CN-121980827-A - Digital twin rapid modeling and simulation system for small medical scene

CN121980827ACN 121980827 ACN121980827 ACN 121980827ACN-121980827-A

Abstract

The invention provides a digital twin rapid modeling and simulation system for small medical scenes, which relates to the technical field of image processing, is based on multi-source perception data and a local space reconstruction mechanism, the local digital twin model is constructed, and the quick modeling and dynamic updating of the medical operation scene are realized through the real-time mapping of key parameters such as the body position, the instrument posture, the propulsion depth and the like of a patient. The system introduces an operation path stability assessment, an instant operation risk response assessment and a twin model effectiveness judgment mechanism, adopts a weighted fusion algorithm to carry out joint calculation on multidimensional parameters, combines threshold judgment and strategy triggering, and realizes real-time identification and intervention on path deviation risks, instrument organization action risks and model failure risks. The invention can realize rapid construction, self-adaptive correction and simulation support of the digital twin model in a small medical scene, and improves the safety, stability and system applicability of interventional operation.

Inventors

LI HUINAN
FENG XUE
ZHANG XIAOKANG
XU KE
Lv Fuchen

Assignees

荣科科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260407

Claims (10)

1. A digital twin rapid modeling and simulation system for a medical small scene is characterized by comprising the following specific steps: The data acquisition module is configured to perform multi-source data acquisition and unified processing on a small medical scene with puncture operation as a core in the bedside interventional treatment process; collecting ultrasonic gray image data and echo intensity distribution information of a target tissue, collecting a patient body position deflection angle value, an interventional instrument axial deflection angle value and an interventional instrument propulsion depth value, collecting an equipment output intensity value of interventional equipment in the current operation process, and forming a standardized original data set; the image processing module is configured to extract boundary coordinate information of the target tissue for the ultrasonic gray image data to form a target tissue boundary coordinate set, and acquire equivalent depth parameters of the target tissue by combining the echo intensity distribution information; the local digital twin model building module is configured to reconstruct a local geometric structure of a target tissue in an intervention operation direction by adopting a modeling method combining parameter-driven local geometric modeling and physical action mapping, model a track and gesture evolution of an intervention instrument propulsion path under a unified space coordinate system, and build a local digital twin model; The interventional operation path stability evaluation module is configured to calculate an operation path stability coefficient through a weighted linear stability evaluation algorithm, judge the stability state of the current interventional operation path and output a corresponding path adjustment strategy according to a comparison result of the operation path stability coefficient and a preset path stability threshold; The immediate operation risk response evaluation module is configured to calculate and acquire an immediate operation risk response coefficient based on the intervention operation path stability evaluation, judge the immediate risk state of the current intervention operation and execute a corresponding risk intervention strategy according to a comparison result of the immediate operation risk response coefficient and a preset risk response threshold; The digital twin model effectiveness judging module is configured to trigger a model effectiveness judging mechanism, calculate a twin model effectiveness coefficient, compare the twin model effectiveness coefficient with a model effectiveness threshold, generate an updated local digital twin model when the applicability of the current local digital twin model is insufficient, and re-input the updated local digital twin model to the intervention operation path stability evaluating module to form an adaptive updating closed loop of the digital twin model.
2. The digital twin rapid modeling and simulation system for small medical scenes according to claim 1, wherein the data acquisition module comprises a local anatomy image acquisition unit, a gesture acquisition unit, an intervention equipment running state acquisition unit and a data preprocessing unit; The system comprises a local anatomy image acquisition unit, a target puncture area acquisition unit, a target tissue acquisition unit and a target tissue acquisition unit, wherein the local anatomy image acquisition unit is used for carrying out real-time monitoring and image data acquisition on a local anatomy area related to a single intervention operation in a small medical scene for guiding the puncture operation in real time in the bedside intervention treatment process; The posture acquisition unit is used for acquiring posture deflection angle data corresponding to the current posture of a patient to obtain the posture deflection angle value by installing a posture sensing device on a body surface or a bedside supporting structure of the patient to sense the posture change of the patient in real time in the interventional operation process; The intervention equipment operation state acquisition unit is used for establishing communication connection with a control interface of the intervention equipment, reading equipment output parameter data in real time, and acquiring output power or energy intensity data of the intervention equipment in the current operation process to obtain an equipment output intensity value; the data preprocessing unit is used for carrying out unified processing on the acquired multi-source data, carrying out time alignment on the data through time synchronization processing, and carrying out preliminary noise suppression and format unified processing on the acquired data to form a standardized original data set.
3. The digital twin rapid modeling and simulation system for the small medical scene as defined in claim 1, wherein the image processing module is used for clipping an image area related to interventional operation by adopting a local area clipping algorithm based on ultrasonic gray image data of a target tissue to obtain an image of the target tissue area, enhancing gray level change of a tissue boundary area by adopting an edge enhancement processing method, analyzing the enhanced image by adopting a contour extraction algorithm, extracting boundary coordinate information of the target tissue to form a boundary coordinate set of the target tissue, and calculating equivalent spatial depth of the target tissue in the current operation direction based on the boundary coordinate set of the target tissue in combination with echo intensity distribution information to obtain equivalent depth parameters of the target tissue.
4. The digital twin rapid modeling and simulation system for small medical scenes according to claim 1, wherein the local digital twin model building module comprises a gesture space mapping unit and a model building unit; The system comprises a gesture space mapping unit, a intervention operation space parameter set, a gesture space mapping unit, a gesture operation space parameter set and a gesture operation space parameter set, wherein the gesture space mapping unit is used for extracting a body position deflection angle value, an instrument axial deflection angle value and an instrument propulsion depth value in a standardized original data set as gesture and displacement basic parameters of the intervention operation; The model building unit is used for reconstructing a local geometric structure of a target tissue in an intervention operation direction by adopting a modeling method combining parameter-driven local geometric modeling and physical action mapping based on equivalent depth parameters of the target tissue, an intervention operation space parameter set and an equipment output intensity value, performing track modeling and gesture evolution description of an intervention instrument propulsion path in a unified space coordinate system, performing space action relation mapping between the equipment output intensity value and an instrument action area, and constructing a local digital twin model facing a current medical small scene.
