CN117393107B - Iterative learning method and system for automatic surgical intervention robot and storage medium - Google Patents
Iterative learning method and system for automatic surgical intervention robot and storage medium Download PDFInfo
- Publication number
- CN117393107B CN117393107B CN202311694424.8A CN202311694424A CN117393107B CN 117393107 B CN117393107 B CN 117393107B CN 202311694424 A CN202311694424 A CN 202311694424A CN 117393107 B CN117393107 B CN 117393107B
- Authority
- CN
- China
- Prior art keywords
- optimal
- stage
- sub
- interventional
- robot
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 133
- 238000011477 surgical intervention Methods 0.000 title claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 43
- 238000004806 packaging method and process Methods 0.000 claims abstract description 10
- 210000004204 blood vessel Anatomy 0.000 claims description 44
- 238000004458 analytical method Methods 0.000 claims description 26
- 230000002792 vascular Effects 0.000 claims description 17
- 238000001356 surgical procedure Methods 0.000 claims description 13
- 210000005036 nerve Anatomy 0.000 claims description 9
- 230000009471 action Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012790 confirmation Methods 0.000 claims description 5
- 230000004048 modification Effects 0.000 claims description 5
- 238000012986 modification Methods 0.000 claims description 5
- 238000013480 data collection Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000009825 accumulation Methods 0.000 description 4
- 230000006378 damage Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000002526 effect on cardiovascular system Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- 208000002177 Cataract Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 208000026106 cerebrovascular disease Diseases 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000013152 interventional procedure Methods 0.000 description 1
- 230000005865 ionizing radiation Effects 0.000 description 1
- 208000032839 leukemia Diseases 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002980 postoperative effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000002285 radioactive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 210000004026 tunica intima Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/40—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- Urology & Nephrology (AREA)
- Surgery (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Manipulator (AREA)
Abstract
The invention discloses an iterative learning method and system for an automatic surgical intervention robot and a storage medium, wherein the method comprises the following steps: collecting related information stored by the operation intervention operation of the individual intervention robots of a plurality of hospitals by applying the latest system once every interval preset time; classifying and storing the collected related information of the individual interventional robots of the hospitals; according to the related information of different classifications, analyzing and processing are carried out to obtain the optimal operation method under different conditions; summarizing and packaging various optimal operation methods, and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system. The method can continuously summarize and analyze the operation method of the interventional operation, continuously and iteratively update the operation method of the interventional robot, and continuously improve the operation capability of the interventional robot along with the increase of time and operation quantity.
Description
Technical Field
The invention relates to the technical field of minimally invasive vascular intervention, in particular to an iterative learning method and system of an automatic surgical intervention robot and a storage medium.
Background
At present, the cardiovascular and cerebrovascular minimally invasive interventional therapy is a main treatment means for cardiovascular and cerebrovascular diseases. Compared with the traditional surgery, the method has the obvious advantages of small incision, short postoperative recovery time and the like. The cardiovascular and cerebrovascular intervention operation is a treatment process by a doctor manually sending the catheter, the guide wire, the bracket and other instruments into a patient.
The intervention operation has the following 2 problems, firstly, in the operation process, as the DSA can emit X rays, the physical strength of doctors is reduced rapidly, the attention and the stability are also reduced, the operation precision is reduced, and accidents such as vascular intima injury, vascular perforation fracture and the like caused by improper pushing force are easy to occur, so that the life of patients is dangerous. Second, long-term ionizing radiation accumulation damage can greatly increase the chances of a doctor suffering from leukemia, cancer, and acute cataract. The phenomenon that doctors continuously accumulate rays due to interventional operations has become a non-negligible problem for damaging the professional lives of doctors and restricting the development of interventional operations.
The problem can be effectively solved by means of the robot technology, the accuracy and stability of operation can be greatly improved, meanwhile, the damage of radioactive rays to interventional doctors can be effectively reduced, and the occurrence probability of accidents in operation is reduced. At present, the interventional robot is operated manually, and through analysis and study of images, the robot can automatically complete operation, and the operation is the direction of future development like unmanned operation in automobiles.
However, there are several problems in iterative learning for interventional surgical robots: (1) The operation experience of doctors is difficult to copy and inherit, other doctors want to master the abundant operation experience, a great deal of time and practice operation are needed, and the learning efficiency is low. (2) The current interventional robot does not have iterative learning capability, and the interventional robot can only finish partial automatic operation at present, and the operation level is lower in actual operation. Only one-sided operation can be mastered and applied, the operation can be only performed in an attempt mode in the operation, the failure rate is high, and the operation efficiency is low. (3) In the operation process of the interventional operation robot, the interventional operation robot lacks guidance and correct standards for operating devices such as guide wires in a specific vascular environment, and has low operation accuracy. (4) The interventional robot lacks a great deal of accumulation process of operation experience, cannot grasp specific operation methods of different operations, and does not have an analysis process of an optimal method. With the technical development and experience accumulation of interventional operations in recent years, the common interventional operations form mature operation experiences, but due to the lack of a good sharing mechanism, the existing operation experiences cannot be summarized and summarized efficiently and can only become an experience island, which is not beneficial to further optimization and popularization of the operation experiences.
