CN-122005100-A - Guidewire autonomous delivery system based on fuzzy logic

CN122005100ACN 122005100 ACN122005100 ACN 122005100ACN-122005100-A

Abstract

A fuzzy logic based guidewire autonomous delivery system is disclosed. The guidewire autonomous delivery system includes an image processing module configured to acquire a blood vessel image and extract feature information including a relative pose of a distal end of a guidewire and a blood vessel from the blood vessel image, a fuzzy logic calculation module configured to calculate a guidewire delivery control parameter including at least one of axial translation and radial rotation based on the feature information and the fuzzy logic, and a delivery control module configured to perform a delivery operation of the guidewire based on the feature information and the guidewire delivery control parameter. The guide wire autonomous delivery system based on the fuzzy logic according to the embodiment of the disclosure does not need to execute training operation due to the use of the fuzzy logic method, has high generalization capability, strong interpretability and high success rate, reduces the control times of a doctor to a robot in operation, and reduces the operation threshold and the workload of the doctor.

Inventors

ZHOU XIAOHU
WU PEILIANG
WANG DINGQIAN
GUI MEIJIANG
XIE XIAOLIANG
LIU SHIQI
LAI ZHICHAO
HOU ZENGGUANG

Assignees

中国科学院自动化研究所

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (10)

1. A fuzzy logic based guidewire autonomous delivery system, the guidewire autonomous delivery system comprising: an image processing module configured to acquire a blood vessel image and extract feature information including a relative pose of a distal end of a guidewire and a blood vessel from the blood vessel image; A fuzzy logic calculation module configured to calculate a guidewire delivery control parameter including at least one of an axial translation and a radial rotation based on the characteristic information and the fuzzy logic, and And a delivery control module configured to perform a delivery operation of the guidewire based on the characteristic information and the guidewire delivery control parameter.
2. The guidewire autonomous delivery system of claim 1, wherein the image processing module is configured to: converting the blood vessel image into gray image and separating blood vessel region, and Feature information including the relative pose of the distal end of the guidewire and the vessel model is extracted from the vessel region.
3. The guidewire autonomous delivery system of claim 1, wherein the fuzzy logic calculation module is configured to: Acquiring values of n input variables based on the characteristic information, wherein n is a positive integer; Determining at least one potential ambiguity value corresponding to a value of each of the n input variables based on a membership function of each input variable; Determining m fuzzy rules associated with m potential fuzzy value combinations in a fuzzy rule base, wherein the m potential fuzzy value combinations refer to all combinations consisting of one potential fuzzy value corresponding to the value of each of the n input variables, and m is a positive integer; Determining an aggregate output variable membership function corresponding to the set of m fuzzy rules; Determining values of the output variables corresponding to the n input variables as guidewire delivery control parameters using a centroid method based on an aggregate output variable membership function, Wherein each fuzzy rule reflects a correspondence between a combination of potential fuzzy values and a particular fuzzy value of the output variable.
4. The guidewire autonomous delivery system of claim 3, wherein the number of m potential ambiguity value combinations corresponds to a product of the number of potential ambiguity values corresponding to the value of each input variable.
5. The guidewire autonomous delivery system of claim 3, wherein the fuzzy logic calculation module is configured to: For each fuzzy rule of the m fuzzy rules, determining a degree of membership of the value of each of the n input variables to a particular fuzzy value indicated by the fuzzy rule and corresponding to the input variable; determining the minimum value of the membership degrees of each of the n input variables as the rule precondition membership degree of the fuzzy rule; obtaining a rule conclusion membership function of the fuzzy rule by using a membership function of a rule precondition membership cutoff output variable of the fuzzy rule; And determining the maximum value among the values of the rule conclusion membership functions of each m fuzzy rules, and taking the rule conclusion membership function corresponding to the determined maximum value as an aggregate output variable membership function corresponding to the set of m fuzzy rules.
6. The guidewire autonomous delivery system of claim 3, wherein the fuzzy logic calculation module is configured to: And determining an abscissa corresponding to the centroid position of the region of the part surrounded by the aggregated output variable membership function and the abscissa axis as the value of the output variable corresponding to the value of the n input variables, and taking the value as a guide wire delivery control parameter.
7. The guidewire autonomous delivery system of claim 3, wherein, The n input variables comprise the distance from the distal end of the guide wire to the current blood vessel branch point in the target direction and the distance from the distal end of the guide wire to the next blood vessel branch point in the target direction, the output variables comprise the guide wire translation distance, the fuzzy value corresponding to the distance from the distal end of the guide wire to the current blood vessel branch point in the target direction comprises short, medium and long, the fuzzy value corresponding to the distance from the distal end of the guide wire to the next blood vessel branch point in the target direction comprises short, medium and long, and/or the fuzzy value corresponding to the guide wire translation distance comprises short, medium and long The n input variables include the number of current vascular branch attempts to repeat and the consistency of the guidewire distal bend direction with the target vascular branch bias, the output variables include the delivery end guidewire rotation angle, the fuzzy values corresponding to the current vascular branch attempts to repeat include very large, medium, and small, the fuzzy values corresponding to the guidewire distal bend direction with the target vascular branch bias include the same and different, and the fuzzy values corresponding to the delivery end guidewire rotation angle include very large, medium, small, and very small.
8. The guidewire autonomous delivery system of claim 1, wherein the delivery control module is configured to: detecting whether the distal end of the guide wire is in a correct intervention path based on the characteristic information; Performing a translation operation without performing a rotation operation on the guidewire based on the guidewire delivery control parameter when the guidewire distal end is in the correct interventional path; when the distal end of the guide wire is not in the correct intervention path, the control is used for retracting the guide wire once, then the control is used for controlling the guide wire to rotate once and forward delivery based on the guide wire delivery, Wherein the control wire is initially rotated in a first direction when the rotating operation is performed, and is rotated in a second direction opposite to the first direction when the distal bending direction of the wire is detected to be opposite to the bias of the target vessel branch.
9. A computer-readable storage medium, wherein instructions stored in the computer-readable storage medium, when executed by at least one processor, cause the at least one processor to perform: collecting a blood vessel image; extracting characteristic information comprising the relative pose of the distal end of the guide wire and the blood vessel from the blood vessel image; Calculating guidewire delivery control parameters including at least one of axial translation and radial rotation based on the characteristic information and the fuzzy logic, and Based on the characteristic information and the guidewire delivery control parameters, a guidewire delivery operation is performed.
10. A computer program product comprising computer instructions which, when executed by at least one processor, cause the at least one processor to perform: collecting a blood vessel image; extracting characteristic information comprising the relative pose of the distal end of the guide wire and the blood vessel from the blood vessel image; Calculating guidewire delivery control parameters including at least one of axial translation and radial rotation based on the characteristic information and the fuzzy logic, and Based on the characteristic information and the guidewire delivery control parameters, a guidewire delivery operation is performed.

