Medication dosage planning method and device, electronic equipment and storage medium

Abstract

The application belongs to the technical field of drug dose planning, and discloses a drug dose planning method, device, electronic equipment and storage medium. The method comprises the following steps: acquiring scanning image information of a subject, dividing the scanning image information to obtain blood vessel perfusion partitions, quantifying the activity of voxels of each blood vessel perfusion partition to obtain voxel activity concentration distribution data and tumor-normal tissue activity concentration ratio data of each blood vessel perfusion partition, predicting the absorbed dose according to the voxel activity concentration distribution data and the tumor-normal tissue activity concentration ratio data to obtain absorbed dose distribution data of a target drug, and calculating the optimal treatment dose based on the absorbed dose distribution data, in combination with a target constraint condition, a lesion priority weight and a target drug target dose to obtain the optimal treatment dose of the target drug in each blood vessel perfusion partition. Through the above method, the drug dose planning precision of radioembolization treatment is improved.

Description

Technical Field

[0001] This application relates to the technical field of drug dosage planning, and more specifically, to a drug dosage planning method, apparatus, electronic device, and storage medium. Background Technology

[0002] In radioembolization therapy (such as yttrium-90 radioembolization), preoperative dose planning is a crucial step in ensuring tumor control and the protection of normal tissues. Currently, the mainstream clinical methods include empirical fixed-dose methods, body surface area methods, and partition models. Among these, the partition model, based on SPECT / CT imaging, divides the liver into three partitions: tumor, normal liver, and lung. It estimates the absorbed dose per unit activity by calculating the uptake fraction of each partition, offering a level of individualization far superior to empirical fixed-dose methods and body surface area methods.

[0003] However, existing technologies still have key limitations. When a patient has two or more lesions independently supplied by different blood vessels, traditional partitioning models treat all tumors as a single "tumor partition," failing to distinguish the dose distribution of different lesions, let alone achieve independent prescription optimization for each lesion. Furthermore, clinical practice typically involves only one SPECT / CT scan. If multiple lesions require fractionated administration via different vessels, a single imaging session can only obtain a mixed perfusion distribution, lacking a method to decouple the mixed contribution into the independent contribution of each vessel. Existing models only provide the average dose for each partition, lacking voxel-level precision, making it difficult to assess intratumoral dose heterogeneity (such as DVH parameters like D90 and V100) and the fine irradiated volume of normal liver tissue.

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] The purpose of this application is to provide a drug dosage planning method, device, electronic device, and storage medium. By using preset target constraints, preset lesion priority weights, and preset target drug doses, combined with the absorbed dose distribution data of the target drug determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, the optimal therapeutic dose is calculated to obtain the optimal therapeutic dose of the target drug in each vascular perfusion zone. This solves the problem that existing radioembolization treatment methods cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion when using traditional zoning models. It can provide voxel-level precision dose assessment, accurately calculate the optimal therapeutic dose of the target drug in different vascular perfusion zones, and improve the drug dosage planning accuracy of radioembolization treatment.

[0006] In a first aspect, this application provides a drug dosage planning method, including: Obtain scanned image information of the examinee; The scanned image information is divided into regions to obtain vascular perfusion zones; Voxel activity was quantified in each of the aforementioned vascular perfusion zones to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each of the aforementioned vascular perfusion zones. Based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data, the absorbed dose is predicted to obtain the absorbed dose distribution data of the target drug. Based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses, the optimal treatment dose is calculated to obtain the optimal treatment dose of the target drug in each of the vascular perfusion zones.

[0007] The drug dosage planning method provided in this application can perform drug dosage planning. By using preset target constraints, preset lesion priority weights, and preset target drug doses, combined with the absorbed dose distribution data of the target drug determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, the optimal therapeutic dose is calculated to obtain the optimal therapeutic dose of the target drug in each vascular perfusion zone. This solves the problem that existing radioembolization treatment methods cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion when using traditional zoning models. It can provide voxel-level precision dose assessment, accurately calculate the optimal therapeutic dose of the target drug in different vascular perfusion zones, and improve the drug dosage planning accuracy of radioembolization treatment.

[0008] Optionally, the subject's scanned image information is acquired, including: Acquire fused images of lung perfusion and anatomical information of the subjects, as well as contrast-enhanced computed tomography images of the subjects; The fused image and the contrast-enhanced computed tomography image are deformably registered to obtain the scan image information of the subject.

[0009] Optionally, voxel activity quantification is performed on each of the aforementioned vascular perfusion zones to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each of the aforementioned vascular perfusion zones, including: The scanned image information is subjected to quantitative correction processing to obtain the activity concentration data of each unit voxel in the scanned image information; Based on the activity concentration data and the number of voxels contained in each of the vascular perfusion zones, the voxel activity concentration distribution data of each of the vascular perfusion zones is calculated. Based on the voxel activity concentration distribution data, lesion calculations are performed to obtain the tumor-to-normal tissue activity concentration ratio data for each of the vascular perfusion zones.

[0010] Optionally, quantitative correction processing is performed on the scanned image information to obtain the activity concentration data of each voxel in the scanned image information, including: The scanned image information is subjected to attenuation correction processing to obtain attenuation-corrected scanned image information; The attenuation-corrected scanned image information is subjected to scattering correction processing to obtain scattering-corrected scanned image information; The scattering-corrected scanned image information is subjected to resolution restoration processing to obtain the resolution-restored scanned image information; Using a preset calibration factor, photon counting processing is performed on the scanned image information after resolution restoration to obtain the activity concentration data of each unit voxel in the scanned image information.

[0011] Optionally, the vascular perfusion zone includes a lesion zone and a corresponding feeding artery perfusion zone; based on the voxel activity concentration distribution data, lesion calculation is performed to obtain the tumor-to-normal tissue activity concentration ratio data for each of the vascular perfusion zones, including: Based on the number of voxels contained in each lesion region and the number of voxels contained in each blood supply artery perfusion region in the voxel activity concentration distribution data, the lesion region activity concentration and the blood supply artery perfusion region activity concentration of each lesion region and each blood supply artery perfusion region are calculated. The percentage of the active concentration of the lesion zone in all vascular perfusion zones is determined as the lesion zone uptake fraction of each lesion zone, and the percentage of the active concentration of the feeding artery perfusion zone in all vascular perfusion zones is determined as the feeding artery perfusion zone uptake fraction of each feeding artery perfusion zone. Based on the uptake fraction of the lesion zone and the uptake fraction of the blood supply artery perfusion zone, a mass ratio is calculated to obtain the tumor-to-normal tissue active concentration ratio data for each of the blood vessel perfusion zones.

[0012] The drug dosage planning method provided in this application can realize drug dosage planning. By distinguishing between lesion zones and blood supply artery perfusion zones and calculating their respective active concentrations, the corresponding uptake fraction is calculated, thereby obtaining tumor-to-normal tissue active concentration ratio data, thus realizing independent assessment of different lesions.

[0013] Optionally, based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses, the optimal treatment dose is calculated to obtain the optimal treatment dose of the target drug in each of the aforementioned vascular perfusion zones, including: Obtain a preset optimal treatment dose calculation function; the optimal treatment dose calculation function is a target function pre-constructed based on the preset target constraints, the preset lesion priority weights, and the preset target drug target dose; Based on the absorbed dose distribution data, and combined with the perfusion overlap area in each of the vascular perfusion zones, the actual drug absorbed dose is calculated to obtain the actual drug absorbed dose of each of the vascular perfusion zones. Based on the actual drug absorption dose, the optimal therapeutic dose of the target drug in each of the vascular perfusion zones is calculated using the optimal therapeutic dose calculation function.