Therefore, there is a large space for iterative learning of the interventional operation robot, which is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the invention provides an iterative learning method and system for an automatic surgical intervention robot and a storage medium, which aim to solve the problems that clinical surgical experience in the present stage cannot be inherited, the intervention robot lacks iterative learning capability, the intervention robot lacks optimal operation methods under different vascular environments, the intervention robot has low automatic surgical operation efficiency and the like.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
in a first aspect, an embodiment of the present invention provides an iterative learning method for an automatic surgical intervention robot, including the steps of:
collecting the latest system operation of individual intervention robot applications of a plurality of hospitals once every interval preset time length
Related information saved by interventional operation;
classifying and storing the collected related information of the individual interventional robots of the hospitals;
according to the related information of different classifications, analyzing and processing are carried out to obtain the optimal operation method under different conditions;
summarizing and packaging various optimal operation methods, and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system.
Further, the interventional robot applies the latest system to operate the relevant information saved by the interventional procedure, including:
the interventional robot records basic information of a surgery, comprising: the type of surgery, patient condition information, vascular status, time of surgery, and outcome;
according to the actual condition of the operation, obtaining a division result of each operation divided into a plurality of stages, starting and ending moments of each stage and an operation target of each stage;
when entering any stage, the blood vessel image information corresponding to the stage is read, and information identification and calculation are carried out; comprising the following steps: calculating an optimal path based on the initial position and the target position, recording the shape of the vascular path, calculating the diameter of each position of the blood vessel, identifying the bifurcation quantity and bifurcation angles of the blood vessel, counting the operation time of the stage, recording the operation result of the stage and judging whether complications exist;
dividing the path into a plurality of sub-stages based on each bifurcation point in the optimal path;
when entering any one of the sub-stages, recording the actual movement condition of the guide wire or catheter head end and the corresponding real-time operation action in the sub-stage;
sequentially completing the record of each sub-stage, and sequentially completing the reading, identifying and calculating operation of each stage;
summarizing the summary information of all stages and sub-stages, and receiving the modification and confirmation information of the manager as the related information stored in the interventional operation.
Further, calculating an optimal path based on the starting and target positions includes:
finding out a branch vessel closest to a next-level vessel corresponding to the target point from the starting point vessel level according to the calibrated vessel classification, taking the branch vessel as an optimal path interval, and recording the interval;
searching step by step until the target point is found; each recording interval is concatenated into a line as the optimal path for this stage.
Further, the classifying and storing the collected related information of the individual interventional robots of the hospitals comprises the following steps:
according to the operation type in each operation basic information, dividing the related information of all individual interventional robots in a plurality of hospitals into three major categories, namely an A heart, a B nerve and a C periphery;
in each large category, the objects are divided into a plurality of small categories, namely P1, P2 and … Pn;
and placing the collected related information of the individual interventional robots of the hospitals in the corresponding small categories under the corresponding large categories, and realizing classified storage.
Further, according to the related information of different classifications, analysis processing is carried out to obtain the optimal operation method under different conditions; comprising the following steps:
analyzing newly entered sub-stages in the same category, analyzing category by category until all data information is completed; the categories include: the periphery of the heart A, the nerve B and the nerve C;
in the sub-stage of newly entering any category, the corresponding vessel diameter and shape are read out, and the optimal path and actual motion path of the guide wire or catheter motion in the sub-stage are read out;
dividing the current category into a plurality of sections according to different blood vessel diameters, and marking the sections as X groups which are X1, X2 and … Xn;
dividing the current category into a plurality of sections according to different blood vessel shapes, and marking the sections as Y groups as groups Y1, Y2 and … Yn;
when each sub-stage is divided into corresponding X and Y groups at the same time, comparing the optimal path and the actual motion path of each sub-stage in the same X group and Y group;
determining success of the sub-stage operation, and calculating the coincidence ratio of the optimal path and the actual motion path of each sub-stage in the group;
finding out the operation method with the highest overlap ratio as the optimal method under the current condition;
and so on until the analysis of the optimal method in all sub-stages is completed;
after analysis is completed, the optimal methods under each condition are stored, and the optimal operation methods under different conditions are obtained.
Further, according to the related information of different classifications, analysis processing is carried out to obtain the optimal operation method under different conditions; further comprises:
after new collected information exists, if the optimal methods exist under the same category, comparing the optimal methods corresponding to the new collected information with the existing optimal methods, and retaining the compared optimal methods.
In a second aspect, an embodiment of the present invention provides an automatic surgical intervention robot iterative learning system, including: the system comprises a cloud processing system and individual intervention robots of a plurality of hospitals;
wherein, the high in the clouds processing system includes:
the data collection module is used for collecting related information stored in the operation intervention operation of the latest system application operation of the individual intervention robots of a plurality of hospitals once every interval preset time;
the data classification module is used for classifying and storing the collected related information of the individual interventional robots of the hospitals;
the analysis obtaining optimal module is used for carrying out analysis processing according to the related information of different classifications to obtain optimal operation methods under different conditions;
the issuing learning module is used for summarizing and packaging various optimal operation methods and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system.
