Tumor target area artery blood vessel volume measuring method and system
By segmenting the arteries in the target area and estimating the distribution of micro-arteries, and combining the vascular anatomy and perfusion characteristics of imaging data, the precise measurement of the arterial volume in the target area was achieved. This solves the problem of inaccurate dose calculation in existing technologies and improves the accuracy and safety of treatment.
Patent Information
- Application Number
- CN202511731882.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-24
- Publication Date
- 2026-02-24
AI Technical Summary
Existing methods for measuring target area arterial volume are not accurate enough in SIRT and TACE treatments, leading to inaccurate dose calculations and affecting treatment efficacy and safety.
By acquiring medical imaging data, fractal and other algorithmic models are used to segment the target area arteries and estimate the distribution of micro-arteries. Combined with vascular anatomy and perfusion characteristics, the embolization dose is calculated, and real-time feedback and user modification functions are provided to achieve automated measurement and personalized treatment plans.
It improves the accuracy of target area arterial volume measurement, reduces human error, enhances the precision and safety of treatment, and lowers medical costs.
Smart Images
Figure CN121565401A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for measuring the volume of arterial vessels in a target area and a system for measuring the volume of arterial vessels in a target area based on this method. Background Technology
[0002] Selective internal radiation therapy (SIRT) is a commonly used treatment for malignant liver tumors. Currently, yttrium-90 is frequently used. 90 Y) Radioactive microsphere SIRT. The basic principle is to adsorb yttrium-90 onto microspheres (such as resin microspheres) or to bombard yttrium-89-containing glass microspheres with a high-flux nuclear reactor to produce yttrium-90-containing glass microspheres. These microspheres, with a prescribed dose of activity, are then selectively injected into the arterial vessels of the tumor target area via an interventional catheter. Because the microspheres are 20-60 μm in diameter, they are essentially embedded within the terminal arteries. This allows for localized internal radiotherapy of the tumor tissue using the pure beta rays released from the decay of yttrium-90.
[0003] Besides selecting suitable patients, precise superselective cannulation, and simulation examinations, the calculation of the prescribed dose is crucial in the entire treatment process. Currently, SIRT prescribed dose calculations are mainly based on the measurement of target volume (including tumor volume + non-tumor volume) and non-target volume. However, the radionuclide carriers relied upon by SIRT are microspheres, which reside within the target artery. Therefore, current SIRT dose calculations are still in their early stages. If the target artery volume can be accurately measured, the number of microspheres that the target artery can hold can be precisely calculated. Combined with traditional target volume measurement methods, a new and more accurate method for calculating SIRT prescribed dose activity can be developed.
[0004] Transarterial chemoembolization (TACE) is currently the most widely used interventional treatment for liver cancer. It is the standard treatment for intermediate-stage liver cancer according to the Barcelona Clinical Liver Cancer Staging (BCLC), and the Chinese National Liver Cancer Staging (CNLC) also uses TACE for liver cancer from stage Ib to IIIb, especially in stages IIb and IIIa, where TACE is recommended as the first-line therapy. Its basic principle is to use a microcatheter for superselective (segmental or subsegmental artery) insertion into the tumor-feeding artery branches, minimizing damage to surrounding normal liver tissue. During the procedure, CBCT (cone-beam computed tomography) guidance is recommended for precise cannulation of the target vessel. The microcatheter is used to intermittently and slowly inject a suspension of iodized oil emulsion or drug-loaded microspheres, chemotherapy drugs, and contrast agents into the tumor-feeding artery at a slow and stable flow rate (determined based on tumor location, patient pain threshold, etc.). Embolization is performed branch by branch according to the tumor's blood supply, and injection is stopped when the endpoint of embolization is reached. By accurately measuring the arterial volume in the tumor target area, doctors can obtain more detailed information to help develop personalized treatment plans. Based on the size and distribution of the vessel volume, appropriate embolization agents and chemotherapy drugs can be selected, the dosage and distribution of embolization agents can be determined more accurately, and the injection speed and dosage can be adjusted, thereby improving the accuracy, safety and effectiveness of TACE treatment. Summary of the Invention
[0005] One objective of this invention is to provide a method for measuring the volume of a target artery, thereby enabling accurate calculation of the embolization dose. Another objective of this invention is to provide a system based on the aforementioned method.
[0006] To achieve the above objectives, a first aspect of the present invention discloses a method for measuring the arterial volume of a tumor target area, characterized by comprising: Obtain medical imaging data of the target object, wherein the medical imaging data includes information on the blood supply vessels of the tumor and the vascular structures and perfusion information surrounding the tumor; The target area arterial volume is calculated based on the medical imaging data, and the required embolization dose is calculated based on the arterial volume.