5. The digital twin rapid modeling and simulation system for small medical scenes according to claim 1, wherein the interventional operation path stability assessment module comprises a first calculation unit and a first analysis unit; The first calculation unit is used for carrying out joint calculation on multidimensional postures and depth parameters which influence path stability in the interventional operation process by adopting a weighted linear stability evaluation algorithm based on a space geometrical relationship and a parameter mapping result in the local digital twin model, and carrying out normalization processing on a posture deflection angle value, an instrument axial deflection angle value and an interventional instrument propulsion depth and a target tissue equivalent depth parameter to calculate and obtain an operation path stability coefficient.
6. The digital twin fast modeling and simulation system for small medical scenes according to claim 5, wherein the first analyzing unit is configured to obtain a first evaluation result by comparing an operation path stability coefficient with a path stability threshold value through a preset path stability threshold value, and includes: When the stability coefficient of the operation path is more than or equal to the path stability threshold, the stability of the current intervention operation path is qualified, an effective path state parameter set is generated, and continuous monitoring is carried out; When the operation path stability coefficient is smaller than the path stability threshold, the current intervention operation path stability is unqualified, the intervention path deviation risk and the tissue accidental injury risk exist, a first early warning instruction is triggered, a first strategy is generated, quantitative path adjustment is conducted, the current intervention instrument propulsion speed is reduced by 10% -30% on the basis of original setting so as to slow down the path deviation accumulation risk, the patient body position angle is prompted to be adjusted to be within a correction range of +/-5% -15% of the reference body position angle according to the deviation degree of the body position deflection angle value relative to the reference body position angle at the starting moment of the intervention operation, dynamic limitation is applied to the intervention instrument propulsion depth, the follow-up single propulsion depth change amplitude is controlled within a range of 70% -90% of an original planning value, and recalculation is conducted after adjustment until the operation path stability coefficient is larger than or equal to the path stability threshold.
7. The digital twin fast modeling and simulation system for small medical scenes according to claim 1, wherein the immediate operational risk response assessment module comprises a second calculation unit and a second analysis unit; The second calculation unit is used for carrying out joint calculation on the multidimensional parameters affecting the instant risk level of the current interventional operation by adopting a weighted risk response evaluation algorithm on the instrument organization action relation and the equipment operation state parameter in the local digital twin model based on the effective path state parameter set, carrying out dimensionless normalization processing on the equipment output intensity value, the interventional instrument axial deflection angle value and the operation path stability coefficient, and calculating to obtain the instant operation risk response coefficient.
8. The digital twin rapid modeling and simulation system for small medical scenes according to claim 7, wherein the second analysis unit is configured to obtain a second evaluation result by presetting a risk response threshold and comparing an immediate risk response coefficient with the risk response threshold, and comprises: when the instant operation risk response coefficient is less than or equal to the risk response threshold value, the current intervention operation is in a risk controllable state, and the current operation flow is allowed to be continuously executed for continuous monitoring; When the immediate operation risk response coefficient > risk response threshold value, the immediate risk exists in the current interventional operation, the rising state of the risk of the action of the instrument and the tissue caused by the superposition of the factors of the output intensity, the deflection state of the instrument or the path stability of the device is represented, a second early warning instruction is triggered, a second strategy is generated, the output intensity value of the device is reduced by 15% -40% on the basis of the current setting so as to weaken the action energy of the instrument, the immediate freezing control is carried out on the propelling action of the interventional instrument, further path deviation or tissue action accumulation is prevented, after adjustment, the calculation is carried out again until the immediate operation risk response coefficient is less than or equal to the risk response threshold value, and a model effectiveness judging mechanism is started.
9. The digital twin rapid modeling and simulation system for small medical scenes according to claim 1, wherein the digital twin model effectiveness judging module comprises a third computing unit and a third analyzing unit; The third calculation unit is used for starting a model effectiveness judging mechanism and is based on a standardized original data set, a timestamp index parameter backtracking and difference calculating method is adopted, a body position deflection angle value corresponding to the current moment and a body position deflection angle value corresponding to the local digital twin model building moment are compared, angle change amounts between the two are calculated, body position deflection angle change amounts are obtained, a time consistency parameter matching and difference calculating method is adopted, an instrument axial deflection angle value at the current intervention operation moment and an instrument axial deflection angle value corresponding to the local digital twin model building moment are compared, a deflection angle difference between the two is calculated, an instrument axial deflection angle change amount is obtained, a difference calculation is carried out on an intervention instrument propulsion depth value at the current moment and a propulsion depth value corresponding to the local digital twin model building moment, a propulsion depth change amount is obtained, normalized change amount of an intervention instrument propulsion depth is obtained, and after dimensionless normalization processing is carried out, a twin model effectiveness coefficient is calculated and obtained.
10. The digital twin rapid modeling and simulation system for small medical scenes according to claim 9, wherein the third analysis unit is configured to obtain a third evaluation result by presetting a model validity threshold and comparing a twin model validity coefficient with the model validity threshold, and comprises: when the availability factor of the twin model is more than or equal to the model availability threshold, the applicability of the current local digital twin model is qualified, and continuous monitoring is performed; When the effectiveness coefficient of the twin model is smaller than the effectiveness threshold value of the model, the applicability of the current local digital twin model is unqualified, the risk of model failure exists, a third early warning instruction is triggered, a third strategy is generated, namely, the corresponding body position deflection angle, instrument posture parameter or propulsion depth parameter is collected again only for the parameter type causing the effectiveness of the twin model to be reduced, the affected local geometric structure and the space mapping relation in the local digital twin model are quickly rebuilt based on updated parameter data, an updated local digital twin model is generated, the updated local digital twin model is input to an intervention operation path stability evaluation module again, and the calculation flow of the operation path stability coefficient is executed again to form a digital twin model self-adaption updating closed loop.

