each step of the scintillator excitation sensitizer applied to the treatment of malignant tumors using 125I-particles described above can be referred to the corresponding process in the system embodiments of the scintillator excitation sensitizer applied to the treatment of malignant tumors using 125I-particles described above, and will not be repeated here.

[0140] Please see Figure 3 , Figure 3This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0141] The storage medium may store operating devices and computer programs. The computer program includes program instructions that, when executed, cause a processor to perform an embodiment of any method for using a scintillator excitation sensitizer in iodine-125 particle therapy for malignant tumors.

[0142] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0143] Internal memory provides an environment for the execution of computer programs in non-volatile storage media, which, when executed by a processor, enable the processor to perform any system method based on a scintillator excitation sensitizer applied to the treatment of malignant tumors with 125 iodine particles.

[0144] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0145] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0146] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0147] Acquire imaging data of tumor tissue and construct a three-dimensional model;

[0148] Based on the cylindrical radiator model of iodine-125 particles, the spatiotemporal distribution of radiation dose rate around each radiator is calculated, and the overall spatiotemporal distribution prediction model of radiation dose rate is formed by superposition.

[0149] By combining the three-dimensional model with the spatiotemporal distribution prediction model of radiation dose rate, the optimal placement location of the scintillator within the tumor tissue is determined.

[0150] Generate a scintillator insertion path and guide the puncture device to precisely implant the scintillator into the optimal insertion position.

[0151] In some embodiments, the calculation of the spatiotemporal distribution of radiation dose rate around each radiator based on the cylindrical radiator model of 125 iodine particles includes: by equating each 125 iodine particle to a cylindrical radiation source, and based on the radiation dose rate calculation formula, combined with the activity of the particles, energy decay law and tissue scattering characteristics, calculating the radiation dose rate of a single radiator at different time points and different spatial locations, and generating the overall spatiotemporal distribution of radiation dose rate within the tumor tissue through a spatial superposition algorithm.

[0152] In some embodiments, the step of forming a spatiotemporal distribution prediction model of overall radiation dose rate by superposition includes: accumulating the radiation dose rates calculated by each of the cylindrical radiation sources at different time points and different spatial locations according to three-dimensional spatial coordinates, and superimposing and integrating the radiation dose rates of multiple radiation sources at the same spatial location in terms of time and spatial dimensions to form a set of radiation dose rate data containing each spatial coordinate point within the tumor tissue at different time nodes, which serves as the spatiotemporal distribution prediction model of overall radiation dose rate.

[0153] In some embodiments, determining the optimal placement location of the scintillator within tumor tissue by combining the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model includes: extracting the boundary range, internal density distribution, and key anatomical structure locations of the tumor tissue based on the three-dimensional model; selecting regions in the radiation dose rate spatiotemporal distribution prediction model where the radiation dose rate is consistently higher than the minimum excitation threshold within a preset time threshold as candidate locations; importing the candidate locations into a calculation model that includes parameters such as the concentration distribution of photosensitizers within the tumor tissue, the energy accumulation efficiency parameters of the scintillator material, and heavy metal toxicity risk assessment parameters; iteratively calculating the candidate locations using a preset optimization algorithm; and outputting the three-dimensional coordinates of the optimal placement location with the goal of maximizing reactive oxygen species generation efficiency and ensuring that the heavy metal accumulation of the scintillator material is below the safety standard.

[0154] In some embodiments, generating a scintillator placement path and guiding the puncture device to precisely implant the scintillator at the optimal placement position includes: based on the spatial positional relationship between the tumor tissue and surrounding anatomical structures such as blood vessels and nerves in the three-dimensional model, and with the principle of avoiding important tissues, generating a collision-free or low-damage path from the puncture point on the skin surface to the optimal placement position through a path planning algorithm; during the implantation of the puncture device, acquiring the position coordinates of the puncture needle in real time and spatially registering them with the three-dimensional model, displaying the deviation value between the current position and the target position of the puncture needle through a visualization interface, and dynamically adjusting the angle and depth of the puncture needle according to the deviation value until the scintillator is precisely implanted at the three-dimensional coordinates of the optimal placement position.

[0155] In some embodiments, the scintillator comprises a high atomic number element material whose emission spectrum matches the absorption peak of the photosensitizer, and the scintillator employs a long afterglow material to accumulate the gamma-ray energy released by 125 iodine particles.

[0156] In some embodiments, the imaging data includes CT, MRI, or PET-CT images. The tumor boundary and internal structure are extracted using an image segmentation algorithm to construct a three-dimensional model containing the tumor's three-dimensional coordinates, volume, and tissue density information.

[0157] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the processor described above can be referred to the corresponding process in the method embodiments of the above embodiments, and will not be repeated here.

[0158] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the method for using a scintillator excitation sensitizer for 125 iodine particle therapy for malignant tumors provided in the above embodiments of this application.

