Auxiliary decision-making system for bronchial artery embolism and spring ring operation method

Abstract

The invention relates to the technical field of medical instruments, in particular to an auxiliary decision-making system for bronchial artery embolism and a spring ring operation method, and the system comprises an image recognition module, a case scene analysis module and an interactive display terminal. The image recognition module receives a bronchial artery angiography image transmitted by the angiography equipment, the case scene analysis module receives a responsible blood vessel recognition result and scene feature information, and the interactive display terminal receives and presents the responsible blood vessel recognition result, the scene feature information, an embolism strategy and decision suggestions. Receiving a correction operation of a doctor on the embolism strategy and the decision suggestion, and outputting a personalized embolism scheme according to the correction operation; the output end of the image recognition module is connected with the input end of the case scene analysis module, and the output end of the case scene analysis module is connected with the input end of the interactive display terminal. Through image processing, strategy generation, correction supporting and spring ring matching operation, disease adaptation, error reduction, safety protection and treatment standardization promotion are realized.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to an auxiliary decision-making system for bronchial artery embolization and a coil operation method. Background Technology

[0002] Bronchial artery embolization is a crucial local treatment method in the current field of lung cancer treatment. 95% of lung cancer blood supply originates from systemic arteries such as the bronchial arteries, clinically referred to as the responsible vessels. Therefore, this technique plays an irreplaceable role in lung cancer treatment. However, bronchial artery chemoembolization carries a serious complication: spinal cord injury. Once this occurs, it can lead to urinary and fecal incontinence, loss of lower limb function, and a very poor prognosis with almost no chance of recovery. This complication is closely related to specific anatomical structures: the right bronchial artery often shares a trunk with the ipsilateral intercostal artery, and the spinal artery originates from a distal branch of the intercostal artery. To address common clinical issues such as the responsible vessel opening being too close, leading to embolic agent reflux, hardening or narrowing of the opening, or sharp-angle deformation, preventing the microcatheter from reaching a safe position, traditional procedures place the microcatheter in the main trunk of the intercostal artery and release coils to block blood flow, preventing embolic particles from entering the distal branches of the intercostal artery and damaging the spinal artery, thus attempting to reduce the risk of spinal cord injury.

[0003] Current bronchial artery embolization techniques still have significant shortcomings and deficiencies in practical applications. On the one hand, the entire process of traditional embolization relies heavily on the physician's individual clinical experience. When faced with different disease types such as benign hemoptysis and malignant tumors (e.g., lung cancer), as well as complex operational scenarios such as unsatisfactory superselective intubation, it is difficult to achieve precise planning of the embolization strategy. This includes a lack of unified and scientific judgment criteria for the selection of embolization materials such as coils and microparticles, and the determination of specific usage methods, resulting in poor adaptability of treatment plans to the specific conditions and anatomical characteristics of different cases. On the other hand, although traditional procedures use coil occlusion of the main intercostal arteries to attempt to avoid the risk of spinal cord injury, the subjective and limited nature of experience-based judgment means that the risk of spinal cord injury is still not effectively controlled, and patient safety is difficult to guarantee. Furthermore, there is a severe lack of auxiliary tools and systems specifically designed for bronchial artery embolization scenarios in the current market, failing to provide physicians with objective data support and operational guidance, further exacerbating the uncertainty of the procedure and ultimately greatly restricting the stability of treatment effects and the safety of the treatment process. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned problems and provide an auxiliary decision-making system and coil operation method for bronchial artery embolism. To achieve the above objective, this invention adopts the following technical solution:

[0005] A decision support system for bronchial artery embolization includes an image recognition module, a case scene analysis module, and an interactive display terminal. The image recognition module receives bronchial artery angiography images transmitted from an angiography device, processes the images, and outputs the responsible vessel identification result and scene feature information. The case scene analysis module receives the responsible vessel identification result and scene feature information, generates an embolization strategy and decision suggestions based on built-in decision logic rules, and the interactive display terminal receives and presents the responsible vessel identification result, scene feature information, embolization strategy, and decision suggestions. It also receives correction operations from doctors on the embolization strategy and decision suggestions and outputs a personalized embolization plan based on the correction operations. The output end of the image recognition module is connected to the input end of the case scene analysis module, and the output end of the case scene analysis module is connected to the input end of the interactive display terminal.