[0014] Optionally, based on the absorbed dose distribution data and combined with the perfusion overlap area in each of the vascular perfusion zones, the actual drug absorbed dose is calculated to obtain the actual drug absorbed dose of each of the vascular perfusion zones, including: From the scanned image information, the perfusion overlap region between each of the vascular perfusion zones is determined; the perfusion overlap region is the area where the lesion zone in the vascular perfusion zone overlaps with the blood supply artery perfusion zone in other vascular perfusion zones; Determine the volume percentage of each of the perfusion overlap regions in each of the vascular perfusion zones; Based on the volume ratio of the overlapping region and the absorbed dose distribution data, the self-absorbed dose of each of the vascular perfusion zones and the cross-absorbed dose of the corresponding perfusion overlapping region are calculated to obtain the actual drug absorbed dose of each of the vascular perfusion zones.

[0015] The drug dosage planning method provided in this application can realize drug dosage planning. By identifying and quantifying the perfusion overlap area and calculating its contribution to the actual drug absorption dose, the accuracy of dosage calculation is improved, making drug dosage planning applicable to multi-vessel fractionated drug administration.

[0016] Secondly, this application provides a drug dosage planning device, comprising: The acquisition module is used to acquire the scanned image information of the examinee; A segmentation module is used to divide the scanned image information into regions to obtain vascular perfusion zones; The quantification module is used to quantify the voxel activity of each of the aforementioned vascular perfusion zones, and obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data of each of the aforementioned vascular perfusion zones. The prediction module is used to predict the absorbed dose based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data, so as to obtain the absorbed dose distribution data of the target drug. The calculation module is used to calculate the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses, so as to obtain the optimal treatment dose of the target drug in each of the vascular perfusion zones.

[0017] This drug dosage planning device, through preset target constraints, preset lesion priority weights, and preset target drug doses, combined with the absorbed dose distribution data of the target drug determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, calculates the optimal treatment dose to obtain the optimal treatment dose of the target drug in each vascular perfusion zone. This solves the problem that existing radioembolization treatment methods using traditional zoning models cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion. It can provide voxel-level precision dose assessment, accurately calculate the optimal treatment dose of the target drug in different vascular perfusion zones, and improve the drug dosage planning accuracy of radioembolization treatment.

[0018] Thirdly, this application provides an electronic device including a processor and a memory, the memory storing a computer program executable by the processor, wherein when the processor executes the computer program, it performs the steps in the drug dosage planning method described above.

[0019] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the drug dosage planning method described above.

[0020] Beneficial effects: The drug dosage planning method, device, electronic equipment, and storage medium provided in this application calculate the optimal therapeutic dose by combining preset target constraints, preset lesion priority weights, and preset target drug target doses with target drug absorption dose distribution data determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data. This yields the optimal therapeutic dose of the target drug in each vascular perfusion zone, solving the problem that existing radioembolization treatment methods cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion when using traditional zoning models. It can provide voxel-level precision dose assessment, accurately calculate the optimal therapeutic dose of the target drug in different vascular perfusion zones, and improve the accuracy of drug dosage planning in radioembolization treatment. Attached Figure Description

[0021] Figure 1 A flowchart of a drug dosage planning method provided in an embodiment of this application.

[0022] Figure 2 This is a schematic diagram of the structure of the drug dosage planning device provided in the embodiments of this application.

[0023] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0024] Labeling Explanation: 1. Acquisition Module; 2. Division Module; 3. Quantization Module; 4. Prediction Module; 5. Calculation Module; 301. Processor; 302. Memory; 303. Communication Bus. Detailed Implementation

[0025] 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 a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0026] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0027] Please refer to Figure 1 , Figure 1 This application provides a drug dosage planning method in some embodiments, used for drug dosage planning, comprising the following steps: Step S101: Obtain the scanned image information of the subject; Step S102: Divide the scanned image information into regions to obtain vascular perfusion zones; Step S103: Voxel activity quantification is performed on each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone. Step S104: Based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data, the absorbed dose is predicted to obtain the absorbed dose distribution data of the target drug. Step S105: Based on the absorbed dose distribution data, combined with the preset target constraints, preset lesion priority weights and preset target drug target doses, the optimal treatment dose is calculated to obtain the optimal treatment dose of the target drug in each vascular perfusion zone.

[0028] This drug dosage planning method, through preset target constraints, preset lesion priority weights, and preset target drug doses, combined with the absorbed dose distribution data of the target drug determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, calculates the optimal treatment dose to obtain the optimal treatment dose of the target drug in each vascular perfusion zone. This solves the problem that existing radioembolization treatment methods using traditional zoning models cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion. It can provide voxel-level precision dose assessment, accurately calculate the optimal treatment dose of the target drug in different vascular perfusion zones, and improve the accuracy of drug dosage planning in radioembolization treatment.

[0029] Specifically, in step S101, the scanned image information of the subject is acquired, including: Acquire fused images of lung perfusion and anatomical information of the subject (i.e., SPECT / CT images), and acquire contrast-enhanced computed tomography images of the subject (i.e., CECT images). Deformation registration was performed on the fused image and the contrast-enhanced computed tomography image to obtain the subject's scan image information.

[0030] In step S101, lung perfusion information can be acquired using single-photon emission computed tomography (SPECT) technology, and anatomical structure information can be acquired using computed tomography (CT) technology. These two technologies are then fused to obtain a fused image of lung perfusion and anatomical structure information, namely a SPECT / CT image. This fused image provides functional information (perfusion status) and structural information (anatomical details) of the lungs, including three-dimensional distribution data of the radioactive tracer in various organs of the body, providing fundamental data for subsequent region division and activity quantification.

[0031] Contrast-enhanced computed tomography (CECT) images, also known as contrast agent-enhanced computed tomography (CT) images, can be obtained by intravenously injecting a contrast agent. CECT images can clearly show the blood supply to vascular structures and lesions, and are of great significance for identifying vascular perfusion zones and lesion areas.

[0032] Existing image processing algorithms are used to spatially align images acquired from different modalities or at different time points to eliminate deformations caused by factors such as subject positioning and respiratory movements. The purpose of deformation registration is to ensure accurate matching of corresponding anatomical structures and functional regions in different images, thereby guaranteeing the accuracy of subsequent analysis. For example, registration can be performed using methods based on feature point matching, mutual information maximization, or deformation field estimation.

[0033] Specifically, in step S102, the scanned image information is divided into regions to obtain at least one vascular perfusion zone. This vascular perfusion zone includes not only lesion regions (i.e., diseased areas, such as the area containing tumor tissue) and the arterial perfusion regions supplying these lesions (i.e., the arteries providing blood supply to the lesion zones and the areas they perfuse, including normal liver tissue surrounding the lesion zones), but also healthy organ regions (i.e., organ tissues not perfused by any lesion-supplying arteries), such as non-target liver tissue regions (i.e., liver tissue not perfused by any lesion-supplying arteries), lung tissue regions (i.e., lung tissue not perfused by any lesion-supplying arteries), and other regions (i.e., other organ regions not perfused by any lesion-supplying arteries, such as the gastrointestinal tract). These vascular perfusion zones are crucial for achieving individualized dosing planning, allowing for independent analysis and processing of different regions.