In a third aspect, an embodiment of the present invention provides an automatic surgical intervention robot iterative learning system, including:
at least one processor; and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any one of the automated surgical intervention robot iterative learning methods of the first aspect.
In a fourth aspect, an embodiment of the present invention further provides a storage medium having stored therein instructions, which when run on a terminal, enable an automatic surgical intervention robot iterative learning method according to any one of the first aspects.
The description of the second to fourth aspects of the present invention may refer to the detailed description of the first aspect; also, the advantageous effects described in the second aspect to the fourth aspect may refer to the advantageous effect analysis of the first aspect, and are not described herein.
Compared with the prior art, the invention discloses an iterative learning method of an automatic operation interventional robot, which can continuously summarize and analyze the operation method of the interventional operation, continuously and iteratively update the operation method of the interventional robot, and continuously improve the operation capability of the interventional robot along with the increase of time and operation quantity.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of an iterative learning method of an automatic surgical intervention robot provided by an embodiment of the invention.
Fig. 2 is an overall schematic diagram of iterative learning of an automatic surgical intervention robot according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a summary experience data flow of an individual interventional robot according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a data classification flow according to an embodiment of the present invention.
Fig. 5 is a schematic flow chart of an analysis optimal operation method according to an embodiment of the present invention.
Fig. 6 is a schematic flow chart of an individual interventional robot using a new system according to an embodiment of the present invention.
Fig. 7 is a schematic structural diagram of an iterative learning system of an automatic surgical intervention robot according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1:
the embodiment of the invention discloses an iterative learning method of an automatic surgical intervention robot, which can be applied to a cloud processing system, and is shown with reference to fig. 1, and comprises the following steps:
s10, collecting relevant information stored by the operation intervention operation of the individual intervention robots of a plurality of hospitals by applying the latest system once every interval preset time; the preset time interval can be the frequency of days, weeks, months and the like; the time length can be the same or different, and the information can be collected in real time according to the instruction.
S20, classifying and storing the collected related information of the individual interventional robots of the hospitals;
s30, analyzing and processing according to the related information of different classifications to obtain the optimal operation method under different conditions;
s40, summarizing and packaging various optimal operation methods, and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system.
In this embodiment, the method may periodically or aperiodically collect summary data of operation experience of a plurality of interventional robots in clinical operations of a plurality of hospitals, upload the summary data to a cloud system for analysis and comparison according to different conditions, and find out an optimal operation method at the present stage. And updating the found optimal operation method of each condition into each interventional robot. Each interventional robot can master the latest and best interventional operation method, and the optimal method can be directly called in the new interventional operation process, so that the operation is completed faster and better. The method can continuously summarize and analyze the operation method of the interventional operation, continuously and iteratively update the operation method of the interventional robot, and continuously improve the operation capability of the interventional robot along with the increase of time and operation quantity.
Referring to fig. 2, an overall schematic diagram of an iterative learning method of an automatic surgical intervention robot is shown; a plurality of hospitals are provided with intervention robots for vascular intervention operation, and all individual intervention robots upload operation data to a cloud processing system; or actively collecting by the cloud processing system; the cloud processing system processes the operation data to obtain an optimal scheme and periodically issues the optimal scheme to each intervention robot, so that each intervention robot can have the latest and best operation skills, and future intervention operation can be completed faster and better.
Specifically, first, the individual interventional robots in a plurality of hospitals can apply the latest interventional operation system to perform automatic interventional operation, in the operation process, each interventional robot stores operation information, intra-operation images, patient information and other data, operation experience data of the interventional operation performed by the individual interventional robot is summarized after the operation, and then the operation experience data are sent to a cloud processing system periodically; or actively collected by the cloud processing system.
The cloud processing system classifies and stores the acquired data from each interventional robot according to the standard. And then analyzing and processing all the collected data according to different data types to find out the optimal operation method under different conditions. And summarizing and packaging various optimal operation methods, and periodically issuing the summarized and packaged optimal operation methods to all individual interventional robots. Then, the individual interventional robot updates and learns the latest version after receiving the update.
Furthermore, the interventional robot can perform self-judgment to judge whether to continuously iteratively update the operating system or the cloud processing system to judge whether an individual interventional robot is the latest operating system; if the operating system of the interventional robot has reached a very perfect condition, the own operating system does not need iterative learning and is not updated. Otherwise, the steps S10 to S40 are continuously executed, and the steps are sequentially circulated. Thus, with the delay of time and the increasing of cases, the cloud processing system can find more types of better operation methods, so that the operation capability of the interventional robot is continuously improved.