Description

Guidewire autonomous delivery system based on fuzzy logic Technical Field The present invention relates to a guidewire autonomous delivery system, and more particularly, to a guidewire autonomous delivery system based on fuzzy logic. Background Vascular intervention is the primary clinical approach to coronary heart disease treatment, and physicians use flexible guidewires to guide interventional instruments to the lesion site for treatment according to Digital Subtraction Angiography (DSA). The physician manipulates the guidewire outside the patient's body to advance the distal tip along the arterial system to the target location by pushing and rotating the proximal end of the guidewire. Compared with the traditional operation, the introduction of the vascular intervention operation robot can reduce radiation exposure and physical fatigue, and improve the operation precision of the guide wire. However, first, current vascular interventional surgical robots still rely entirely on continuous control by the physician during the surgical procedure, with a low degree of intelligence. Because the current vascular interventional operation robot cannot directly interact with the guide wire, and the remarkable nonlinear relation between the guide wire delivery operation and the distal movement, even a doctor with abundant experience needs a great deal of training to master. Second, the next target stage of vascular interventional surgical robot intellectualization is task-level autonomy, i.e., the robotic system is able to automatically complete the surgical task with limited adjustments and controls by the physician. At this autonomous level, the vascular interventional robot can reduce the complexity of the operation and the workload of the doctor, avoiding mental fatigue and reduced accuracy of manual operation due to high concentration for a long time. Thirdly, the core challenge of autonomous delivery of a guidewire is that model-based approaches are difficult to implement due to the physical properties of the flexible guidewire, complex mechanical interactions with the vasculature, and the inability of DSA images to provide three-dimensional information. Meanwhile, the existing learning-based method often lacks of interpretability, and potential safety hazards exist in clinical application. Furthermore, such methods require a lot of training and have limited generalization ability. Disclosure of Invention The invention aims to provide a guide wire autonomous delivery system based on fuzzy logic, which has the advantages of no need of training operation, high generalization capability, strong interpretability, high success rate, reduction of the control times of a doctor on a robot in operation, and reduction of operation threshold and the workload of the doctor. According to one embodiment of the inventive concept, a wire autonomous delivery system based on fuzzy logic is provided, which includes an image processing module configured to acquire a blood vessel image and extract feature information including a relative pose of a distal end of a wire and a blood vessel from the blood vessel image, a fuzzy logic calculation module configured to calculate a wire delivery control parameter including at least one of axial translation and radial rotation based on the feature information and the fuzzy logic, and a delivery control module configured to perform a delivery operation of the wire based on the feature information and the wire delivery control parameter. Optionally, the image processing module is configured to convert the blood vessel image into a grayscale image and separate a blood vessel region, and extract feature information from the blood vessel region including a relative pose of the distal end of the guidewire and the blood vessel model. Optionally, the fuzzy logic calculation module is configured to obtain values of n input variables based on the characteristic information, wherein n is a positive integer, determine at least one potential fuzzy value corresponding to the value of each of the n input variables based on a membership function of each of the n input variables, determine m fuzzy rules associated with m potential fuzzy value combinations in a fuzzy rule base, wherein the m potential fuzzy value combinations refer to all combinations composed of one potential fuzzy value corresponding to the value of each of the n input variables and m is a positive integer, determine an aggregate output variable membership function corresponding to a set of the m fuzzy rules, determine the value of the output variable corresponding to the n input variables as a guidewire delivery control parameter using a centroid method based on the aggregate output variable membership function, wherein each fuzzy rule reflects a correspondence between one potential fuzzy value combination and a specific fuzzy value of the output variable. Optionally, the number of m potential ambiguity value combinations corresponds to