[0034] For example, experienced doctors can manually delineate different regions within the liver based on historical scan images. Then, using existing machine learning algorithms (such as neural networks), they can build and train a region segmentation model based on the feature information of different regions. This allows the trained region segmentation model to automatically identify and segment vascular perfusion zones based on scan image information.

[0035] Specifically, in step S103, voxel activity quantification is performed on each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone, including: Quantitative correction processing is performed on the scanned image information to obtain the activity concentration data of each unit voxel in the scanned image information; Based on the active concentration data and the number of voxels contained in each vascular perfusion zone, the voxel active concentration distribution data of each vascular perfusion zone are calculated. Based on voxel activity concentration distribution data, lesion calculations were performed to obtain the tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone.

[0036] Specifically, in step S103, quantitative correction processing is performed on the scanned image information to obtain the activity concentration data of each voxel in the scanned image information, including: Attenuation correction is performed on the scanned image information to obtain attenuation-corrected scanned image information; The attenuation-corrected scanned image information is subjected to scattering correction processing to obtain the scattering-corrected scanned image information. The resolution-restored scanned image information after scattering correction is processed to obtain the resolution-restored scanned image information. Using a preset calibration factor, photon counting is performed on the scanned image information after resolution restoration to obtain the activity concentration data of each unit voxel in the scanned image information.

[0037] In step S103, attenuation correction processing refers to processing the scanned image information to compensate for signal loss caused by the attenuation of radionuclides within the subject's body. In practical applications, gamma rays undergo absorption and scattering when penetrating human tissue, resulting in a signal intensity received by the detector that is lower than the actual radioactivity. The purpose of attenuation correction is to eliminate this attenuation effect, ensuring that the signal intensity of radioactivity in the image accurately reflects the distribution of radionuclides within the body, and avoiding underestimation of the radioactivity in deep tissues. For example, information provided by the subject's CT images can be used to calculate the attenuation coefficients of different tissues for gamma rays, and voxel-by-voxel attenuation compensation can be performed on the SPECT images.

[0038] Scattering correction refers to the processing of attenuation-corrected scanned image information to eliminate artifacts caused by scattering effects. When gamma photons interact with tissues in vivo, their direction and energy may change. These scattered photons are incorrectly recorded by the detector, thus reducing image contrast and quantitative accuracy. Scattering correction improves the signal-to-noise ratio and quantitative accuracy of the image by identifying and removing these scattered signals, reducing image background noise and artifacts, resulting in purer activity concentration data for each voxel. For example, scattering correction can be performed using dual-energy window methods, Monte Carlo simulations, or model-based methods.

[0039] Resolution restoration processing refers to the processing of scatter-corrected scanned image information to compensate for the inherent spatial resolution limitations of the imaging system. Due to the physical characteristics of the detector and collimator, as well as partial volume effects, the original image is often blurry and cannot accurately reflect the true distribution of radionuclides. Resolution restoration processing aims to improve image sharpness and detail, making details in the image clearer and boundaries sharper, thereby more accurately locating the distribution of radionuclides and avoiding activity dilution caused by partial volume effects. For example, iterative reconstruction algorithms (such as the OSEM algorithm) combined with the point spread function (PSF) model can be used for resolution restoration.

[0040] Photon counting processing refers to the process of processing the scanned image information after resolution restoration using a preset calibration factor to convert the image pixel values ​​into actual activity concentration data. The calibration factor is obtained in advance by scanning and measuring a known activity standard source, establishing a quantitative relationship between image pixel values ​​and actual radioactivity. By applying this calibration factor, the image signal, after the aforementioned series of corrections, can be converted into physically meaningful radioactivity units (e.g., Bq / mL), i.e., activity concentration data, thereby achieving absolute quantification of the radiotracer (radiopharmaceutical) within each unit voxel.

[0041] Attenuation correction processing, scattering correction processing, resolution restoration processing, and photon counting processing are existing technologies and will not be described in detail here.

[0042] Quantitative correction processing (i.e., attenuation correction processing, scattering correction processing, resolution restoration processing, and photon counting processing) can eliminate or reduce various errors that may be introduced during the scanning process, so as to ensure that the obtained active concentration data can accurately reflect the true distribution of radiopharmaceuticals in the subject's body.

[0043] Specifically, in step S103, the active concentration data of each unit voxel is combined with the number of voxels in the corresponding vascular perfusion zone to quantify the active concentration of radiopharmaceutical in each voxel in each vascular perfusion zone, thereby obtaining the voxel active concentration distribution data of each vascular perfusion zone, which truly reflects the distribution of the drug in different regions and provides basic data for subsequent lesion analysis.

[0044] The total voxel activity concentration in each vascular perfusion zone can be calculated based on the voxel activity concentration distribution data. The specific formula for calculating the total voxel activity concentration is as follows: ; in, denoted as the total voxel activity concentration in the i-th vascular perfusion zone; x represents the x-th unit voxel. The region is defined for the i-th vascular perfusion zone; This represents the activity concentration data for the xth voxel; The volume of a unit voxel.

[0045] Specifically, in step S103, based on the voxel activity concentration distribution data, lesion calculation is performed to obtain the tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone, including: Based on the number of voxels contained in each lesion zone and the number of voxels contained in each blood supply artery perfusion zone in the voxel activity concentration distribution data, the lesion zone activity concentration and the blood supply artery perfusion zone activity concentration of each lesion zone are calculated. The proportion of the first active concentration of the lesion zone in all vascular perfusion zones is determined as the lesion zone uptake fraction of each lesion zone, and the proportion of the second active concentration of the feeding artery perfusion zone in all vascular perfusion zones is determined as the feeding artery perfusion zone uptake fraction of each feeding artery perfusion zone. Based on the uptake fraction of the lesion zone and the uptake fraction of the perfusion zone of the feeding artery, the quality ratio is calculated to obtain the tumor-to-normal tissue active concentration ratio data for each vascular perfusion zone.

[0046] In step S103, based on the number of voxels contained in each partition (lesion partition or feeding artery perfusion partition), the activity concentration values ​​of all voxels within the partition are summed to calculate the activity concentration of each partition, i.e., the lesion partition activity concentration of each lesion partition and the feeding artery perfusion partition activity concentration of each feeding artery perfusion partition. By quantifying the activity concentration of each lesion partition and its corresponding feeding artery perfusion partition, a preliminary distinction between different lesions and their blood supply conditions is achieved, laying a data foundation for subsequent refined calculations.

[0047] The uptake fraction refers to the ability of each zone to take up the radiotracer, usually expressed as the percentage of active concentration. One method is to use the percentage of the active concentration of each zone in all vascular perfusion zones as the uptake fraction. Therefore, the percentage of active concentration corresponding to each zone is calculated (i.e., the first percentage of active concentration of the lesion zone in all vascular perfusion zones and the second percentage of active concentration of the feeding artery zone in all vascular perfusion zones), and this calculated percentage is determined as the uptake fraction for each zone (i.e., the lesion zone uptake fraction and the feeding artery zone uptake fraction). Quantifying the relative active contribution of each lesion and its feeding artery in all vascular perfusion zones as an uptake fraction helps to more accurately assess the distribution of the drug in different lesions and feeding arteries.

[0048] The intake score can be obtained through the intake score calculation formula, which is as follows: ; in, The uptake fraction of partition c (the lesion partition or the perfusion partition of the feeding artery in the c-th vascular perfusion partition); The region divided by partition c; This represents the total area of ​​all vascular perfusion zones.