[0006] Furthermore, the image recognition module processes bronchial artery angiography images including preprocessing to remove noise and extracting vascular features and tumor-related features through deep learning algorithms. The vascular features include the opening location, course, and morphology of the responsible vessel, and the tumor-related features include the spatial correlation between the responsible vessel and the tumor tissue. The responsible vessel identification result includes the opening location, course, and morphology information of the responsible vessel, and the scene feature information includes the results of distinguishing between benign and malignant tumor blood supply vessels and superselective cannulation status information.

[0007] Furthermore, the case scenario analysis module incorporates decision logic rules that are linked to clinical bronchial artery embolism case data, which includes angiographic images, embolism treatment plans, and treatment prognoses. The case scenario analysis module optimizes the decision logic rules by adding new clinical bronchial artery embolism case data.

[0008] When the scene features information is benign hemoptysis, the embolization strategy is to fill the distal end of the target vessel with microparticles combined with the main trunk coil occlusion of the target vessel.

[0009] When the scene feature information is a lung malignant tumor scene, the embolization strategy is to fill the distal end of the target vessel with microparticles and not to perform target vessel trunk coil occlusion.

[0010] When the scenario features unsatisfactory superselective cannulation, the embolization strategy is protective occlusion of the main intercostal artery with coils; the decision recommendations include information on the selection of embolization materials and guidance on the operation procedures.

[0011] Furthermore, it is deployed in the imaging workstation of the interventional operating room; the image recognition module establishes a real-time connection with the angiography equipment through the data interface. The bronchial artery angiography images received by the image recognition module are real-time dynamic image data, which includes the angiography image of the responsible vessel and the superselective cannulation image; the image recognition module processes the real-time dynamic image data in real time, and the processed responsible vessel recognition result and scene feature information are synchronized with the clinical operation.

[0012] Furthermore, the interactive display terminal presents content including a visually labeled image of the responsible vessel identification result, graphic and textual explanations of scene feature information, a flowchart of the embolization strategy, and detailed text of decision-making suggestions; embolization material selection information includes microparticle type selection information and information on whether or not coils are used; operation procedure guidance information includes the order of microparticle filling and the timing of coil occlusion; the interactive display terminal allows doctors to adjust the embolization material selection information and operation procedure guidance information, resulting in a personalized embolization plan.

[0013] Furthermore, a method for operating a coil for bronchial artery embolization includes the following steps:

[0014] Step S1: Obtain bronchial artery angiography images from the angiography equipment. During bronchial artery embolization treatment, determine whether the microcatheter can reach a safe position based on the bronchial artery angiography images.

[0015] Step S2: When the microcatheter cannot reach a safe position, use 3-4 coils to release into the main intercostal artery through the microcatheter. The blood flow in the main intercostal artery is completely blocked, and the embolic particles cannot flow to the spinal artery. There is no backflow when injecting the embolic particles.

[0016] Step S3: Determine the type of disease of the patient based on the bronchial arteriography images. The types of diseases include benign hemoptysis and malignant lung tumors.

[0017] Step S31: When the symptom type is benign hemoptysis, microparticles are used to fill the distal end of the target vessel. After confirming that the filling effect is satisfactory based on the bronchial artery angiography images, the main trunk of the target vessel is sealed with coils.

[0018] Step S32: When the disease type is malignant lung tumor, microparticles are used to fill the distal end of the target vessel. After confirming that the filling effect is satisfactory based on bronchial artery angiography images, coils are not used to block the main trunk of the target vessel.