The following describes each of the above steps in detail:
in one embodiment, step S10 collects relevant information saved by the interventional operation of the individual interventional robots of a plurality of hospitals by applying the latest system operation every interval for a preset period of time; wherein, the relevant information stored in the interventional operation is operated by the interventional robot by using the latest system, and the relevant information comprises:
s101, recording operation basic information by the interventional robot, wherein the operation basic information comprises the following steps of: the type of surgery, patient condition information, vascular status, time of surgery, and outcome;
s102, according to the actual condition of the operation, obtaining a division result of each operation divided into a plurality of stages, starting and ending moments of each stage and an operation target of each stage;
s103, when entering any one of the stages, the blood vessel image information corresponding to the stage is read, and information identification and calculation are carried out; comprising the following steps: calculating an optimal path based on the initial position and the target position, recording the shape of the vascular path, calculating the diameter of each position of the blood vessel, identifying the bifurcation quantity and bifurcation angles of the blood vessel, counting the operation time of the stage, recording the operation result of the stage and judging whether complications exist;
s104, dividing the path into a plurality of sub-stages based on each bifurcation point in the optimal path;
s105, when entering any one of the sub-stages, recording the actual movement condition of the head end of the guide wire or the catheter and the corresponding real-time operation action in the sub-stage;
s106, sequentially completing the recording of each sub-stage, and sequentially completing the reading, identifying and calculating operation of each stage;
and S107, summarizing the summary information of all the stages and sub-stages, and receiving the modification and confirmation information of the manager as the related information stored in the interventional operation.
Referring to fig. 3, in the individual interventional robot summary experience data flow, when the interventional robot completes the summary of the operation, first, basic information in each operation including operation type, patient condition information, blood vessel state, operation time, result and the like is recorded. The information is used as a label of the operation, and can be correspondingly found at any stage in the operation. Then, each operation is divided into several stages according to the actual treatment and treatment conditions of the operation. The starting and ending time and position of each stage are calibrated by assistance of doctors, and the operation target of each stage is calibrated.
For example, the physician marks one of the phases as the derivation of the silk to the anterior descending branch, the start and end times being 15:20 and 18:34, the corresponding positions being the guide catheter exit and the anterior descending branch tip. After dividing the phases in the operation, the interventional robot will analyze each phase one by one in the time sequence of the operation. From the first stage, first, the blood vessel image information of the stage is read, and the identification and calculation of the related information are performed. Comprising the following steps:
1. based on the start and end positions within the phase, an optimal path is calculated. The calculation method is that after the blood vessel classification is calibrated, the branch blood vessel closest to the blood vessel of the next level corresponding to the target point is found out from the blood vessel level of the starting point and is used as the optimal path interval, and the interval is recorded. The search is performed step by step until the target point is found. And connecting each recording interval into a line, namely the optimal path at the stage.
2. The shape of the vascular path is recorded, and after the blood vessel in the stage is identified, the path of the blood vessel is calculated, for example, the blood vessel is in a straight line or the blood vessel is in a 30-degree arc shape.
3. The diameter of each position of the blood vessel is calculated by extracting the central line of the blood vessel, making a vertical line along the central line, finding out a line segment formed by 2 intersection points of the vertical line and the wall of the blood vessel, and calculating the length of the line segment to obtain the diameter of the blood vessel.
4, the bifurcation number of the blood vessel and the angle of each bifurcation are identified.
5. And counting the operation time of the current stage.
6. The outcome of the procedure at this stage and whether complications are present are recorded.
Then, the interventional robot searches all bifurcation points on the optimal path, takes the adjacent bifurcation points as a sub-stage, and divides the path into a plurality of sub-stages. The position of the guidewire or catheter tip is then observed, and each sub-stage is recorded one by one in time sequence within the stage until all sub-stages are completed. In each sub-phase, the actual movement of the guide wire or catheter tip and the corresponding interventional robot operating actions are recorded. After the summary of all the stages is completed, the robot marks the information, integrates the data according to a fixed format and gives the information to the attending doctor for confirmation. If the physician finds a problem, the problem is modified. After confirming that the information is not wrong, the robot packs the summarized information and prepares for uploading.
In this embodiment, the operation experience library of the interventional robot may be continuously updated in an iterative manner over time by periodically or aperiodically collecting data of the operation method, so that the operation experience library is more comprehensive and the optimal method is more complete, thereby improving the operation capability of the interventional robot.
In one embodiment, step S20 classifies and saves the collected related information of the individual interventional robots of the plurality of hospitals; the classification criteria may refer to fig. 4, where in the data classification flow of the cloud processing system, after the cloud system receives the data information sent by each interventional robot:
(1) Firstly, according to the operation types in the basic information of each operation, all operation experiences are divided into 3 large categories which are respectively marked as an A heart, a B nerve and a C periphery.
(2) In each large category, it is divided into several small categories, denoted as P1, P2, … Pn, according to the main object. For example, the main goal is to place the stent may be noted as a class P1, with all phases belonging to the stent placement phase, all divided in P1.
(3) After the newly collected data enter the cloud processing system, the basic operation information is marked, the data firstly enter a large class (ABC), then enter different small classes (Pn) at different stages of each operation, and the information of a plurality of sub-stages is also included in each stage, so that further detailed comparison analysis can be performed.
In one embodiment, step S30 performs analysis processing according to the related information of different classifications to obtain the optimal operation method under different conditions; referring to fig. 5, in a flow of the method for analyzing the optimal operation of the cloud processing system:
1) And analyzing the newly entered sub-stages in the same category by the cloud processing system, and analyzing the sub-stages category by category until all the data information is completed.