[0049] The average active concentration of the tumor is calculated based on the ratio of the uptake fraction of the lesion zone to the tumor mass within the corresponding lesion zone (the volume of the lesion zone (number of voxels contained) multiplied by the density of the corresponding tissue). Similarly, the average active concentration of the normal tissue within the corresponding feeding artery perfusion zone is calculated based on the ratio of the uptake fraction of the feeding artery perfusion zone to the normal tissue mass within the corresponding feeding artery perfusion zone (the volume of the feeding artery perfusion zone (number of voxels contained) multiplied by the density of the corresponding tissue). By using the ratio of the average active concentration of the tumor to the average active concentration of the normal tissue belonging to the same vascular perfusion zone, the enrichment degree of the target drug in different lesion regions can be more accurately reflected. This allows for the determination of the tumor-to-normal tissue active concentration ratio data for vascular perfusion zones, providing a more reliable basis for subsequent absorption dose prediction and optimal therapeutic dose calculation.

[0050] The specific formula for calculating the tumor-to-normal tissue active concentration ratio is as follows: ; in, The tumor-to-normal tissue active concentration ratio data for the i-th vascular perfusion zone; The lesion uptake score is the lesion region score in the i-th vascular perfusion region. The tumor quality of the lesion zone in the i-th vascular perfusion zone; The uptake score of the feeding artery perfusion zone in the i-th vascular perfusion zone; This represents the normal tissue quality of the blood supply artery perfusion zone within the i-th vascular perfusion zone.

[0051] Specifically, in step S104, based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, absorbed dose prediction is performed to calculate the absorbed dose (i.e., unit activity dose) generated by a unit of administered activity (e.g., per GBq or per mCi) for each lesion region, thus obtaining the absorbed dose distribution data of the target drug. For example, the Monte Carlo simulation method can be used, combined with the physical decay characteristics of the radiopharmaceutical and the anatomical information of the subject, to simulate the transport process of radioactive particles in the body, thereby calculating the absorbed dose for each voxel or each vascular perfusion region. Alternatively, the S-factor method can be used to estimate the absorbed dose based on parameters such as voxel activity concentration and tissue density. The absorbed dose distribution data is a core output for evaluating the effectiveness and safety of the treatment regimen, as it intuitively reflects the energy deposition of the radiopharmaceutical in the body. The Monte Carlo simulation method and the S-factor method are existing technologies and will not be described in detail here.

[0052] Specifically, in step S105, based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses, the optimal treatment dose is calculated to obtain the optimal treatment dose of the target drug in each vascular perfusion zone, including: Obtain the preset optimal treatment dose calculation function; the optimal treatment dose calculation function is a pre-constructed objective function based on preset target constraints, preset lesion priority weights, and preset target drug target doses; Based on the absorbed dose distribution data, combined with the perfusion overlap area in each vascular perfusion zone, the actual drug absorbed dose is calculated to obtain the actual drug absorbed dose of each vascular perfusion zone. Based on the actual drug absorption dose, the optimal therapeutic dose of the target drug in each vascular perfusion zone is calculated using the optimal therapeutic dose calculation function.

[0053] In step S105, the preset target constraints are multiple restrictions that need to be met during drug dosage planning. For example, they may include dosage constraints for healthy organs (such as dosage constraints for a normal liver and normal lungs), the minimum effective dose for the lesion area, and the upper limit of drug concentration (upper limit of activity). The preset target constraints can ensure the effectiveness and safety of treatment.

[0054] Preset lesion priority weights refer to the relative importance coefficients assigned to different lesions based on factors such as their type, size, location, malignancy, or impact on the patient's life and health. For example, life-threatening lesions can be assigned higher priority weights to ensure they receive priority and more aggressive treatment.

[0055] The preset target drug dose refers to the range or specific value of the target drug dose that is set at the beginning of the treatment plan. It can be set according to expert opinions or relevant treatment standards and can provide a benchmark or reference for subsequent calculation of the optimal treatment dose.

[0056] Based on satisfying the target constraints, considering lesion priority weights, and the target drug dose, an optimal treatment dose calculation function is constructed. Using this function, the optimal drug dose for each vascular perfusion zone can be calculated, thus finding an optimal dose allocation scheme under complex physiological and therapeutic constraints.

[0057] The specific function for calculating the optimal treatment dose is as follows: ; in, This represents the optimal dose allocation scheme; n is the total number of vascular perfusion zones; The lesion priority weight is the lesion partition in the i-th vascular perfusion partition; This represents the actual drug absorption dose in the lesion zone within the i-th vascular perfusion zone; The target drug dose is the target dose for the lesion zone in the i-th vascular perfusion zone.

[0058] This optimal treatment dose calculation function, based on all preset optimization conditions, calculates the minimum treatment dose of the target drug during treatment, so that the actual treatment dose of the lesion zone in each vascular perfusion zone is as close as possible to the preset target drug dose, thereby obtaining the optimal treatment dose of the target drug in each vascular perfusion zone.

[0059] The specific formula for calculating the actual drug absorption dose is as follows: ; ; in, This is the unit activity dose of the lesion zone in the i-th vascular perfusion zone (which can be extracted from the absorbed dose distribution data); is the cross-perfusion coefficient between the feeding artery perfusion zone in the j-th vascular perfusion zone and the lesion zone in the i-th vascular perfusion zone; The total voxel activity concentration of the j-th vascular perfusion zone; The number of voxels contained in the lesion region within the i-th vascular perfusion region; The number of voxels contained in the perfusion zone of the feeding artery in the j-th vascular perfusion zone; This represents the dose contribution of the lesion region to the i-th vascular perfusion region; This represents the dose contribution of the feeding artery perfusion zone in the j-th vascular perfusion zone (other vascular perfusion zones) to the i-th vascular perfusion zone.

[0060] Specifically, in step S105, based on the absorbed dose distribution data and combined with the perfusion overlap area in each vascular perfusion zone, the actual absorbed dose is calculated to obtain the actual absorbed dose of each vascular perfusion zone, including: From the scanned image information, the perfusion overlap area between each vascular perfusion zone is determined; the perfusion overlap area is the area where the lesion zone in a vascular perfusion zone overlaps with the perfusion zone of the feeding artery in other vascular perfusion zones; Determine the volume percentage of each overlapping perfusion region in each vascular perfusion zone; Based on the volume ratio of the overlapping area and the absorption dose distribution data, the self-absorbed dose of each vascular perfusion zone and the cross-absorbed dose of the corresponding perfusion overlapping area are calculated to obtain the actual absorbed dose of each vascular perfusion zone.

[0061] In step S105, existing image recognition technology is used to perform detailed analysis and image segmentation of the subject's scanned image information to identify overlapping areas in three-dimensional space between different vascular perfusion zones. The perfusion overlap area is the location where a lesion zone within one vascular perfusion zone overlaps with a feeding artery perfusion zone within another vascular perfusion zone. This overlap may mean that the distribution and absorption of the drug in these areas are affected by multiple blood supply sources, and therefore requires special consideration in dosage calculation.

[0062] After identifying the perfusion overlap regions, the proportion of these overlap regions within their respective vascular perfusion zones is quantified to obtain the volume percentage of each overlap region within its respective vascular perfusion zone. For example, this can be obtained by calculating the ratio of the number of voxels in the overlap region to the total number of voxels in the entire vascular perfusion zone. The volume percentage of the overlap region provides a quantified weighting parameter for subsequent absorbed dose calculations, accurately reflecting the contribution of the overlap region to the total absorbed dose.