[0007] Preferably, the calculation of the target area arterial vascular volume includes: The target area arteries in the medical image data are segmented, and based on the segmented arteries, the distribution and structure of micro-arteries in the image data are estimated using fractal and other algorithm models to generate a complete arterial vascular bed in the target area. For the complete arterial vascular bed in the target area, vascular anatomical features are extracted and generated, and perfusion features of the target area are extracted from the temporal density curve of the image data. The target area arterial vascular volume algorithm model takes the vascular anatomical features and perfusion features as inputs and outputs the predicted value of the embolization dose. The system displays the identification results of the target area artery and its volume, along with the predicted embolization dose, to the user. The user can then modify any inaccurate identification results, and the system will recalculate the corresponding predicted embolization dose based on the modified results using the target area artery volume algorithm model.
[0008] Preferably, when training the target area arterial volume algorithm model, the total injection dose when the embolic material is injected into the patient is obtained as a modeling label.
[0009] A second aspect of the present invention discloses a system for measuring the volume of arterial vessels in a tumor target area, characterized in that it is used to implement the above-described measurement method, and characterized in that it comprises: The acquisition module is used to obtain medical image data of the target object, wherein the medical image data includes information on the blood supply vessels of the tumor and the vascular structures and perfusion information around the tumor; The calculation module, used to identify the target area arterial vessels and their volume based on the medical image data, and to calculate the required embolization dose, further includes an image preprocessing unit, a target area arterial vessel volume algorithm model, and a judgment unit, wherein: The image preprocessing unit is used to preprocess the medical image data; The target area arterial volumetric algorithm model is used to segment visible target area arteries from preprocessed medical image data, and based on the segmented visible arteries, to estimate the distribution and structure of invisible micro-arteries in the image data using fractal and other algorithm models, generating a complete target area arterial vascular bed. For the complete target area arterial vascular bed, the vascular anatomy features are extracted and generated, and the perfusion features of the target area are extracted from the image data. The target area arterial volumetric algorithm model takes the vascular anatomy features and perfusion features as inputs and outputs the predicted value of the embolization dose. Based on the output of the target area arterial vessel volume algorithm model, the user uses the judgment unit to judge and modify the target target area arterial vessel, its volume, and the embolization dose. The display module is used to display the prediction results of the calculation module, as well as the target area arterial vascular bed modified by the user through the judgment unit.
[0010] Preferably, the medical imaging data is DSA image, CBCT image, or preoperative CTA, MRA, or CTP image.
[0011] Preferably, when the target area arterial volume algorithm model segments the visible target area arterial vessels, it uses a multi-scale segmentation algorithm to automatically or manually segment the blood supply arteries in the medical image data.
[0012] Preferably, the vascular anatomical features include vessel diameter and vessel length, and the perfusion features include arrival time, peak time, arrival value, and peak value.
[0013] Preferably, a regression method using machine learning or deep learning is used to construct and train the target area arterial volume algorithm model.
[0014] Preferably, the judgment unit displays the automatically generated target area arterial vascular bed and embolization dose through the display module, and the user modifies the inaccurately segmented target area arterial vascular bed. The modified target area arterial vascular bed is then fed back to the target area arterial vascular volume algorithm model, which then re-predicts the tumor embolization dose.
[0015] Preferably, the measurement system is used either by integration into a hardware device system or by installation on a standalone workstation.
[0016] This invention enables precise calculation of radiation dose for selective internal radiotherapy (SIRT) of liver malignant tumors using radioactive microspheres and precise embolization of liver malignant tumors via transarterial chemoembolization (TACE). The technical solution disclosed in this invention can improve the automation and accuracy of measurements, reduce human error, achieve personalized calculation of treatment doses, and improve treatment efficacy and patient safety.
[0017] Compared with existing technical solutions, the present invention has the following advantages: 1. Improved accuracy: The new measurement method provides higher measurement accuracy and reduces errors in embolization dose calculation; 2. High degree of automation: Automated measurement and dosage calculation reduce the workload of doctors and improve treatment efficiency; 3. Real-time feedback: The system can provide real-time measurement results and feedback, enabling doctors to adjust treatment plans immediately; 4. High adaptability: The model can adapt to different types and shapes of blood vessels, improving the universality of the method; 5. Cost-effectiveness: In the long run, by reducing treatment errors and improving treatment outcomes, medical costs and the burden on patients can be reduced. Attached Figure Description
[0018] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a flowchart of the present invention. Detailed Implementation
[0019] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.