2) In the sub-stage of the new entering category, the information of the diameter, the shape and the like of the blood vessel is read, the image information is read, the optimal path of the movement of the guide wire or the guide tube in the sub-stage is found, and meanwhile, the actual movement path of the guide wire or the guide tube is recorded by using different arrays.
3) Based on the read information, the blood vessel is divided into a plurality of sections based on different blood vessel diameters, and for example, a blood vessel having a diameter of about 3mm is denoted as an X1 group.
4) The blood vessel is classified into a plurality of sections based on different blood vessel shapes, and for example, a blood vessel having a curvature of about 30 ° is denoted as a Y1 group.
5) I.e. each sub-phase can be divided into corresponding X and Y groups, and in the same X and Y groups, a comparison is made of all sub-phases in the group.
6) Firstly judging whether the operation fails or not in the group, if so, indicating that the method is not good, and excluding the method.
7) Secondly, in all successful sub-phases, the optimal path and actual motion path overlap ratio for each sub-phase within the group is calculated. Finding the operation method with the highest overlap ratio, taking the method as the optimal method under the current condition, and recording the operation method in the operation action.
And so on until all analysis of the optimal method is completed. After the analysis is completed, the best method under each condition is saved and recorded as the latest scheme. After new acquired data enter the cloud processing system, if the optimal method exists under the same category, the cloud processing system only needs to compare the newly entered acquired data with the original optimal method, and the method of the cloud processing system is better, so that the cloud processing system remains as the better method.
In this embodiment, a large number of operation methods of the interventional operation can be summarized through a large number of operations performed by a plurality of interventional robots in a plurality of hospitals, and the cloud processing system can find an optimal operation method under the same type through comparing and analyzing a large number of operation methods.
In one embodiment, step S40 performs summary packaging on each type of optimal operation method, and issues the summary packaging to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system
Referring to fig. 6, in the new system flow of the individual robot, the interventional robot is used in a clinical operation after a new interventional operation system is installed:
1. firstly, the interventional robot acquires a blood vessel image of the current operation and the like, and after confirming an operation path with the assistance of a doctor, the path can be divided into a plurality of sub-stages based on each bifurcation point.
2. The interventional robot analyzes the sub-phases one by one. After completing one sub-phase, the next sub-phase is entered until after all sub-phases are completed, the procedure is completed.
3. After entering a sub-stage, the interventional robot firstly analyzes the blood vessel characteristics of the sub-stage, calculates the blood vessel shape and the blood vessel diameter, quickly searches whether the operation system of the interventional robot contains the operation with the same characteristics, and if the operation system is found to be the same, performs the operation of the current stage based on the optimal operation method.
4. If the same features are not found, the interventional surgical robot adjusts and attempts to push the guide wire or catheter on the basis of the iterative learning effort on the operation method of the similar vascular features.
5. After the intervention robot tries for 3 times without success, the intervention robot searches for manual intervention of a doctor, the doctor completes the operation through manual operation, and the system records the operation of the doctor and also serves as operation experience; so as to upload the cloud processing system for sharing use of all interventional robots in interventional operations.
In this embodiment, by establishing an intervention operation experience sharing mechanism, the cloud processing system issues the operation method with the latest multiple scenes to each intervention robot, so that the intervention robots of different individuals can master the latest and the latest operation experience, and the experience can be inherited quickly.
According to the iterative learning method of the automatic surgical intervention robot, provided by the embodiment of the invention, through induction of various vascular conditions and corresponding operations in an intervention operation scene, the intervention robot can rapidly use an optimal operation method under different vascular conditions, and the success rate and the efficiency of the operation are improved.
Example 2:
based on the same inventive concept, the automatic surgical intervention robot iterative learning system provided by the invention is described below, and the iterative learning system described below and the iterative learning method described above can be referred to correspondingly.
An automated surgical intervention robot iterative learning system, comprising: the system comprises a cloud processing system and individual intervention robots of a plurality of hospitals;
wherein, the high in the clouds processing system includes:
the data collection module is used for collecting related information stored in the operation intervention operation of the latest system application operation of the individual intervention robots of a plurality of hospitals once every interval preset time;
the data classification module is used for classifying and storing the collected related information of the individual interventional robots of the hospitals;
the analysis obtaining optimal module is used for carrying out analysis processing according to the related information of different classifications to obtain optimal operation methods under different conditions;
the issuing learning module is used for summarizing and packaging various optimal operation methods and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system.
Referring to fig. 7, the iterative learning system of the automatic surgical intervention robot mainly comprises 2 parts, a plurality of individual intervention robots (a hospital a intervention robot, a hospital B intervention robot, a … hospital N intervention robot) and a cloud processing system from the composition point of view.
After the interventional robots are popular, many hospitals will have clinical operations of the interventional robots. Different individual interventional robots encounter different surgical cases in clinical surgery. According to the automatic surgical intervention robot iterative learning system, based on the principle of big data sharing, a plurality of individual robots can respectively summarize respective surgical operation methods, and then the individual robots respectively upload own experiences (surgical operation methods) into a cloud processing system.