[0063] After considering the volume ratio of the perfusion overlap area, the absorbed dose within each vascular perfusion zone is divided into two parts: one part is the dose absorbed by the zone itself (self-absorbed dose), and the other part is the additional absorption or influence caused by overlap with other zones (cross-absorbed dose). The sum of these two parts is then used to determine the actual absorbed dose of the vascular perfusion zone (which can be calculated using the formula for actual drug absorption dose mentioned above). This method more accurately reflects the actual absorption of the drug within each vascular perfusion zone, especially in areas with complex perfusion overlap.

[0064] Specifically, in step S105, the actual drug absorption dose of each vascular perfusion zone is optimized using a pre-constructed optimal treatment dose calculation function. This function can adjust the actual drug absorption dose of each vascular perfusion zone based on all preset optimization conditions, so that the actual drug absorption dose of all lesion zones is as close as possible to the preset target drug target dose. Thus, the actual drug absorption dose of each vascular perfusion zone when the optimal treatment dose calculation function is at its minimum value is determined as the optimal treatment dose of the target drug in each vascular perfusion zone.

[0065] As described above, this drug dosage planning method acquires the subject's scanned image information, divides the scanned image information into regions to obtain vascular perfusion zones, quantifies voxel activity in each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone, predicts the absorbed dose based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data to obtain the absorbed dose distribution data of the target drug, and calculates the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses to obtain the optimal treatment dose of the target drug in each vascular perfusion zone. The optimal therapeutic dose is calculated by combining preset target constraints, preset lesion priority weights, and preset target drug doses with target drug absorption dose distribution data determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data. This yields the optimal therapeutic dose of the target drug in each vascular perfusion zone, solving the problem that existing radioembolization treatment methods using traditional zoning models cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion. It can provide voxel-level precision dose assessment, accurately calculate the optimal therapeutic dose of the target drug in different vascular perfusion zones, and improve the accuracy of drug dose planning in radioembolization treatment.

[0066] refer to Figure 2 This application provides a drug dosage planning device for drug dosage planning, comprising: Module 1 is used to acquire scanned image information of the subject; The segmentation module 2 is used to divide the scanned image information into regions to obtain vascular perfusion zones; Quantization module 3 is used to quantify voxel activity in each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data in each vascular perfusion zone. Prediction module 4 is used to predict the absorbed dose based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, and obtain the absorbed dose distribution data of the target drug. The calculation module 5 is used to calculate the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights and preset target drug target doses, so as to obtain the optimal treatment dose of the target drug in each vascular perfusion zone.

[0067] This drug dosage planning device, through preset target constraints, preset lesion priority weights, and preset target drug doses, combined with the absorbed dose distribution data of the target drug determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data, calculates the optimal treatment dose to obtain the optimal treatment dose of the target drug in each vascular perfusion zone. This solves the problem that existing radioembolization treatment methods using traditional zoning models cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion. It can provide voxel-level precision dose assessment, accurately calculate the optimal treatment dose of the target drug in different vascular perfusion zones, and improve the drug dosage planning accuracy of radioembolization treatment.

[0068] Specifically, when acquiring the scanned image information of the subject, module 1 performs the following: Acquire fused images of lung perfusion and anatomical information of the subject (i.e., SPECT / CT images), and acquire contrast-enhanced computed tomography images of the subject (i.e., CECT images). Deformation registration was performed on the fused image and the contrast-enhanced computed tomography image to obtain the subject's scan image information.

[0069] When module 1 is executed, it can acquire lung perfusion information using single-photon emission computed tomography (SPECT) and anatomical structure information using computed tomography (CT). These two technologies are then fused to obtain a fused image of lung perfusion and anatomical structure information, i.e., a SPECT / CT image. This fused image provides functional information (perfusion status) and structural information (anatomical details) of the lungs, including three-dimensional distribution data of the radioactive tracer in various organs of the body, providing foundational data for subsequent region segmentation and activity quantification.

[0070] Contrast-enhanced computed tomography (CECT) images, also known as contrast agent-enhanced computed tomography (CT) images, can be obtained by intravenously injecting a contrast agent. CECT images can clearly show the blood supply to vascular structures and lesions, and are of great significance for identifying vascular perfusion zones and lesion areas.

[0071] Existing image processing algorithms are used to spatially align images acquired from different modalities or at different time points to eliminate deformations caused by factors such as subject positioning and respiratory movements. The purpose of deformation registration is to ensure accurate matching of corresponding anatomical structures and functional regions in different images, thereby guaranteeing the accuracy of subsequent analysis. For example, registration can be performed using methods based on feature point matching, mutual information maximization, or deformation field estimation.

[0072] Specifically, during execution, segmentation module 2 divides the scanned image information into regions, obtaining at least one vascular perfusion zone. This vascular perfusion zone includes not only lesion regions (i.e., diseased areas, such as the area containing tumor tissue) and the arterial perfusion regions supplying these lesions (i.e., the arteries providing blood supply to the lesion zones and the areas they perfuse, including normal liver tissue surrounding the lesion regions), but also healthy organ regions (i.e., organ tissues not perfused by any lesion-supplying arteries), such as non-target liver tissue regions (i.e., liver tissue not perfused by any lesion-supplying arteries), lung tissue regions (i.e., lung tissue not perfused by any lesion-supplying arteries), and other regions (i.e., other organ regions not perfused by any lesion-supplying arteries, such as the gastrointestinal tract). These vascular perfusion zones are crucial for achieving individualized dosing planning, allowing for independent analysis and processing of different regions.

[0073] For example, experienced doctors can manually delineate different regions within the liver based on historical scan images. Then, using existing machine learning algorithms (such as neural networks), they can build and train a region segmentation model based on the feature information of different regions. This allows the trained region segmentation model to automatically identify and segment vascular perfusion zones based on scan image information.

[0074] Specifically, when quantification module 3 performs voxel activity quantification on each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone, it executes the following: Quantitative correction processing is performed on the scanned image information to obtain the activity concentration data of each unit voxel in the scanned image information; Based on the active concentration data and the number of voxels contained in each vascular perfusion zone, the voxel active concentration distribution data of each vascular perfusion zone are calculated. Based on voxel activity concentration distribution data, lesion calculations were performed to obtain the tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone.

[0075] Specifically, when quantization module 3 performs quantitative correction processing on the scanned image information to obtain the activity concentration data of each unit voxel in the scanned image information, it executes: Attenuation correction is performed on the scanned image information to obtain attenuation-corrected scanned image information; The attenuation-corrected scanned image information is subjected to scattering correction processing to obtain the scattering-corrected scanned image information. The resolution-restored scanned image information after scattering correction is processed to obtain the resolution-restored scanned image information. Using a preset calibration factor, photon counting is performed on the scanned image information after resolution restoration to obtain the activity concentration data of each unit voxel in the scanned image information.

[0076] During the execution of quantization module 3, attenuation correction processing refers to the processing of scanned image information to compensate for signal loss caused by the attenuation of radionuclides within the subject's body. In practical applications, gamma rays undergo absorption and scattering when penetrating human tissue, resulting in a signal intensity received by the detector that is lower than the actual radioactivity. The purpose of attenuation correction is to eliminate this attenuation effect, ensuring that the signal intensity of radioactivity in the image accurately reflects the distribution of radionuclides within the body, and avoiding underestimation of the radioactivity in deep tissues. For example, information provided by the subject's CT images can be used to calculate the attenuation coefficients of different tissues for gamma rays, and voxel-by-voxel attenuation compensation can be performed on the SPECT images.