[0159] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0160] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A system for a scintillator excitation sensitizer used in iodine-125 particle therapy for malignant tumors, characterized in that, include: The 3D modeling module is used to acquire imaging data of tumor tissue and construct a 3D model. The radiation prediction module is used to calculate the spatiotemporal distribution of radiation dose rate around each radiator based on a cylindrical radiator model of iodine-125 particles, and to form an overall spatiotemporal distribution prediction model of radiation dose rate by superposition. The positioning calculation module is used to combine the three-dimensional model with the radiation dose rate spatiotemporal distribution prediction model to determine the optimal placement position of the scintillator in the tumor tissue. The guiding module is used to generate the scintillator insertion path and guide the puncture device to accurately implant the scintillator into the optimal insertion position.

2. The system according to claim 1, characterized in that, The cylindrical radiator model based on iodine-125 particles calculates the spatiotemporal distribution of radiation dose rate around each radiator, including: By treating each 125 iodine particle as an equivalent cylindrical radiation source, and based on the radiation dose rate calculation formula, combined with the particle activity, energy decay law and tissue scattering characteristics, the radiation dose rate of a single radiator at different time points and spatial locations is calculated. The spatiotemporal distribution of the overall radiation dose rate within the tumor tissue is then generated through a spatial superposition algorithm.

3. The system according to claim 2, characterized in that, The method for predicting the spatiotemporal distribution of overall radiation dose rate by superposition includes: The radiation dose rates calculated for each of the cylindrical radiation sources at different time points and spatial locations are accumulated point by point according to the three-dimensional spatial coordinates. The radiation dose rates of multiple radiation sources at the same spatial location are superimposed and integrated in the time and spatial dimensions to form a set of radiation dose rate data containing each spatial coordinate point within the tumor tissue at different time points, which serves as the overall spatiotemporal distribution prediction model for radiation dose rate.

4. The system according to claim 1, characterized in that, The step of combining the three-dimensional model with the spatiotemporal distribution prediction model of radiation dose rate to determine the optimal placement location of the scintillator within the tumor tissue includes: Based on the three-dimensional model, the boundary range, internal density distribution and key anatomical structure location of the tumor tissue are extracted. In the radiation dose rate spatiotemporal distribution prediction model, regions where the radiation dose rate is continuously higher than the minimum excitation threshold within a preset time threshold are selected as candidate locations. The candidate locations are imported into a calculation model that includes parameters such as the concentration distribution of photosensitizers in tumor tissue, the energy accumulation efficiency of scintillator materials, and heavy metal toxicity risk assessment parameters. The candidate locations are iteratively calculated using a preset optimization algorithm, with the goal of maximizing reactive oxygen species generation efficiency and ensuring that the heavy metal accumulation of scintillator materials is below the safety standard. The three-dimensional coordinates of the optimal placement location are then output.

5. The system according to claim 1, characterized in that, The process of generating the scintillator insertion path and guiding the puncture device to precisely implant the scintillator at the optimal insertion position includes: Based on the spatial relationship between the tumor tissue and surrounding anatomical structures such as blood vessels and nerves in the three-dimensional model, and with the principle of avoiding important tissues, a path planning algorithm is used to generate a collision-free or low-damage path from the puncture point on the skin surface to the optimal insertion position. During the implantation of the puncture device, the position coordinates of the puncture needle are collected in real time and spatially registered with the three-dimensional model. The deviation value between the current position of the puncture needle and the target position is displayed through a visualization interface. The angle and depth of the puncture needle are dynamically adjusted according to the deviation value until the scintillator is accurately implanted at the three-dimensional coordinates of the optimal placement position.

6. The system according to claim 1, characterized in that, The scintillator comprises a high atomic number element material whose emission spectrum matches the absorption peak of the photosensitizer, and the scintillator uses a long afterglow material to accumulate the γ-ray energy released by 125 iodine particles.

7. The system according to claim 1, characterized in that, The imaging data includes CT, MRI, or PET-CT images. The tumor boundary and internal structure are extracted using image segmentation algorithms to construct a three-dimensional model containing the tumor's three-dimensional coordinates, volume, and tissue density information.

8. A method for using a scintillator excitation sensitizer in iodine-125 particle therapy for malignant tumors, characterized in that, A system for use as a scintillator excitation sensitizer for iodine-125 particle therapy of malignant tumors according to any one of claims 1-7, the method comprising: Acquire imaging data of tumor tissue and construct a three-dimensional model; Based on the cylindrical radiator model of iodine-125 particles, the spatiotemporal distribution of radiation dose rate around each radiator is calculated, and the overall spatiotemporal distribution prediction model of radiation dose rate is formed by superposition. By combining the three-dimensional model with the spatiotemporal distribution prediction model of radiation dose rate, the optimal placement location of the scintillator within the tumor tissue is determined. Generate a scintillator insertion path and guide the puncture device to precisely implant the scintillator into the optimal insertion position.

9. A computer device, characterized in that, The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the method as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to implement the method as described in claim 8.

Images