[0019] The advantages of this invention are:

[0020] 1. This invention receives bronchial artery angiography images transmitted by an angiography device through the image recognition module of the auxiliary decision-making system, processes them, and outputs the responsible vessel identification result and scene feature information. The case scene analysis module generates corresponding embolization strategies for benign hemoptysis, malignant lung tumors, and unsatisfactory superselective intubation scenarios based on built-in decision logic rules. The interactive display terminal presents the information and supports doctors to make corrections. At the same time, combined with the operation of 3-4 coils to block the main intercostal artery in the coil operation method, the bronchial artery embolization treatment is accurately adapted to different diseases, reduces human decision-making errors, effectively prevents recurrence of benign hemoptysis, preserves the subsequent treatment pathway for malignant tumors, protects the spinal cord, and improves the accuracy and safety of treatment.

[0021] 2. This invention deploys an auxiliary decision-making system in the imaging workstation of the interventional operating room. The image recognition module is connected to the angiography equipment in real time and processes dynamic images synchronously. The case scenario analysis module relies on the decision logic rules of related clinical angiography images, treatment plans and prognostic effects, and can be optimized by adding new cases. Combined with the interactive terminal, it outputs personalized embolization plans, realizing the transformation of the traditional treatment mode that relies on personal experience into a reusable system decision-making mode. This provides support for the standardization of bronchial artery embolization treatment process and helps to improve the overall level of interventional treatment in different medical institutions.

[0022] 3. This invention designs a bronchial artery embolization auxiliary decision-making system that includes an image recognition module, a case scene analysis module, and an interactive display terminal. It integrates functions such as real-time image processing, scene-based embolization strategy generation, and personalized doctor correction. It is not limited to the coil operation method. It breaks through the limitations of patents for simple treatment methods and builds a technical protection system at the system software level, providing solid intellectual property protection for the clinical transformation and industrial application of this innovative technology. Attached Figure Description

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application.

[0024] In the attached diagram:

[0025] Figure 1 This is a framework diagram of an auxiliary decision-making system for bronchial artery embolism in Example 1.

[0026] Figure 2 This is a flowchart of a coil operation method for bronchial artery embolization in Example 1. Detailed Implementation

[0027] The present invention will now be described in detail and specifically through specific embodiments to enable a better understanding of the invention. However, the following embodiments do not limit the scope of protection of the present invention.

[0028] Example 1

[0029] like Figure 1-2 As shown, an auxiliary decision-making system for bronchial artery embolization includes an image recognition module, a case scene analysis module, and an interactive display terminal. The image recognition module receives bronchial artery angiography images transmitted by an angiography device, processes the images, and outputs the responsible vessel identification result and scene feature information. The case scene analysis module receives the responsible vessel identification result and scene feature information, generates an embolization strategy and decision suggestions based on built-in decision logic rules, and the interactive display terminal receives and presents the responsible vessel identification result, scene feature information, embolization strategy, and decision suggestions. It also receives correction operations from doctors on the embolization strategy and decision suggestions and outputs a personalized embolization plan based on the correction operations. The output end of the image recognition module is connected to the input end of the case scene analysis module, and the output end of the case scene analysis module is connected to the input end of the interactive display terminal.

[0030] In a specific embodiment, the bronchial artery embolization auxiliary decision-making system constructs a closed-loop decision support process from image processing to strategy generation and personalized adjustment through the orderly connection of an image recognition module, a case scenario analysis module, and an interactive display terminal. The image recognition module is responsible for extracting key diagnostic and treatment information, the case scenario analysis module outputs adaptation strategies based on built-in rules, and the interactive display terminal provides doctors with an interactive operation window. These three components work together to replace the traditional judgment mode that relies on the doctor's personal experience. This architecture significantly reduces human decision-making errors and avoids strategy deviations caused by insufficient experience. It not only improves the overall accuracy of bronchial artery embolization treatment but also provides system-level assurance for treatment safety and reduces the risk of complications caused by improper decision-making.

[0031] Furthermore, the image recognition module processes bronchial artery angiography images including preprocessing to remove noise and extracting vascular features and tumor-related features through deep learning algorithms. The vascular features include the opening location, course, and morphology of the responsible vessel, and the tumor-related features include the spatial correlation between the responsible vessel and the tumor tissue. The responsible vessel identification result includes the opening location, course, and morphology information of the responsible vessel, and the scene feature information includes the results of distinguishing between benign and malignant tumor blood supply vessels and superselective cannulation status information.