[0020] like Figure 1 As shown, the first aspect of this invention discloses a method for measuring the arterial vascular volume in a tumor target area, capable of extracting the target area arterial vascular bed from patient images and calculating the vascular volume and embolization dose, specifically including the following steps: The first step is to acquire image data of the target object during the injection of contrast agent and preprocess the image data.
[0021] The image data should include information on the blood vessels supplying the tumor, the vascular structures surrounding the tumor, and perfusion and other functional information, such as DSA video images, CBCT images, preoperative CTA, MRA, CTP images, and other related multimodal imaging data. Image processing algorithms are used to preprocess image data, including but not limited to image denoising and image enhancement.
[0022] The second step involves generating visible and invisible target arterial beds for the patient using a target arterial angiogenesis model. Based on the generated target arterial beds, a target arterial vascular volume algorithm model is used to calculate the vascular volume and embolization dose.
[0023] The target area arterial volumetric algorithm model segments visible target area arteries from preprocessed image data. Based on the segmented visible arteries, it uses fractal and other algorithms to estimate the distribution and structure of invisible micro-arteries in the image data, generating a complete target area arterial vascular bed. For the complete target area arterial vascular bed, it extracts vascular anatomical features such as vessel diameter and vessel length, and extracts perfusion features of the target area from the temporal density curve of the image data. The target area arterial volumetric algorithm model takes vascular anatomical features and perfusion features as input and outputs a predicted value of embolization dose.
[0024] In a preferred embodiment of the present invention, when training the target area arterial vascular volume algorithm model, the total injection dose when the patient injects microspheres with a diameter of 20-60 μm is obtained and used as a modeling label for subsequent training of the target area arterial vascular volume algorithm model.
[0025] The third step is to display the identified target arterial vascular bed and the calculated embolization dose to the user, who can then modify the target arterial vascular information and embolization dose.
[0026] Users can determine and modify the target artery, its volume, and the embolization dose based on the output of the target artery volumetric algorithm model. In a preferred embodiment of the invention, the system automatically generates and displays the target artery vascular bed and the embolization dose. The user then modifies any inaccurately segmented target artery vascular beds and feeds the modified beds back to the target artery volumetric algorithm model, which then re-predicts the tumor embolization dose.
[0027] like Figure 2 As shown, a second aspect of the present invention discloses a measurement system for the arterial volume of a tumor target area, including an acquisition module S1, a calculation module S2, and a display module S3.
[0028] The acquisition module S1 acquires target object data from hardware devices or information systems such as PACS via network connection. For example, it acquires multimodal imaging data related to the patient, such as DSA video images, CBCT images, preoperative CTA, MRA, and CTP images. The imaging data must include information on the blood supply vessels of the tumor and the vascular structures around the tumor, as well as perfusion and other functional information.
[0029] The calculation module S2 identifies the target area arterial vessels and their volume based on multimodal image data, and calculates the required embolization dose. In this embodiment of the invention, the calculation module S2 further includes an image preprocessing unit, a target area arterial vessel volume algorithm model, and a judgment unit.
[0030] The image preprocessing unit applies image processing algorithms to preprocess the image data acquired by the acquisition module S1, including but not limited to image denoising and image enhancement.
[0031] The target area arterial vascular bed generation model segments visible target area arteries from preprocessed image data. Based on the segmented visible arteries, it uses fractal and other algorithms to estimate the distribution and structure of invisible micro-arteries in the image data, generating a complete target area arterial vascular bed. In a preferred embodiment of the invention, when segmenting visible target area arteries, the target area arterial vascular bed generation model uses a vascular segmentation algorithm to automatically or manually segment the feeding arteries in the medical image.
[0032] For the intact arterial vascular bed in the target area, vascular anatomical features such as vessel diameter and length are extracted and generated. Perfusion features of the target area are extracted from the temporal density curves of the image data. The target area arterial vascular volume algorithm model takes these vascular anatomical and perfusion features as input and outputs a predicted value for the embolization dose. For example, using the total microsphere injection dose recorded intraoperatively as the modeling label, a target area arterial vascular volume algorithm model is constructed and trained using machine learning or deep learning regression methods. Different correction parameters can be set to adjust the embolization dose for embolization materials with different properties and sizes, such as iodized oil and large-diameter microspheres. The correction coefficients can be derived from clinical experience, volume conversion, and fluid simulation.