[0077] Scattering correction refers to the processing of attenuation-corrected scanned image information to eliminate artifacts caused by scattering effects. When gamma photons interact with tissues in vivo, their direction and energy may change. These scattered photons are incorrectly recorded by the detector, thus reducing image contrast and quantitative accuracy. Scattering correction improves the signal-to-noise ratio and quantitative accuracy of the image by identifying and removing these scattered signals, reducing image background noise and artifacts, resulting in purer activity concentration data for each voxel. For example, scattering correction can be performed using dual-energy window methods, Monte Carlo simulations, or model-based methods.

[0078] Resolution restoration processing refers to the processing of scatter-corrected scanned image information to compensate for the inherent spatial resolution limitations of the imaging system. Due to the physical characteristics of the detector and collimator, as well as partial volume effects, the original image is often blurry and cannot accurately reflect the true distribution of radionuclides. Resolution restoration processing aims to improve image sharpness and detail, making details in the image clearer and boundaries sharper, thereby more accurately locating the distribution of radionuclides and avoiding activity dilution caused by partial volume effects. For example, iterative reconstruction algorithms (such as the OSEM algorithm) combined with the point spread function (PSF) model can be used for resolution restoration.

[0079] Photon counting processing refers to the process of processing the scanned image information after resolution restoration using a preset calibration factor to convert the image pixel values ​​into actual activity concentration data. The calibration factor is obtained in advance by scanning and measuring a known activity standard source, establishing a quantitative relationship between image pixel values ​​and actual radioactivity. By applying this calibration factor, the image signal, after the aforementioned series of corrections, can be converted into physically meaningful radioactivity units (e.g., Bq / mL), i.e., activity concentration data, thereby achieving absolute quantification of the radiotracer (radiopharmaceutical) within each unit voxel.

[0080] Attenuation correction processing, scattering correction processing, resolution restoration processing, and photon counting processing are existing technologies and will not be described in detail here.

[0081] Quantitative correction processing (i.e., attenuation correction processing, scattering correction processing, resolution restoration processing, and photon counting processing) can eliminate or reduce various errors that may be introduced during the scanning process, so as to ensure that the obtained active concentration data can accurately reflect the true distribution of radiopharmaceuticals in the subject's body.

[0082] Specifically, when the quantification module 3 is executed, it combines the active concentration data of each unit voxel with the number of voxels in the corresponding vascular perfusion zone to quantify the active concentration of radiopharmaceuticals in each voxel in each vascular perfusion zone, thereby obtaining the voxel active concentration distribution data of each vascular perfusion zone. This truly reflects the distribution of the drug in different regions and provides basic data for subsequent lesion analysis.

[0083] The total voxel activity concentration in each vascular perfusion zone can be calculated based on the voxel activity concentration distribution data. The specific formula for calculating the total voxel activity concentration is as follows: ; in, denoted as the total voxel activity concentration in the i-th vascular perfusion zone; x represents the x-th unit voxel. The region is defined for the i-th vascular perfusion zone; This represents the activity concentration data for the xth voxel; The volume of a unit voxel.

[0084] Specifically, when quantification module 3 calculates lesions based on voxel activity concentration distribution data and obtains the tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone, it executes the following: Based on the number of voxels contained in each lesion zone and the number of voxels contained in each blood supply artery perfusion zone in the voxel activity concentration distribution data, the lesion zone activity concentration and the blood supply artery perfusion zone activity concentration of each lesion zone are calculated. The proportion of the first active concentration of the lesion zone in all vascular perfusion zones is determined as the lesion zone uptake fraction of each lesion zone, and the proportion of the second active concentration of the feeding artery perfusion zone in all vascular perfusion zones is determined as the feeding artery perfusion zone uptake fraction of each feeding artery perfusion zone. Based on the uptake fraction of the lesion zone and the uptake fraction of the perfusion zone of the feeding artery, the quality ratio is calculated to obtain the tumor-to-normal tissue active concentration ratio data for each vascular perfusion zone.

[0085] During execution, Quantization Module 3 sums the activity concentration values ​​of all voxels within each partition (lesion partition or feeding artery perfusion partition) based on the number of voxels contained in that partition, thus calculating the activity concentration of each partition. This yields the lesion partition activity concentration and the feeding artery perfusion partition activity concentration. By quantifying the activity concentration of each lesion partition and its corresponding feeding artery perfusion partition, a preliminary distinction is made between different lesions and their blood supply conditions, laying a data foundation for subsequent refined calculations.

[0086] The uptake fraction refers to the ability of each zone to take up the radiotracer, usually expressed as the percentage of active concentration. One method is to use the percentage of the active concentration of each zone in all vascular perfusion zones as the uptake fraction. Therefore, the percentage of active concentration corresponding to each zone is calculated (i.e., the first percentage of active concentration of the lesion zone in all vascular perfusion zones and the second percentage of active concentration of the feeding artery zone in all vascular perfusion zones), and this calculated percentage is determined as the uptake fraction for each zone (i.e., the lesion zone uptake fraction and the feeding artery zone uptake fraction). Quantifying the relative active contribution of each lesion and its feeding artery in all vascular perfusion zones as an uptake fraction helps to more accurately assess the distribution of the drug in different lesions and feeding arteries.

[0087] The intake score can be obtained through the intake score calculation formula, which is as follows: ; in, The uptake fraction of partition c (the lesion partition or the perfusion partition of the feeding artery in the c-th vascular perfusion partition); The region divided by partition c; This represents the total area of ​​all vascular perfusion zones.

[0088] The average active concentration of the tumor is calculated based on the ratio of the uptake fraction of the lesion zone to the tumor mass within the corresponding lesion zone (the volume of the lesion zone (number of voxels contained) multiplied by the density of the corresponding tissue). Similarly, the average active concentration of the normal tissue within the corresponding feeding artery perfusion zone is calculated based on the ratio of the uptake fraction of the feeding artery perfusion zone to the normal tissue mass within the corresponding feeding artery perfusion zone (the volume of the feeding artery perfusion zone (number of voxels contained) multiplied by the density of the corresponding tissue). By using the ratio of the average active concentration of the tumor to the average active concentration of the normal tissue belonging to the same vascular perfusion zone, the enrichment degree of the target drug in different lesion regions can be more accurately reflected. This allows for the determination of the tumor-to-normal tissue active concentration ratio data for vascular perfusion zones, providing a more reliable basis for subsequent absorption dose prediction and optimal therapeutic dose calculation.

[0089] The specific formula for calculating the tumor-to-normal tissue active concentration ratio is as follows: ; in, The tumor-to-normal tissue active concentration ratio data for the i-th vascular perfusion zone; The lesion uptake score is the lesion region score in the i-th vascular perfusion region. The tumor quality of the lesion zone in the i-th vascular perfusion zone; The uptake score of the feeding artery perfusion zone in the i-th vascular perfusion zone; This represents the normal tissue quality of the blood supply artery perfusion zone within the i-th vascular perfusion zone.