[0032] In a specific embodiment, the image recognition module preprocesses the bronchial artery angiography images by removing noise to effectively eliminate image interference factors and ensure the accuracy of subsequent feature extraction. Relying on deep learning algorithms to extract vascular features and tumor-related features, it can accurately capture the spatial attributes of the responsible vessel and its relationship with the tumor, laying a reliable data foundation for the output of the responsible vessel identification results and scene feature information. Accurate responsible vessel identification results and scene feature information can directly avoid errors in subsequent strategy generation caused by data bias, enabling the case scenario analysis module to formulate plans based on real diagnostic and treatment data, further improving the scientific nature of the entire system's decision-making and ensuring that the embolization operation is consistent with the patient's anatomy and actual condition.

[0033] Furthermore, the case scenario analysis module incorporates decision logic rules that are linked to clinical bronchial artery embolism case data, which includes angiographic images, embolism treatment plans, and treatment prognoses. The case scenario analysis module optimizes the decision logic rules by adding new clinical bronchial artery embolism case data.

[0034] When the scene features information is benign hemoptysis, the embolization strategy is to fill the distal end of the target vessel with microparticles combined with the main trunk coil occlusion of the target vessel.

[0035] When the scene feature information is a lung malignant tumor scene, the embolization strategy is to fill the distal end of the target vessel with microparticles and not to perform target vessel trunk coil occlusion.

[0036] When the scenario features unsatisfactory superselective cannulation, the embolization strategy is protective occlusion of the main intercostal artery with coils; the decision recommendations include information on the selection of embolization materials and guidance on the operation procedures.

[0037] In a specific embodiment, the decision logic rules built into the case scenario analysis module are linked to actual clinical case data and can be continuously optimized through the addition of new cases, enabling dynamic adaptation to clinical scenarios and continuously improving the suitability of the strategy. Differential embolization strategies are developed for three typical scenarios: benign hemoptysis, malignant lung tumors, and unsatisfactory superselective catheterization. These strategies can achieve precise, scenario-based responses. For benign hemoptysis, the strategy effectively reduces the probability of recurrence; for malignant lung tumors, the strategy preserves the target vessel access to support subsequent treatment; and for unsatisfactory superselective catheterization, the strategy avoids spinal cord damage from embolic particles. Furthermore, clear guidelines for embolization material selection and operational procedures can mitigate risks caused by improper material selection or operational sequence, improving the targetedness and safety of treatment in different scenarios.

[0038] Furthermore, it is deployed in the imaging workstation of the interventional operating room; the image recognition module establishes a real-time connection with the angiography equipment through the data interface. The bronchial artery angiography images received by the image recognition module are real-time dynamic image data, which includes the angiography image of the responsible vessel and the superselective cannulation image; the image recognition module processes the real-time dynamic image data in real time, and the processed responsible vessel recognition result and scene feature information are synchronized with the clinical operation.

[0039] In a specific embodiment, the system is deployed on the imaging workstation in the interventional operating room, allowing physicians to directly access system support at the treatment site without switching between devices, thus improving the convenience of clinical operations. The image recognition module connects to the angiography equipment in real time via a data interface, enabling it to synchronously receive and process real-time dynamic image data. This ensures that the output rhythm of the responsible vessel identification result and scene feature information matches the clinical operation process, avoiding operational delays caused by information lag. Furthermore, the real-time dynamic image data covers images related to the responsible vessel and superselective cannulation, providing comprehensive diagnostic and treatment information for decision-making. This effectively avoids strategic oversights due to missing information, further ensuring treatment efficiency and accuracy.