Description

Digital twin rapid modeling and simulation system for small medical scene Technical Field The invention relates to the technical field, in particular to a digital twin rapid modeling and simulation system for a small medical scene. Background With the development of interventional therapy, bedside minimally invasive procedures and accurate medical techniques, medical procedures are evolving from traditional empirical dominant modes to data-driven and model-aided decision-making modes. The digital twin technology provides a new technical means for risk assessment, path planning and decision support in the medical operation process by modeling and simulating the physical object, the operation process and the running state thereof. The digital twin research and application in the existing medical field mainly focuses on complete organ modeling, whole body structure modeling or large medical equipment operation simulation, generally depends on high-precision image data, multi-mode physiological parameters and a complex physical modeling method, and has the characteristics of long model construction period, high data acquisition requirement and high calculation complexity. The technology is more suitable for preoperative planning, scientific research analysis or macroscopic management evaluation, and is difficult to meet the application requirements of real-time, light weight and instant feedback in small medical scenes such as bedside interventional therapy and the like. During actual bedside intervention procedures, a single intervention often involves only a local anatomical region, a single interventional instrument, and continuous operational adjustments for a short period of time. Meanwhile, the body position change of the patient, the posture adjustment of the instrument and the output state of the equipment can be changed rapidly in the operation process, so that the traditional static or one-time built digital twin model is difficult to continuously adapt to the current operation state. In the prior art, the defects that a digital twin model is high in dependence on data integrity, is difficult to quickly establish through limited acquisition data in a small scene, a simulation target is focused on scientific research analysis or macroscopic evaluation, path stability, instant risk and the like which face a specific interventional operation process are lacked, quantitative feedback which can directly guide operation is provided, a model updating mechanism is imperfect, when an operation state changes, an integral reconstruction model is often needed, real-time performance and clinical usability are insufficient, a system structure is complex, deployment cost is high, a using threshold is high, and popularization and application in high-frequency medical small scenes such as bedside intervention are difficult. Disclosure of Invention The invention aims to provide a digital twin rapid modeling and simulation system for a small medical scene, which aims to solve the problems in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions: a digital twin rapid modeling and simulation system for medical small scenes specifically comprises: The data acquisition module is configured to perform multi-source data acquisition and unified processing on a small medical scene with puncture operation as a core in the bedside interventional treatment process; collecting ultrasonic gray image data and echo intensity distribution information of a target tissue, collecting a patient body position deflection angle value, an interventional instrument axial deflection angle value and an interventional instrument propulsion depth value, collecting an equipment output intensity value of interventional equipment in the current operation process, and forming a standardized original data set; the image processing module is configured to extract boundary coordinate information of the target tissue for the ultrasonic gray image data to form a target tissue boundary coordinate set, and acquire equivalent depth parameters of the target tissue by combining the echo intensity distribution information; the local digital twin model building module is configured to reconstruct a local geometric structure of a target tissue in an intervention operation direction by adopting a modeling method combining parameter-driven local geometric modeling and physical action mapping, model a track and gesture evolution of an intervention instrument propulsion path under a unified space coordinate system, and build a local digital twin model; The interventional operation path stability evaluation module is configured to calculate an operation path stability coefficient through a weighted linear stability evaluation algorithm, judge the stability state of the current interventional operation path and output a corresponding path adjustment strategy according to a comparison result of t