And the cloud processing system performs data collection, data classification and data comparison analysis and then transmits the optimized data to all the interventional robots. Specifically, the cloud processing system can compare and analyze the operation methods under different conditions to find out the optimal method after receiving the operation under each condition. And then issuing the optimal methods to a system of each interventional operation robot to upgrade the operating system of the interventional operation robot. Thus, each surgical intervention robot will complete a round of improvement of the operative capability. With the progress of time, the operation capacity of the interventional robot can be continuously improved due to the accumulation of the operation quantity, so that the iterative learning process is realized.
Example 3:
based on the same inventive concept, the invention also provides an automatic surgical intervention robot iterative learning system, comprising:
at least one processor; and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform an automated surgical intervention robot iterative learning method of embodiment 1.
Example 4:
based on the same inventive concept, the embodiment of the present invention further provides a computer readable storage medium, in which instructions are stored, which when executed on a terminal, can implement an automatic surgical intervention robot iterative learning method as in the above embodiment 1.
The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access Memory (Random Access Memory, RAM), a Read-Only Memory (ROM), an erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), a register, a hard disk, an optical fiber, a portable electronic device
Compact disk read-Only Memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the preceding, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an application specific integrated circuit (Application Specific Integrated Circuit, ASIC). In this embodiment, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. An iterative learning method of an automatic surgical intervention robot is characterized by comprising the following steps:
collecting related information stored by the operation intervention operation of the individual intervention robots of a plurality of hospitals by applying the latest system once every interval preset time;
classifying and storing the collected related information of the individual interventional robots of the hospitals;
according to the related information of different classifications, analyzing and processing are carried out to obtain the optimal operation method under different conditions;
summarizing and packaging various optimal operation methods, and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system;
wherein, the related information stored in the interventional operation is operated by the interventional robot by using the latest system, and the related information comprises:
the interventional robot records basic information of a surgery, comprising: the type of surgery, patient condition information, vascular status, time of surgery, and outcome;
according to the actual condition of the operation, obtaining a division result of each operation divided into a plurality of stages, starting and ending moments of each stage and an operation target of each stage;
when entering any stage, the blood vessel image information corresponding to the stage is read, and information identification and calculation are carried out; comprising the following steps: calculating an optimal path based on the initial position and the target position, recording the shape of the vascular path, calculating the diameter of each position of the blood vessel, identifying the bifurcation quantity and bifurcation angles of the blood vessel, counting the operation time of the stage, recording the operation result of the stage and judging whether complications exist;
dividing the path into a plurality of sub-stages based on each bifurcation point in the optimal path;
when entering any one of the sub-stages, recording the actual movement condition of the guide wire or catheter head end and the corresponding real-time operation action in the sub-stage;
sequentially completing the record of each sub-stage, and sequentially completing the reading, identifying and calculating operation of each stage;
summarizing the summarized information of all stages and sub-stages, and receiving the modification and confirmation information of the manager as the related information stored in the interventional operation;
analyzing and processing the related information according to different classifications to obtain optimal operation methods under different conditions; comprising the following steps:
analyzing newly entered sub-stages in the same category, analyzing category by category until all data information is completed; the categories include: the periphery of the heart A, the nerve B and the nerve C;
in the sub-stage of newly entering any category, the corresponding vessel diameter and shape are read out, and the optimal path and actual motion path of the guide wire or catheter motion in the sub-stage are read out;
dividing the current category into a plurality of sections according to different blood vessel diameters, and marking the sections as X groups which are X1, X2 and … Xn;
dividing the current category into a plurality of sections according to different blood vessel shapes, and marking the sections as Y groups as groups Y1, Y2 and … Yn;
when each sub-stage is divided into corresponding X and Y groups at the same time, comparing the optimal path and the actual motion path of each sub-stage in the same X group and Y group;
determining success of the sub-stage operation, and calculating the coincidence ratio of the optimal path and the actual motion path of each sub-stage in the group;
finding out the operation method with the highest overlap ratio as the optimal method under the current condition;
and so on until the analysis of the optimal method in all sub-stages is completed;
after analysis is completed, the optimal methods under each condition are stored, and the optimal operation methods under different conditions are obtained.
2. An automated surgical intervention robot iterative learning method according to claim 1, wherein calculating an optimal path based on the starting and target positions comprises:
finding out a branch vessel closest to a next-level vessel corresponding to the target point from the starting point vessel level according to the calibrated vessel classification, taking the branch vessel as an optimal path interval, and recording the interval;
searching step by step until the target point is found; each recording interval is concatenated into a line as the optimal path for this stage.
3. The iterative learning method of an automatic surgical intervention robot according to claim 1, wherein classifying and storing the collected information about individual intervention robots of a plurality of hospitals comprises:
according to the operation type in each operation basic information, dividing the related information of all individual interventional robots in a plurality of hospitals into three major categories, namely an A heart, a B nerve and a C periphery;
in each large category, the objects are divided into a plurality of small categories, namely P1, P2 and … Pn;
and placing the collected related information of the individual interventional robots of the hospitals in the corresponding small categories under the corresponding large categories, and realizing classified storage.