[0040] Furthermore, the interactive display terminal presents content including a visually labeled image of the responsible vessel identification result, graphic and textual explanations of scene feature information, a flowchart of the embolization strategy, and detailed text of decision-making suggestions; embolization material selection information includes microparticle type selection information and information on whether or not coils are used; operation procedure guidance information includes the order of microparticle filling and the timing of coil occlusion; the interactive display terminal allows doctors to adjust the embolization material selection information and operation procedure guidance information, resulting in a personalized embolization plan.

[0041] In specific embodiments, the interactive display terminal presents visualized annotated images, scene feature illustrations, and schematic diagrams of embolization strategies. This transforms abstract diagnostic data into intuitive visual information, eliminating the need for doctors to spend excessive time interpreting complex data, significantly shortening the information acquisition cycle and improving decision-making efficiency. Clear information on embolization material selection and operational steps, combined with functions supporting doctor adjustments, allows the treatment plan to adhere to standard procedures while adapting to individual patient differences, avoiding the problem of a uniform plan failing to address special circumstances. This combination of visual presentation and personalized adjustment reduces the operational difficulty for doctors while ensuring the individual adaptability of the treatment plan, further improving treatment outcomes.

[0042] Furthermore, a method for operating a coil for bronchial artery embolization includes the following steps:

[0043] Step S1: Obtain bronchial artery angiography images from the angiography equipment. During bronchial artery embolization treatment, determine whether the microcatheter can reach a safe position based on the bronchial artery angiography images.

[0044] Step S2: When the microcatheter cannot reach a safe position, use 3-4 coils to release into the main intercostal artery through the microcatheter. The blood flow in the main intercostal artery is completely blocked, and the embolic particles cannot flow to the spinal artery. There is no backflow when injecting the embolic particles.

[0045] Step S3: Determine the type of disease of the patient based on the bronchial arteriography images. The types of diseases include benign hemoptysis and malignant lung tumors.

[0046] Step S31: When the symptom type is benign hemoptysis, microparticles are used to fill the distal end of the target vessel. After confirming that the filling effect is satisfactory based on the bronchial artery angiography images, the main trunk of the target vessel is sealed with coils.

[0047] Step S32: When the disease type is malignant lung tumor, microparticles are used to fill the distal end of the target vessel. After confirming that the filling effect is satisfactory based on bronchial artery angiography images, coils are not used to block the main trunk of the target vessel.

[0048] In a specific embodiment, the coil manipulation method for bronchial artery embolization provides a clear premise for subsequent operations by determining the microcatheter position in step S1, avoiding the risks caused by blind operation; in step S2, 3-4 coils are used to block the blood flow of the main intercostal artery, which can directly cut off the path of the embolic particles to the spinal artery, avoiding the serious complication of spinal cord injury from an operational perspective; step S3 determines the type of disease and subsequent differentiated operations for benign hemoptysis and malignant lung tumors. In the case of benign hemoptysis, the main target vessel is blocked to reduce the risk of recurrence, while in the case of malignant lung tumors, it is not blocked to preserve the subsequent treatment pathway, making the operation more targeted. The entire process is logically rigorous, avoiding the arbitrariness of traditional operations, greatly improving the safety and treatment adaptability of the operation, and ensuring the treatment effect for patients with different diseases.

[0049] The specific embodiments of the present invention have been described in detail above, but they are merely examples, and the present invention is not equivalent to the specific embodiments described above. For those skilled in the art, any equivalent modifications and substitutions to the present invention are also within the scope of the present invention. Therefore, all equivalent transformations and modifications made without departing from the spirit and scope of the present invention should be covered within the scope of the present invention.

Claims

1. A decision support system for bronchial artery embolization, characterized in that, The system comprises an image recognition module, a case scene analysis module and an interactive display terminal; the image recognition module receives bronchial artery angiography images transmitted by a blood vessel angiography device, processes the bronchial artery angiography images, and outputs responsibility blood vessel recognition results and scene characteristic information; the case scene analysis module receives the responsibility blood vessel recognition results and scene characteristic information, generates embolization strategies and decision suggestions based on built-in decision logic rules; the interactive display terminal receives and presents the responsibility blood vessel recognition results, scene characteristic information, embolization strategies and decision suggestions, and receives correction operations of the embolization strategies and decision suggestions by doctors, and outputs individualized embolization schemes according to the correction operations. The output end of the image recognition module is connected with the input end of the case scene analysis module, and the output end of the case scene analysis module is connected with the input end of the interactive display terminal.