[0090] Specifically, during execution, prediction module 4 predicts absorbed dose based on voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data to calculate the absorbed dose (i.e., unit activity dose) generated by a unit of administered activity (e.g., per GBq or per mCi) for each lesion region, thus obtaining the absorbed dose distribution data of the target drug. For example, Monte Carlo simulation can be used, combining the physical decay characteristics of the radiopharmaceutical with the anatomical information of the subject, to simulate the transport process of radioactive particles in the body, thereby calculating the absorbed dose for each voxel or each vascular perfusion region. Alternatively, the S-factor method can be used to estimate the absorbed dose based on parameters such as voxel activity concentration and tissue density. The absorbed dose distribution data is a core output for evaluating the effectiveness and safety of the treatment regimen, as it intuitively reflects the energy deposition of the radiopharmaceutical in the body. The Monte Carlo simulation method and the S-factor method are existing technologies and will not be detailed here.

[0091] Specifically, when calculation module 5 calculates the optimal treatment dose based on absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug doses, and obtains the optimal treatment dose of the target drug in each vascular perfusion zone, it executes: Obtain the preset optimal treatment dose calculation function; the optimal treatment dose calculation function is a pre-constructed objective function based on preset target constraints, preset lesion priority weights, and preset target drug target doses; Based on the absorbed dose distribution data, combined with the perfusion overlap area in each vascular perfusion zone, the actual drug absorbed dose is calculated to obtain the actual drug absorbed dose of each vascular perfusion zone. Based on the actual drug absorption dose, the optimal therapeutic dose of the target drug in each vascular perfusion zone is calculated using the optimal therapeutic dose calculation function.

[0092] When the calculation module 5 is executed, the preset target constraints are multiple restrictions that need to be met during the drug dosage planning process. For example, these constraints may include dosage constraints for healthy organs (such as dosage constraints for normal liver and normal lungs), the minimum effective dose for the lesion area, and the upper limit of drug concentration (upper limit of activity). The preset target constraints ensure the effectiveness and safety of the treatment.

[0093] Preset lesion priority weights refer to the relative importance coefficients assigned to different lesions based on factors such as their type, size, location, malignancy, or impact on the patient's life and health. For example, life-threatening lesions can be assigned higher priority weights to ensure they receive priority and more aggressive treatment.

[0094] The preset target drug dose refers to the range or specific value of the target drug dose that is set at the beginning of the treatment plan. It can be set according to expert opinions or relevant treatment standards and can provide a benchmark or reference for subsequent calculation of the optimal treatment dose.

[0095] Based on satisfying the target constraints, considering lesion priority weights, and the target drug dose, an optimal treatment dose calculation function is constructed. Using this function, the optimal drug dose for each vascular perfusion zone can be calculated, thus finding an optimal dose allocation scheme under complex physiological and therapeutic constraints.

[0096] The specific function for calculating the optimal treatment dose is as follows: ; in, This represents the optimal dose allocation scheme; n is the total number of vascular perfusion zones; The lesion priority weight is the lesion partition in the i-th vascular perfusion partition; This represents the actual drug absorption dose in the lesion zone within the i-th vascular perfusion zone; The target drug dose is the target dose for the lesion zone in the i-th vascular perfusion zone.

[0097] This optimal treatment dose calculation function, based on all preset optimization conditions, calculates the minimum treatment dose of the target drug during treatment, so that the actual treatment dose of the lesion zone in each vascular perfusion zone is as close as possible to the preset target drug dose, thereby obtaining the optimal treatment dose of the target drug in each vascular perfusion zone.

[0098] The specific formula for calculating the actual drug absorption dose is as follows: ; ; in, This is the unit activity dose of the lesion zone in the i-th vascular perfusion zone (which can be extracted from the absorbed dose distribution data); is the cross-perfusion coefficient between the feeding artery perfusion zone in the j-th vascular perfusion zone and the lesion zone in the i-th vascular perfusion zone; The total voxel activity concentration of the j-th vascular perfusion zone; The number of voxels contained in the lesion region within the i-th vascular perfusion region; The number of voxels contained in the perfusion zone of the feeding artery in the j-th vascular perfusion zone; This represents the dose contribution of the lesion region to the i-th vascular perfusion region; This represents the dose contribution of the feeding artery perfusion zone in the j-th vascular perfusion zone (other vascular perfusion zones) to the i-th vascular perfusion zone.

[0099] Specifically, during execution, the calculation module 5 calculates the actual absorbed dose based on the absorbed dose distribution data and the perfusion overlap area in each vascular perfusion zone, obtaining the actual absorbed dose for each vascular perfusion zone, including: From the scanned image information, the perfusion overlap area between each vascular perfusion zone is determined; the perfusion overlap area is the area where the lesion zone in a vascular perfusion zone overlaps with the perfusion zone of the feeding artery in other vascular perfusion zones; Determine the volume percentage of each overlapping perfusion region in each vascular perfusion zone; Based on the volume ratio of the overlapping area and the absorption dose distribution data, the self-absorbed dose of each vascular perfusion zone and the cross-absorbed dose of the corresponding perfusion overlapping area are calculated to obtain the actual absorbed dose of each vascular perfusion zone.

[0100] During execution, the calculation module 5 utilizes existing image recognition technology to perform detailed analysis and image segmentation of the subject's scanned images, identifying overlapping areas in three-dimensional space between different vascular perfusion zones. These overlapping areas are defined as locations where a lesion zone within one vascular perfusion zone spatially overlaps with a feeding artery perfusion zone within another. This overlap may indicate that drug distribution and absorption in these areas are influenced by multiple blood supply sources, thus requiring special consideration in dosage calculations.

[0101] After identifying the perfusion overlap regions, the proportion of these overlap regions within their respective vascular perfusion zones is quantified to obtain the volume percentage of each overlap region within its respective vascular perfusion zone. For example, this can be obtained by calculating the ratio of the number of voxels in the overlap region to the total number of voxels in the entire vascular perfusion zone. The volume percentage of the overlap region provides a quantified weighting parameter for subsequent absorbed dose calculations, accurately reflecting the contribution of the overlap region to the total absorbed dose.

[0102] After considering the volume ratio of the perfusion overlap area, the absorbed dose within each vascular perfusion zone is divided into two parts: one part is the dose absorbed by the zone itself (self-absorbed dose), and the other part is the additional absorption or influence caused by overlap with other zones (cross-absorbed dose). The sum of these two parts is then used to determine the actual absorbed dose of the vascular perfusion zone (which can be calculated using the formula for actual drug absorption dose mentioned above). This method more accurately reflects the actual absorption of the drug within each vascular perfusion zone, especially in areas with complex perfusion overlap.

[0103] Specifically, when the calculation module 5 is executed, it uses a pre-built optimal treatment dose calculation function to optimize the actual drug absorption dose of each vascular perfusion zone. This allows the function to adjust the actual drug absorption dose of each vascular perfusion zone based on all preset optimization conditions, so that the actual drug absorption dose of all lesion zones is as close as possible to the preset target drug target dose. Thus, the actual drug absorption dose of each vascular perfusion zone when the optimal treatment dose calculation function is at its minimum value is determined as the optimal treatment dose of the target drug in each vascular perfusion zone.

[0104] As described above, this drug dosage planning device acquires the subject's scanned image information, divides the scanned image information into regions to obtain vascular perfusion zones, quantifies voxel activity in each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone, predicts the absorbed dose based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data to obtain the absorbed dose distribution data of the target drug, and calculates the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses to obtain the optimal treatment dose of the target drug in each vascular perfusion zone. The optimal therapeutic dose is calculated by combining preset target constraints, preset lesion priority weights, and preset target drug doses with target drug absorption dose distribution data determined based on voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data. This yields the optimal therapeutic dose of the target drug in each vascular perfusion zone, solving the problem that existing radioembolization treatment methods using traditional zoning models cannot distinguish the dose distribution of different lesions and cannot achieve independent prescription optimization for each lesion. It can provide voxel-level precision dose assessment, accurately calculate the optimal therapeutic dose of the target drug in different vascular perfusion zones, and improve the accuracy of drug dose planning in radioembolization treatment.