[0033] Based on the output of the target area arterial vascular volume algorithm model, the user uses a judgment unit to determine and modify the target area arterial vessels, their volume, and the embolization dose. In a preferred embodiment of the invention, the judgment unit displays the automatically generated target area arterial vascular bed and embolization dose through a display module. The user then modifies any inaccurately segmented target area arterial vascular beds and feeds the modified beds back to the target area arterial vascular volume algorithm model, which then re-predicts the tumor embolization dose.
[0034] The display module S4 is used to display the prediction results of the calculation module S2, as well as the target area arterial vascular bed modified by the user through the judgment module S3.
[0035] The tumor target area arterial volume measurement system disclosed in this invention can be used by integrating it into a hardware device system or by installing it on a standalone workstation.
Claims
1. A method for measuring the arterial volume of a tumor target area, characterized in that, Includes the following steps: Obtain medical imaging data of the target object, the medical imaging data including information on the blood supply vessels of the tumor and the vascular structures and perfusion information around the tumor; calculate the arterial volume of the target area based on the medical imaging data, and calculate the required embolization dose based on the arterial volume.
2. The method for measuring the arterial volume of a tumor target area according to claim 1, characterized in that, The calculation of the target area arterial vascular volume includes: segmenting the target area arterial vessels in the medical image data, and estimating the distribution and structure of micro-arteries in the image data using fractal and other algorithm models based on the segmented arterial vessels to generate a complete arterial vascular bed in the target area; extracting vascular anatomical features from the complete arterial vascular bed in the target area, and extracting perfusion features of the target area from the temporal density curve of the image data; the target area arterial vascular volume algorithm model takes the vascular anatomical features and perfusion features as input and outputs a predicted value of the embolization dose; displaying the identification results of the target area arterial vessels and vascular volume, as well as the predicted value of the embolization dose, to the user, allowing the user to modify inaccurate identification results, and then recalculating the corresponding predicted value of the embolization dose based on the modified identification results by the target area arterial vascular volume algorithm model.
3. A system for measuring the volume of arterial vessels in a tumor target area, characterized in that, include: The acquisition module is used to acquire medical image data of the target object, which includes information on the blood supply vessels of the tumor and the vascular structures and perfusion information around the tumor. The calculation module is used to identify the target area arteries and their volumes based on the medical image data and to calculate the required embolization dose. It further includes an image preprocessing unit, a target area artery volume algorithm model, and a judgment unit. The image preprocessing unit preprocesses the medical image data. The target area artery volume algorithm model segments the visible target area arteries from the preprocessed medical image data and applies fractal and other algorithm models based on the segmented visible arteries. The system estimates the distribution and structure of invisible micro-arteries in the image data, generates a complete arterial vascular bed in the target area, extracts vascular anatomical features for the complete arterial vascular bed in the target area, and extracts perfusion features of the target area from the temporal density curve of the image data. The target area arterial vascular volume algorithm model takes the vascular anatomical features and perfusion features as input and outputs the predicted value of the embolization dose. Based on the output of the target area arterial vascular volume algorithm model, the user uses a judgment unit to judge and modify the target area arterial vessels, their volume, and the embolization dose. The display module is used to display the prediction results of the calculation module and the target area arterial vascular bed modified by the user through the judgment unit.
4. The system for measuring the arterial volume of a tumor target area as described in claim 3, characterized in that, The medical imaging data includes DSA video images, CBCT images, or preoperative CTA, MRA, and CTP images.
5. The system for measuring the arterial volume of a tumor target area as described in claim 3, characterized in that, When the target area arterial volume algorithm model segments the visible target area arterial vessels, it uses a vessel segmentation algorithm to automatically or manually segment the blood supply arteries in the medical image data.
6. The system for measuring the arterial volume of a tumor target area as described in claim 3, characterized in that, The vascular anatomical features include vessel diameter and vessel length, and the perfusion features include arrival time, peak time, arrival value, and peak value.
7. The system for measuring the arterial volume of a tumor target area as described in claim 3, characterized in that, The judgment unit displays the automatically generated target area arterial vascular bed and embolization dose through the display module. The user can modify the inaccurately segmented target area arterial vascular bed and feed the modified target area arterial vascular bed back to the target area arterial vascular volume algorithm model, which then re-predicts the tumor embolization dose.
8. The system for measuring the arterial volume of a tumor target area as described in claim 3, characterized in that, The measurement system can be used either by integrating it into a hardware device system or by installing it on a standalone workstation.