4. The iterative learning method of an automatic surgical intervention robot according to claim 1, wherein the analysis processing is performed according to the related information of different classifications to obtain the optimal operation method under different conditions; further comprises:
after new collected information exists, if the optimal methods exist under the same category, comparing the optimal methods corresponding to the new collected information with the existing optimal methods, and retaining the compared optimal methods.
5. An automated surgical intervention robot iterative learning system, comprising: the system comprises a cloud processing system and individual intervention robots of a plurality of hospitals;
wherein, the high in the clouds processing system includes:
the data collection module is used for collecting related information stored in the operation intervention operation of the latest system application operation of the individual intervention robots of a plurality of hospitals once every interval preset time;
the data classification module is used for classifying and storing the collected related information of the individual interventional robots of the hospitals;
the analysis obtaining optimal module is used for carrying out analysis processing according to the related information of different classifications to obtain optimal operation methods under different conditions;
the issuing learning module is used for summarizing and packaging various optimal operation methods and issuing the summarized and packaged optimal operation methods to all individual intervention robot systems; for the interventional robot to perform iterative updating of the self-system;
wherein, the related information stored in the interventional operation is operated by the interventional robot by using the latest system, and the related information comprises:
the interventional robot records basic information of a surgery, comprising: the type of surgery, patient condition information, vascular status, time of surgery, and outcome;
according to the actual condition of the operation, obtaining a division result of each operation divided into a plurality of stages, starting and ending moments of each stage and an operation target of each stage;
when entering any stage, the blood vessel image information corresponding to the stage is read, and information identification and calculation are carried out; comprising the following steps: calculating an optimal path based on the initial position and the target position, recording the shape of the vascular path, calculating the diameter of each position of the blood vessel, identifying the bifurcation quantity and bifurcation angles of the blood vessel, counting the operation time of the stage, recording the operation result of the stage and judging whether complications exist;
dividing the path into a plurality of sub-stages based on each bifurcation point in the optimal path;
when entering any one of the sub-stages, recording the actual movement condition of the guide wire or catheter head end and the corresponding real-time operation action in the sub-stage;
sequentially completing the record of each sub-stage, and sequentially completing the reading, identifying and calculating operation of each stage;
summarizing the summarized information of all stages and sub-stages, and receiving the modification and confirmation information of the manager as the related information stored in the interventional operation;
the analysis obtains an optimal module, which is specifically used for:
analyzing newly entered sub-stages in the same category, analyzing category by category until all data information is completed; the categories include: the periphery of the heart A, the nerve B and the nerve C;
in the sub-stage of newly entering any category, the corresponding vessel diameter and shape are read out, and the optimal path and actual motion path of the guide wire or catheter motion in the sub-stage are read out;
dividing the current category into a plurality of sections according to different blood vessel diameters, and marking the sections as X groups which are X1, X2 and … Xn;
dividing the current category into a plurality of sections according to different blood vessel shapes, and marking the sections as Y groups as groups Y1, Y2 and … Yn;
when each sub-stage is divided into corresponding X and Y groups at the same time, comparing the optimal path and the actual motion path of each sub-stage in the same X group and Y group;
determining success of the sub-stage operation, and calculating the coincidence ratio of the optimal path and the actual motion path of each sub-stage in the group;
finding out the operation method with the highest overlap ratio as the optimal method under the current condition;
and so on until the analysis of the optimal method in all sub-stages is completed;
after analysis is completed, the optimal methods under each condition are stored, and the optimal operation methods under different conditions are obtained.
6. An automated surgical intervention robot iterative learning system, comprising:
at least one processor; and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the automated surgical intervention robot iterative learning method of any of claims 1-4.
7. A storage medium having stored therein instructions which, when run on a terminal, enable an automated surgical intervention robot iterative learning method according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311694424.8A CN117393107B (en) | 2023-12-12 | 2023-12-12 | Iterative learning method and system for automatic surgical intervention robot and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311694424.8A CN117393107B (en) | 2023-12-12 | 2023-12-12 | Iterative learning method and system for automatic surgical intervention robot and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117393107A CN117393107A (en) | 2024-01-12 |
| CN117393107B true CN117393107B (en) | 2024-03-15 |
Family
ID=89467038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311694424.8A Active CN117393107B (en) | 2023-12-12 | 2023-12-12 | Iterative learning method and system for automatic surgical intervention robot and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117393107B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104582624A (en) * | 2012-06-29 | 2015-04-29 | 儿童国家医疗中心 | Automated surgical and interventional procedures |