[0105] Please refer to Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device includes a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other via a communication bus 303 and / or other connection mechanisms (not shown). The memory 302 stores a computer program executable by the processor 301. When the electronic device is running, the processor 301 executes the computer program to perform the drug dosage planning method in any optional implementation of the above embodiments, to achieve the following functions: acquiring scanned image information of the subject; dividing the scanned image information into regions to obtain vascular perfusion zones; quantifying voxel activity in each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone; predicting the absorbed dose based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data to obtain the absorbed dose distribution data of the target drug; and calculating the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses to obtain the optimal treatment dose of the target drug in each vascular perfusion zone.

[0106] This application provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it executes the drug dosage planning method in any optional implementation of the above embodiments to achieve the following functions: acquiring scanned image information of the subject; dividing the scanned image information into regions to obtain vascular perfusion zones; quantifying voxel activity in each vascular perfusion zone to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each vascular perfusion zone; predicting the absorbed dose based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data to obtain the absorbed dose distribution data of the target drug; and calculating the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses to obtain the optimal treatment dose of the target drug in each vascular perfusion zone. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0107] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0108] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0109] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0110] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0111] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A drug dosage planning method, characterized in that, include: Obtain scanned image information of the examinee; The scanned image information is divided into regions to obtain vascular perfusion zones; Voxel activity was quantified in each of the aforementioned vascular perfusion zones to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each of the aforementioned vascular perfusion zones. Based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data, the absorbed dose is predicted to obtain the absorbed dose distribution data of the target drug. Based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses, the optimal treatment dose is calculated to obtain the optimal treatment dose of the target drug in each of the vascular perfusion zones.

2. The drug dosage planning method according to claim 1, characterized in that, Obtain the subject's scanned image information, including: Acquire fused images of lung perfusion and anatomical information of the subjects, as well as contrast-enhanced computed tomography images of the subjects; The fused image and the contrast-enhanced computed tomography image are deformably registered to obtain the scan image information of the subject.

3. The drug dosage planning method according to claim 1, characterized in that, Voxel activity quantification was performed on each of the aforementioned vascular perfusion zones to obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data for each of the aforementioned vascular perfusion zones, including: The scanned image information is subjected to quantitative correction processing to obtain the activity concentration data of each unit voxel in the scanned image information; Based on the activity concentration data and the number of voxels contained in each of the vascular perfusion zones, the voxel activity concentration distribution data of each of the vascular perfusion zones is calculated. Based on the voxel activity concentration distribution data, lesion calculations are performed to obtain the tumor-to-normal tissue activity concentration ratio data for each of the vascular perfusion zones.

4. The drug dosage planning method according to claim 3, characterized in that, The scanned image information is subjected to quantitative correction processing to obtain the activity concentration data of each voxel in the scanned image information, including: The scanned image information is subjected to attenuation correction processing to obtain attenuation-corrected scanned image information; The attenuation-corrected scanned image information is subjected to scattering correction processing to obtain scattering-corrected scanned image information; The scattering-corrected scanned image information is subjected to resolution restoration processing to obtain the resolution-restored scanned image information; Using a preset calibration factor, photon counting processing is performed on the scanned image information after resolution restoration to obtain the activity concentration data of each unit voxel in the scanned image information.

5. The drug dosage planning method according to claim 3, characterized in that, The vascular perfusion zones include lesion zones and corresponding feeding artery perfusion zones; based on the voxel activity concentration distribution data, lesion calculations are performed to obtain the tumor-to-normal tissue activity concentration ratio data for each of the vascular perfusion zones, including: Based on the number of voxels contained in each lesion region and the number of voxels contained in each blood supply artery perfusion region in the voxel activity concentration distribution data, the lesion region activity concentration and the blood supply artery perfusion region activity concentration of each lesion region and each blood supply artery perfusion region are calculated. The percentage of the active concentration of the lesion zone in all vascular perfusion zones is determined as the lesion zone uptake fraction of each lesion zone, and the percentage of the active concentration of the feeding artery perfusion zone in all vascular perfusion zones is determined as the feeding artery perfusion zone uptake fraction of each feeding artery perfusion zone. Based on the uptake fraction of the lesion zone and the uptake fraction of the blood supply artery perfusion zone, a mass ratio is calculated to obtain the tumor-to-normal tissue active concentration ratio data for each of the blood vessel perfusion zones.

6. The drug dosage planning method according to claim 5, characterized in that, Based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug doses, the optimal treatment dose is calculated to obtain the optimal treatment dose of the target drug in each of the aforementioned vascular perfusion zones, including: Obtain a preset optimal treatment dose calculation function; the optimal treatment dose calculation function is a target function pre-constructed based on the preset target constraints, the preset lesion priority weights, and the preset target drug target dose; Based on the absorbed dose distribution data, and combined with the perfusion overlap area in each of the vascular perfusion zones, the actual drug absorbed dose is calculated to obtain the actual drug absorbed dose of each of the vascular perfusion zones. Based on the actual drug absorption dose, the optimal therapeutic dose of the target drug in each of the vascular perfusion zones is calculated using the optimal therapeutic dose calculation function.

7. The drug dosage planning method according to claim 6, characterized in that, Based on the absorbed dose distribution data, and combined with the perfusion overlap areas in each of the aforementioned vascular perfusion zones, the actual drug absorbed dose is calculated to obtain the actual drug absorbed dose for each of the aforementioned vascular perfusion zones, including: From the scanned image information, the perfusion overlap region between each of the vascular perfusion zones is determined; the perfusion overlap region is the area where the lesion zone in the vascular perfusion zone overlaps with the blood supply artery perfusion zone in other vascular perfusion zones; Determine the volume percentage of each of the perfusion overlap regions in each of the vascular perfusion zones; Based on the volume ratio of the overlapping region and the absorbed dose distribution data, the self-absorbed dose of each of the vascular perfusion zones and the cross-absorbed dose of the corresponding perfusion overlapping region are calculated to obtain the actual drug absorbed dose of each of the vascular perfusion zones.

8. A drug dosage planning device, characterized in that, include: The acquisition module is used to acquire the scanned image information of the examinee; A segmentation module is used to divide the scanned image information into regions to obtain vascular perfusion zones; The quantification module is used to quantify the voxel activity of each of the aforementioned vascular perfusion zones, and obtain voxel activity concentration distribution data and tumor-to-normal tissue activity concentration ratio data of each of the aforementioned vascular perfusion zones. The prediction module is used to predict the absorbed dose based on the voxel activity concentration distribution data and the tumor-to-normal tissue activity concentration ratio data, so as to obtain the absorbed dose distribution data of the target drug. The calculation module is used to calculate the optimal treatment dose based on the absorbed dose distribution data, combined with preset target constraints, preset lesion priority weights, and preset target drug target doses, so as to obtain the optimal treatment dose of the target drug in each of the vascular perfusion zones.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing a computer program executable by the processor, which, when executing the computer program, performs the steps of the drug dosing planning method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it performs the steps in the drug dosing planning method as described in any one of claims 1-7.

Images