2. The decision support system for bronchial artery embolization according to claim 1, wherein, The processing of the bronchial artery angiography images by the image recognition module comprises pre-processing to remove noise, extracting blood vessel features and tumor correlation features by a deep learning algorithm; the blood vessel features comprise opening position, shape and morphology of the responsibility blood vessel, and the tumor correlation features comprise spatial position correlation degree between the responsibility blood vessel and tumor tissue; the responsibility blood vessel recognition results comprise opening position, shape and morphology information of the responsibility blood vessel, and the scene characteristic information comprises benign and malignant tumor blood supply vessel differentiation results and super-selective intubation state information.

3. The decision support system for bronchial artery embolization according to claim 2, wherein, The case scene analysis module has built-in decision logic rules, the decision logic rules are associated with clinical bronchial artery embolization case data, the clinical bronchial artery embolization case data comprises angiography image data, embolization treatment schemes and treatment prognosis effects; the case scene analysis module optimizes the decision logic rules by adding clinical bronchial artery embolization case data; when the scene characteristic information is a benign hemoptysis scene, the embolization strategy is target blood vessel distal particle filling combined with target blood vessel trunk coil occlusion; when the scene characteristic information is a lung malignant tumor scene, the embolization strategy is target blood vessel distal particle filling without target blood vessel trunk coil occlusion; when the scene characteristic information is an unsatisfactory super-selective intubation scene, the embolization strategy is intercostal artery trunk coil protective occlusion; the decision suggestion comprises embolization material selection information and operation step guide information.

4. The decision support system for bronchial artery embolization according to claim 3, wherein, The system is deployed in an interventional operating room image workstation; the image recognition module establishes real-time connection with the blood vessel angiography device through a data interface; the bronchial artery angiography images received by the image recognition module are real-time dynamic image data, which comprises responsibility blood vessel angiography images and super-selective intubation images; the image recognition module processes the real-time dynamic image data in real time, and the processed responsibility blood vessel recognition results and scene characteristic information are synchronized with clinical operations.

5. The decision support system for bronchial artery embolization according to claim 4, wherein, The content presented by the interactive display terminal includes a visual annotation image of the responsibility blood vessel identification result, a graphic and text description of scene feature information, a flowchart of embolization strategy, and detailed text of decision suggestions; the embolization material selection information includes particle type selection information and judgment information of whether to use a spring ring, and the operation step guide information includes particle filling sequence information and spring ring occlusion timing information; the interactive display terminal supports the doctor to adjust the embolization material selection information and the operation step guide information, and after adjustment, a personalized embolization plan is formed.

6. A method of coil manipulation for bronchial artery embolization, characterized by, The method comprises the following steps: Step S1: acquiring bronchial arteriography images collected by a vascular imaging device, and determining whether a microcatheter can reach a safe position during bronchial artery embolization treatment according to the bronchial arteriography images; Step S2: when the microcatheter cannot reach the safe position, 3-4 spring rings are released into the intercostal artery trunk through the microcatheter, the blood flow in the intercostal artery trunk is completely blocked, embolic particles cannot flow to the spinal artery, and there is no reflux when injecting the embolic particles; Step S3: determining the disease type of the treatment object according to the bronchial arteriography images, wherein the disease type includes benign hemoptysis and lung malignant tumor; Step S31: when the disease type is benign hemoptysis, the target blood vessel distal end is filled with particles, and after the filling effect is confirmed to be satisfactory according to the bronchial arteriography images, a spring ring is used to occlude the target blood vessel trunk; Step S32: when the disease type is lung malignant tumor, the target blood vessel distal end is filled with particles, and after the filling effect is confirmed to be satisfactory according to the bronchial arteriography images, a spring ring is not used to occlude the target blood vessel trunk.

Images