| CN109806004A (en) * | 2019-03-18 | 2019-05-28 | 汪俊霞 | A kind of surgical robot system and operating method based on cloud data technique |
| CN111527556A (en) * | 2017-12-28 | 2020-08-11 | 爱惜康有限责任公司 | Adaptive control program update for surgical devices |
| JP6760676B1 (en) * | 2019-12-18 | 2020-09-23 | Ｊｅインターナショナル株式会社 | Chatbot server device, learning device, chatbot system, chatbot server device operating method, learning device operating method, program, and recording medium |
| CN111805546A (en) * | 2020-07-20 | 2020-10-23 | 中国人民解放军国防科技大学 | Human-multi-robot sharing control method and system based on brain-computer interface |
| CN114917029A (en) * | 2022-07-22 | 2022-08-19 | 北京唯迈医疗设备有限公司 | Interventional surgical robot system, control method and medium |
| CN114948199A (en) * | 2022-05-17 | 2022-08-30 | 天津大学 | Surgical operation auxiliary system and operation path planning method |
| CN115422838A (en) * | 2022-09-01 | 2022-12-02 | 清华大学 | Autonomous learning method, apparatus, device and medium for surgical robot |
| CN116196099A (en) * | 2023-02-23 | 2023-06-02 | 同济大学 | Cardiovascular intervention operation path planning method, system, storage medium and terminal |
| CN116562392A (en) * | 2023-04-17 | 2023-08-08 | 同济大学 | Vascular intervention operation action sensing and learning system, method and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2712607A1 (en) * | 2008-01-25 | 2009-07-30 | Mcmaster University | Surgical guidance utilizing tissue feedback |
| US11158415B2 (en) * | 2017-02-16 | 2021-10-26 | Mako Surgical Corporation | Surgical procedure planning system with multiple feedback loops |
| US20210065870A1 (en) * | 2019-09-04 | 2021-03-04 | Medtech S.A. | Robotically-assisted surgical procedure feedback techniques based on care management data |
-
2023
- 2023-12-12 CN CN202311694424.8A patent/CN117393107B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104582624A (en) * | 2012-06-29 | 2015-04-29 | 儿童国家医疗中心 | Automated surgical and interventional procedures |
| CN111527556A (en) * | 2017-12-28 | 2020-08-11 | 爱惜康有限责任公司 | Adaptive control program update for surgical devices |
| CN109806004A (en) * | 2019-03-18 | 2019-05-28 | 汪俊霞 | A kind of surgical robot system and operating method based on cloud data technique |
| JP6760676B1 (en) * | 2019-12-18 | 2020-09-23 | Ｊｅインターナショナル株式会社 | Chatbot server device, learning device, chatbot system, chatbot server device operating method, learning device operating method, program, and recording medium |
| CN111805546A (en) * | 2020-07-20 | 2020-10-23 | 中国人民解放军国防科技大学 | Human-multi-robot sharing control method and system based on brain-computer interface |
| CN114948199A (en) * | 2022-05-17 | 2022-08-30 | 天津大学 | Surgical operation auxiliary system and operation path planning method |
| CN114917029A (en) * | 2022-07-22 | 2022-08-19 | 北京唯迈医疗设备有限公司 | Interventional surgical robot system, control method and medium |
| CN115422838A (en) * | 2022-09-01 | 2022-12-02 | 清华大学 | Autonomous learning method, apparatus, device and medium for surgical robot |
| CN116196099A (en) * | 2023-02-23 | 2023-06-02 | 同济大学 | Cardiovascular intervention operation path planning method, system, storage medium and terminal |
| CN116562392A (en) * | 2023-04-17 | 2023-08-08 | 同济大学 | Vascular intervention operation action sensing and learning system, method and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117393107A (en) | 2024-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113748468B (en) | System and method for filling out postoperative report of surgical operation, computer readable medium | |
| CN110996748B (en) | Machine learning-oriented surgical video analysis system | |
| CN100517344C (en) | Method and apparatus for medical intervention procedure planning | |
| WO2015066565A1 (en) | System and method for a situation and awareness-based intelligent surgical system | |
| CN110604616B (en) | Interventional operation path planning method and system based on graph search and electronic equipment | |
| CN112074912B (en) | Interactive coronary artery labeling using interventional X-ray images and depth learning | |
| CN113808125A (en) | Medical image processing method, focus type identification method and related product | |
| CN117393107B (en) | Iterative learning method and system for automatic surgical intervention robot and storage medium | |
| CN115565667A (en) | Prediction method, device, equipment and medium for success rate of CTO interventional therapy | |
| CN112599214A (en) | Method and system for extracting ICD (interface control document) codes from electronic medical records | |
| CN110853716B (en) | Medical record template creation method and device | |
| CN115905960B (en) | Adverse event detection method and device based on ventricular assist device | |
| US20220409309A1 (en) | Method for generating an actuation signal for a robotic system for robot-assisted navigation of a medical object and overall system | |
| CN113710190A (en) | Device and method for evaluating moving path of guide wire | |
| JP2008027093A (en) | Medical practice support method and medical practice support device | |
| CN113792798B (en) | Model training method and device based on multi-source data and computer equipment | |
| CN113392633B (en) | Medical named entity identification method, device and storage medium | |
| CN116052176A (en) | Text extraction method based on cascade multitask learning | |
| CN114464315A (en) | Cerebral apoplexy typing method and device based on blood vessel image completion | |
| CN113506003A (en) | Method, device and equipment for determining student level label and storage medium | |
| CN108803994B (en) | Retinal blood vessel management method and retinal blood vessel management device | |
| EP4165593A1 (en) | Image segmentation for sets of objects | |
| JP2013069111A (en) | Information processor and program | |
| CN118662230B (en) | Three-dimensional modeling intelligent planning method for coronary intervention operation path | |
| US20230190212A1 (en) | Actuation method for x-ray device and x-ray device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |