Model fusion system for preoperative and intraoperative image data of cardiac surgery cardiac intervention

Abstract

The invention discloses a model fusion system for image data before and during cardiac intervention in cardiac surgery, and relates to the technical field of fusion of cardiac surgery image data models. An image acquisition module is used for acquiring image data before and after cardiac intervention, and a fusion and processing module is used for acquiring image data after cardiac intervention; and sequentially calculating and outputting an image data preliminary fusion feature value RT1, an optimized image data fusion feature value RT2 and an adjusted preoperative image data feature value QTnew, and based on the difference between the image data preliminary fusion feature value RT1 and the adjusted preoperative image data feature value QTnew, adjusting the preoperative image data of the next same type of cardiac interventional operation. According to the method, the accuracy and safety of the operation are improved, the image data processing flow is optimized, the flexibility and expandability of the system are enhanced, accurate fusion and optimization of image data models before and during the operation can be achieved through overall cooperation of the system, and more reliable and efficient image data support is provided for the cardiac interventional operation.

Description

Technical Field

[0001] The present invention relates to the technical field of fusion of cardiac surgery image data models, and in particular to a model fusion system for preoperative and intraoperative image data of cardiac surgery. Background Art

[0002] With the continuous development of medical technology, cardiac interventional surgery has become an important means of treating heart disease. However, cardiac interventional surgery has extremely high requirements for the accuracy and real-time performance of image data. Therefore, model fusion technology of preoperative and intraoperative image data has emerged.

[0003] Although existing image data processing methods have achieved certain results in cardiac interventional surgery, there are still many problems and shortcomings. Specifically, existing image data processing methods often rely only on a single preoperative or intraoperative image data, resulting in incomplete information, which makes it difficult to accurately reflect the true structure and functional status of the heart. Due to the lack of effective fusion algorithms and formulas, existing image data fusion methods often have the problem of insufficient accuracy, resulting in low quality of fused image data and difficulty in flexible adjustment, thereby affecting the accuracy of surgical decisions. In addition, traditional image data processing methods often lack an effective feedback mechanism, making it difficult to evaluate the fusion quality and guide subsequent image data adjustments. In addition, it should be noted that intraoperative image data are often interfered with by noise, and these interference factors will affect the accuracy and reliability of image data. Summary of the invention

[0004] The purpose of the present invention is to provide a model fusion system for image data before and during cardiac intervention in cardiac surgery, which solves the problems raised in the above-mentioned background technology.

[0005] To achieve the above purpose, the present invention provides the following technical solution, and the specific steps of implementing model fusion and adjustment are as follows: Step 1: Use the image acquisition module to collect image data before and after cardiac surgery, the image data including features of preoperative and intraoperative image data, key features, emphasis on surgical sites, and noise levels during surgery; Step 2: Using the fusion and processing module, calculate the output image data preliminary fusion feature value RT1, optimized image data fusion feature value RT2, and adjusted preoperative image data feature value QT new ; Step 3: Preliminary fusion of the eigenvalue RT1 based on the image data and the adjusted eigenvalue QT of the preoperative image data new The difference between the two images is calculated and the result output module is used to adjust the preoperative image data of the next cardiac interventional surgery of the same type. Wherein, the fusion and processing module includes a unit for realizing preliminary fusion of preoperative and intraoperative image data, a unit for optimizing intraoperative key features based on preliminary fusion, and a unit for providing guidance for adjusting preoperative image data; The image acquisition module includes a preoperative acquisition unit and an intraoperative acquisition unit. The preoperative acquisition unit extracts the features, key features, and emphasis of the surgical site of the preoperative image data. The intraoperative acquisition unit acquires the features, key features, and noise level of the intraoperative image data in real time and stores them.

[0006] Optionally, the calculation formula for implementing the preliminary fusion unit of preoperative and intraoperative image data is as follows: RT1=QT×(1+(CZ / 10))-|TCY / TCY max |×QT avg ; QT=SQRT(QT x 2 +QT y 2 +QT z 2 ); TCY=|QT key -HT key |; QT avg =(QT1+QT2+QT3+......+QT n ) / n; in: Initial fusion eigenvalues ​​of RT1 image data; QT is the characteristic value of preoperative image data; QT x , QT y and QT z They are the characteristic values ​​of pixels in three dimensions (x, y, z) of the preoperative image data; CZ is the emphasis weight value, which is flexibly adjusted according to the surgical site, and its value range is {0-1}, 0 means that the surgical emphasis is not considered, and 1 means that the surgical emphasis is important; TCY is the key feature difference value, which reflects the degree of difference between preoperative and intraoperative image data in key features; QT key is the key characteristic value before operation, HT key is the key characteristic value during the operation, and the key characteristic value before the operation is QT key and intraoperative key feature value HT key The calculation formula is the same as that of the characteristic value QT of preoperative image data; TCY max is the maximum difference value of the key feature; QT avg is the average eigenvalue of preoperative image data; n is the total amount of preoperative image data; QT1 is the first image data characteristic value, QT2 is the second image data characteristic value, QT3 is the third image data characteristic value, QT n is the nth image data feature value.

[0007] Optionally, the calculation formula of the key feature unit in the preliminary fusion optimization is as follows: RT2=(RT1+QT key ×SQRT(CZ))-[(RT1×(1-QT key / QT key,max )) / 2]-[(ZS / 10)×SQRT(RT1 2 / (QT key 2 +1))]; ZS=YHT key -HT key ; in: RT2 is the optimized image data fusion feature value; QT key,max is the maximum critical characteristic value before surgery; ZS is the noise level of intraoperative image data; YHT key Expected values ​​of key intraoperative features.

[0008] Optionally, the calculation formula of the preoperative image data adjustment guidance unit is as follows: QT new =[RT2+(RT1×QT) / QT max ]-[(RT2×(1-RT1 / RT1 max )) / CZ]-SQRT[(RT2 2 -RT1 2 )+TCY 2 ]; in: QT new is the adjusted feature value of preoperative image data; QT max is the maximum eigenvalue of the preoperative image data; RT1 max Initially fuse the maximum eigenvalue for the image data.

[0009] Optionally, based on the adjusted preoperative image data characteristic value QT new The fusion analysis of the preliminary fusion feature value RT1 of the image data is as follows: If QT new =RT1, which reflects that the fusion effect of preoperative and intraoperative image data is good, and the preoperative and intraoperative image data have achieved good matching in key features; If QT new ≠RT1, it reflects that there is a deviation in the fusion process of preoperative and intraoperative image data, and the preoperative and intraoperative image data do not achieve a good match in key features. When the feature value QT of the preoperative image data after adjustment is new , adjust the preoperative image data.

[0010] Optionally, the preoperative key characteristic value QT key The calculation formula is as follows: QT key =SQRT(QT key，x 2 +QT key，y 2 +QT key，z 2 ); QT key，x , QT key，y and QT key，z They are the characteristic values ​​of pixels in the key areas of the preoperative image data in three dimensions (key,x,key, y, key,z); The intraoperative key characteristic value HT key The calculation formula is as follows: HT key =SQRT(HT key，x 2 +HT key，y 2 +HT key，z 2 ); HT key，x , HT key，y and HT key，z They are the feature values ​​of pixels in the key areas of the intraoperative image data in three dimensions (key, x, key, y, key, z).

[0011] Optionally, the devices used by the image acquisition module include imaging devices and storage devices; The equipment used by the fusion and processing module includes computer processing equipment; The devices used by the result output module include input and output devices.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The preliminary fusion unit of the preoperative and intraoperative image data of the present invention improves the comprehensiveness of the data by fusing the feature values ​​of the preoperative image data in three dimensions and considering the weight values ​​of the main focus of the surgery. At the same time, by introducing the key feature difference value TCY and the key feature maximum difference value TCY max Normalization is performed to further improve the accuracy and reliability of the data.

[0013] 2. The present invention optimizes the intraoperative image data after preliminary fusion by introducing the eigenvalues ​​of key features and noise level parameters based on preliminary fusion optimization of intraoperative key feature units. By adjusting these parameters, the fusion accuracy can be further improved and noise interference can be reduced.

[0014] Among them, in the key feature unit based on preliminary fusion optimization, by introducing the intraoperative image data noise level ZS and the preoperative key feature value QT key The square parameter of the image is used to effectively suppress noise interference, which helps to improve the accuracy and reliability of intraoperative image data and provide more reliable support for surgical decision-making.

[0015] 3. The preoperative image data adjustment guidance unit of the present invention compares the adjusted preoperative image data characteristic value QT new The value of the eigenvalue RT1 is initially fused with the image data, thereby establishing an effective feedback mechanism that can evaluate the fusion quality of preoperative and intraoperative image data and guide subsequent image data adjustments, thereby directly affecting the processing of the next round of preoperative image data, thereby achieving continuous optimization and improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a method flow chart of the model fusion system for preoperative and intraoperative image data; Figure 2 This is a schematic diagram of the overall structure of the model fusion system for preoperative and intraoperative image data; Figure 3 It is a structural schematic diagram of the image acquisition module of the present invention; Figure 4 It is a structural schematic diagram of the fusion and processing module of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The model fusion system of preoperative and intraoperative image data of cardiac surgery is different from the existing image data processing methods. The existing image data processing methods are difficult to accurately reflect the real structure and functional status of the heart, and lack effective fusion algorithms and formulas, and thus cannot be flexibly adjusted and guide subsequent image data adjustment work. The purpose of this algorithm unit is to improve the accuracy and safety of the operation, optimize the image data processing process, and enhance the flexibility and scalability of the system. Through the overall collaboration of the system, it can achieve accurate fusion and optimization of preoperative and intraoperative image data models, providing more reliable and efficient image data support for cardiac interventional surgery.

[0019] For example, see Figures 1 to 4 This embodiment provides a model fusion system for preoperative and intraoperative image data of cardiac surgery. The specific steps for implementing model fusion and adjustment are as follows: Step 1: Use the image acquisition module to collect image data before and after cardiac surgery, the image data including features of preoperative and intraoperative image data, key features, emphasis on surgical sites, and noise levels during surgery; Step 2: Using the fusion and processing module, calculate the output image data preliminary fusion feature value RT1, optimized image data fusion feature value RT2, and adjusted preoperative image data feature value QT new ; Step 3: Preliminary fusion of the eigenvalue RT1 based on the image data and the adjusted eigenvalue QT of the preoperative image data new The difference between the two images is calculated and the result output module is used to adjust the preoperative image data of the next cardiac interventional surgery of the same type. The fusion and processing module includes a unit for realizing preliminary fusion of preoperative and intraoperative image data, a unit for optimizing intraoperative key features based on preliminary fusion, and a unit for providing guidance for adjusting preoperative image data; The image acquisition module includes a preoperative acquisition unit and an intraoperative acquisition unit. The preoperative acquisition unit extracts the features, key features, and emphasis of the surgical site of the preoperative image data. The intraoperative acquisition unit acquires the features, key features, and noise level of the intraoperative image data in real time and stores them. The devices used in the image acquisition module include imaging devices and storage devices; The equipment used in the fusion and processing module includes computer processing equipment; The devices used by the result output module include input and output devices.

[0020] In this embodiment, the system cooperates with three algorithm units and combines RT1, RT2 and QT newThe three operation results realize the fusion and optimization of preoperative and intraoperative image data, and provide more accurate and efficient image data support for cardiac interventional surgery. Specifically, RT1 is the initial fusion feature value of image data, which is obtained through mathematical operation. This value not only contains the anatomical information of preoperative image data, but also integrates the specific needs of the operation to a certain extent, ensuring the stability and reliability of the fusion result. RT2 is the fusion feature value of optimized image data, which can extract and optimize the key features during the operation and reduce noise interference. This process helps to improve the clarity and accuracy of image data and provide more reliable image support for surgical operations. In addition, the RT2 algorithm also adopts nonlinear processing methods to further improve the fusion effect of image data. new To adjust the feature value of preoperative image data, this value not only reflects the current operation situation, but also provides more accurate and efficient image data support for the next operation. new The calculation results of QT can also affect the calculation of RT1 and RT2, making the three algorithms of this system have significant beneficial effects in the model fusion system of image data before and during cardiac intervention in cardiac surgery. new The cyclic impact mechanism established under the comparison results of RT1 provides strong support for the continuous learning and improvement of the system, helps to promote technological innovation and development, and improves the success rate and safety of surgery.

[0021] See also Figures 1 to 4 , the calculation formula for realizing the preliminary fusion unit of preoperative and intraoperative image data is as follows: RT1=QT×(1+(CZ / 10))-|TCY / TCY max |×QT avg ; QT=SQRT(QT x 2 +QT y 2 +QT z 2 ); TCY=|QT key -HT key |; QT avg =(QT1+QT2+QT3+......+QT n ) / n; in: Initial fusion eigenvalues ​​of RT1 image data; QT is the characteristic value of preoperative image data; QT x , QT y and QT zThey are the characteristic values ​​of pixels in three dimensions (x, y, z) of the preoperative image data; CZ is the emphasis weight value, which is flexibly adjusted according to the surgical site, and its value range is {0-1}, 0 means that the surgical emphasis is not considered, and 1 means that the surgical emphasis is important; TCY is the key feature difference value, which reflects the degree of difference between preoperative and intraoperative image data in key features; QT key is the key characteristic value before operation, HT key is the key characteristic value during the operation, and the key characteristic value before the operation is QT key and intraoperative key feature value HT key The calculation formula is the same as that of the characteristic value QT of preoperative image data; TCY max is the maximum difference value of the key feature; QT avg is the average eigenvalue of preoperative image data; n is the total amount of preoperative image data; QT1 is the first image data characteristic value, QT2 is the second image data characteristic value, QT3 is the third image data characteristic value, QT n is the nth image data feature value; Preoperative key characteristic value QT key The calculation formula is as follows: QT key =SQRT(QT key，x 2 +QT key，y 2 +QT key，z 2 ); QT key，x , QT key，y and QT key，z They are the characteristic values ​​of pixels in the key areas of the preoperative image data in three dimensions (key,x,key, y, key,z); Intraoperative key feature value HT key The calculation formula is as follows: HT key =SQRT(HT key，x 2 +HT key，y 2 +HT key，z 2 ); HT key，x , HT key，y and HT key，zThey are the feature values ​​of pixels in the key areas of the intraoperative image data in three dimensions (key, x, key, y, key, z).

[0022] In this embodiment: First, in this algorithm unit, "SQRT (QT x 2 +QT y 2 +QT z 2 )” calculation part represents the square root of the Euclidean distance of the eigenvalues ​​of the preoperative image data in three-dimensional space (x, y, z), that is, the modulus of the space vector. It is used to comprehensively evaluate the feature strength of the preoperative image data in each dimension. As the main component of the preliminary fusion unit of the preoperative and intraoperative image data, it reflects the overall feature strength of the preoperative image data and is combined with the focus weight value CZ to jointly affect the calculation of the preliminary fusion feature value RT1 of the fused image data. The calculation part of "(1+(CZ / 10))" adjusts the influence of preoperative image data on fusion results by dividing the emphasis weight value CZ by 10 and adding a constant 1. The larger the emphasis weight value CZ is, the more attention the surgeon pays to the area, and the greater the influence of the fusion result on the area. As a weight factor, CZ adjusts the contribution of preoperative image data in the fusion process to ensure that the main emphasis of the surgery is reflected in the fusion result. This algorithm unit comprehensively considers the eigenvalues ​​of the preoperative image data in three dimensions and the weight values ​​of the main focus of the operation, so that the unit for realizing the preliminary fusion of the preoperative and intraoperative image data can generate a preliminary fused intraoperative image data eigenvalue and a preliminary fusion eigenvalue RT1 of the image data. This process not only considers the spatial distribution of the image data, but also incorporates the specific needs of the operation, thereby improving the accuracy of the fusion. The unit realizes the preliminary fusion of preoperative and intraoperative image data by introducing the key feature difference value TCY of the key feature maximum difference value TCY max Normalization is performed to ensure the stability and reliability of the fusion results. Normalization helps to eliminate the dimensional differences between different data, making the fusion results more unified and comparable. The focus weight value CZ parameter in the preliminary fusion unit for preoperative and intraoperative image data can be adjusted according to the surgical site. This flexibility makes the preliminary fusion unit for preoperative and intraoperative image data applicable to cardiac interventional surgeries of different types and complexities.

[0023] See also Figures 1 to 4 , the calculation formula of the key feature unit in the preliminary fusion optimization is as follows: RT2=(RT1+QT key×SQRT(CZ))-[(RT1×(1-QT key / QT key,max )) / 2]-[(ZS / 10)×SQRT(RT1 2 / (QT key 2 +1))]; ZS=YHT key -HT key ; in: RT2 is the optimized image data fusion feature value; QT key,max is the maximum critical characteristic value before surgery; ZS is the noise level of intraoperative image data; YHT key Expected values ​​of key intraoperative features.

[0024] In this embodiment, first, "(RT1+QT key ×SQRT(CZ))” calculation part combines the image data preliminary fusion eigenvalue RT1 with the preoperative key eigenvalue QT in the preoperative image data key Combined with the focus weight value CZ, in this way, it aims to enhance the expression of key features in the fusion result. As part of the optimization process, it ensures that the key features are more clearly reflected in the optimized intraoperative image data; "[(RT1×(1-QT key / QT key,max )) / 2]” The calculation part is calculated by multiplying a value with the preoperative key characteristic value QT key and the maximum critical characteristic value QT before surgery key,max Related factors are used to adjust the contribution of the initial fusion eigenvalue RT1 of the image data. When the preoperative key eigenvalue QT key Close to the maximum critical characteristic value QT before surgery key,max When , the value of this factor approaches 0, thereby reducing the contribution of the preliminary fusion eigenvalue RT1 of the image data. On the contrary, it increases the contribution of the preliminary fusion eigenvalue RT1 of the image data. This calculation part is used to balance the contribution of key features and other features in the optimization results to ensure the balanced expression of the overall features. “[(ZS / 10)×SQRT(RT1 2 / (QT key 2 The calculation part takes into account the intraoperative image data noise level ZS and the preoperative key eigenvalue QT keyThe square of and the inverse of the square of the preliminary fusion eigenvalue RT1 of the image data. In this way, it aims to reduce the impact of noise on the optimization results and emphasizes the importance of the preliminary fusion results. As another part of the optimization process, it ensures that the optimized intraoperative image data maintains consistency with the preliminary fusion results while reducing noise. This algorithm unit introduces the preoperative key characteristic value QT key , and the intraoperative image data noise level ZS in the intraoperative image data, so that the intraoperative key feature unit based on the preliminary fusion optimization can extract and optimize the key features during the operation. This process helps to reduce noise interference and improve the clarity and accuracy of the image data; Based on the nonlinear processing method used in the key feature units in the preliminary fusion optimization, the fusion effect of image data was further improved by introducing the nonlinear terms of quadratic terms and reciprocal terms. Nonlinear processing can better capture the complex relationships and features in the image data, thereby improving the accuracy and robustness of fusion. Optimize the key feature unit during surgery based on preliminary fusion. By comprehensively considering the preliminary fusion result image data preliminary fusion feature value RT1 and preoperative key feature value QT key factors, and generates an image data fusion feature value RT2. This process not only improves the accuracy of the fusion result, but also ensures the stability of the optimization result.

[0025] See also Figures 1 to 4 , the calculation formula for providing the preoperative image data adjustment guidance unit is as follows: QT new =[RT2+(RT1×QT) / QT max ]-[(RT2×(1-RT1 / RT1 max )) / CZ]-SQRT[(RT2 2 -RT1 2 )+TCY 2 ]; in: QT new is the adjusted feature value of preoperative image data; QT max is the maximum eigenvalue of the preoperative image data; RT1 max Initially fuse the maximum eigenvalue for the image data.

[0026] In this embodiment, the algorithm unit first "[RT2+(RT1×QT) / QT max]” The calculation part combines the optimized image data fusion feature value RT2 with the image data preliminary fusion feature value RT1 and the feature value QT for the preoperative image data. In this way, the preoperative image data of the next operation is adjusted according to the optimized intraoperative data. As part of the adjustment process, it is ensured that the preoperative image data of the next operation can reflect the changes of the optimized intraoperative data; "[(RT2×(1-RT1 / RT1 max The calculation part is calculated by dividing the optimized image data fusion eigenvalue RT2 and the image data preliminary fusion eigenvalue RT1 by the image data preliminary fusion maximum eigenvalue RT1 max The contribution of the optimized image data fusion feature value RT2 is adjusted by dividing it by the factor of the emphasis weight value CZ. In this way, it aims to fine-tune the optimized image data fusion feature value RT2 according to surgical requirements. It is used to balance the contribution of the optimized intraoperative data in the adjustment result according to surgical requirements. "SQRT[(RT2 2 -RT1 2 )+TCY 2 The calculation part calculates the square of the difference between the optimized image data fusion feature value RT2 and the image data preliminary fusion feature value RT1, as well as the square of the key feature difference value TCY. In this way, the degree of data change during the adjustment process is evaluated. As an evaluation indicator of the adjustment process, it ensures that the adjusted preoperative image data of the next surgery is consistent with the optimized intraoperative data and reflects the changes before and after the adjustment. In this algorithm unit, the preoperative image data feature value QT is adjusted by comparison new The value of the RT1 eigenvalue of the initial fusion of the image data enables the preoperative image data adjustment guidance unit to establish an effective feedback mechanism, which can evaluate the fusion quality of the preoperative and intraoperative image data and guide the subsequent image data adjustment work. The feedback mechanism helps to timely discover and correct problems in the fusion process, thereby improving the success rate and safety of the operation; Providing the adjusted preoperative image data characteristic value QT of the preoperative image data adjustment guidance unit new It can associate and cyclically influence the initial fusion unit of preoperative and intraoperative image data, and then directly influence the processing of the next round of preoperative image data. This cyclic influence and continuous optimization mechanism enables the system to continuously learn and improve, thereby improving the accuracy and efficiency of fusion; The preoperative image data adjustment guidance unit not only considers the image data fusion characteristic value RT2, but also introduces the difference value factor before and after adjustment. This enables the system to make personalized preoperative adjustments based on the patient's specific situation and surgical needs, thereby improving the pertinence and effectiveness of the surgery.

[0027] For example 2, please refer to Figures 1 to 4 , based on the adjusted preoperative image data feature value QT new The fusion analysis of the preliminary fusion feature value RT1 of the image data is as follows: If QT new =RT1, which reflects that the fusion effect of preoperative and intraoperative image data is good, and the preoperative and intraoperative image data have achieved good matching in key features; If QT new ≠RT1, it reflects that there is a deviation in the fusion process of preoperative and intraoperative image data, and the preoperative and intraoperative image data do not achieve a good match in key features. When the feature value QT of the preoperative image data after adjustment is new , adjust the preoperative image data.

[0028] In this embodiment, by providing a feedback mechanism and a cyclic influence mechanism of the preoperative image data adjustment guidance unit, the system can continuously learn and improve the fusion algorithm, which helps the system adapt to cardiac interventional surgeries of different types and complexities and improve the accuracy and efficiency of fusion. By continuously optimizing preoperative image data, providing a preoperative image data adjustment guidance unit helps to improve the success rate of surgery. More accurate image data means more precise surgical operations and lower risks of complications, thereby improving the treatment effect and quality of life of patients. Providing a cyclic influence mechanism of the preoperative image data adjustment guidance unit on the unit for realizing the preliminary fusion of preoperative and intraoperative image data provides impetus for technological innovation and development. With the continuous advancement of technology and the deepening of application, the system can be continuously upgraded and improved to provide more efficient and reliable image data support for cardiac interventional surgery. Specifically, by comparing the adjusted preoperative image data feature value QT new The value of the initial fusion feature value RT1 of the image data can intuitively understand the fusion quality of the preoperative and intraoperative image data. If the preoperative image data feature value QT new The value of the initial fusion feature value RT1 of the image data is equal, indicating that the fusion effect is good, and the preoperative and intraoperative image data have achieved a good match in key features. On the contrary, if the adjusted preoperative image data feature value QT new If the value of the initial fusion feature value RT1 of the image data is significantly different, it indicates that there are deviations and inconsistencies in the fusion process, which requires further adjustment and optimization; Adjusted preoperative image data feature value QT new The comparison with the initial fusion characteristic value RT1 of the image data not only provides a basis for the evaluation of the fusion quality, but also provides guidance for the subsequent image data adjustment work. When the fusion quality is found to be poor, the system can adjust the image data according to the adjusted preoperative image data characteristic value QT newThe difference from the initial fusion characteristic value RT1 of the image data is optimized by adjusting the parameters in the preoperative image data adjustment guidance unit to make it closer to the actual situation during the operation. This adjustment process can be iterated until a satisfactory fusion effect is achieved. In the unit for realizing the preliminary fusion of preoperative and intraoperative image data, the unit for optimizing the key intraoperative features based on the preliminary fusion, and the unit for providing guidance for adjusting preoperative image data, the hand-side focus weight value CZ and the noise level factor are taken into consideration. This enables the system to comprehensively consider the surgical needs and actual conditions when fusing preoperative and intraoperative image data, thereby obtaining a more accurate and reliable fusion result. Through the precise fusion of preoperative and intraoperative image data, we can have a clearer understanding of the heart structure and pathological conditions, so as to formulate a more accurate and safe surgical plan. During the operation, we can also adjust the surgical strategy in time according to the intraoperative image data obtained by real-time imaging technology to ensure the smooth progress of the operation. This improvement in accuracy and safety is of great significance to the success rate of cardiac interventional surgery and the prognosis of patients. In summary, the adjusted preoperative image data feature value QT new As an effective feedback mechanism, the comparison with the preliminary fusion characteristic value RT1 of image data plays an important role in evaluating the quality of preoperative and intraoperative image data fusion, guiding the subsequent image data adjustment work, and promoting the development and innovation of cardiac surgery technology. The realization of this mechanism is inseparable from the realization of the preliminary fusion unit of preoperative and intraoperative image data, the close combination and mutual correlation of the intraoperative key feature unit based on preliminary fusion optimization, and the preoperative image data adjustment guidance unit. Therefore, they together constitute the core and foundation of the preoperative and intraoperative image data model fusion system for cardiac intervention in cardiac surgery.

[0029] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A model fusion system for preoperative and intraoperative image data of cardiac surgery, characterized in that: The specific steps to achieve model fusion and adjustment are as follows: Step 1: Use the image acquisition module to collect image data before and after cardiac surgery, the image data including features of preoperative and intraoperative image data, key features, emphasis on surgical sites, and noise levels during surgery; Step 2: Using the fusion and processing module, calculate the output image data preliminary fusion feature value RT1, optimized image data fusion feature value RT2, and adjusted preoperative image data feature value QT new ; Step 3: Preliminary fusion of the eigenvalue RT1 based on the image data and the adjusted eigenvalue QT of the preoperative image data new The difference between the two images is calculated and the result output module is used to adjust the preoperative image data of the next cardiac interventional surgery of the same type. The fusion and processing module includes a unit for realizing preliminary fusion of preoperative and intraoperative image data, a unit for optimizing intraoperative key features based on preliminary fusion, and a unit for providing guidance for adjusting preoperative image data.

2. A model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 1, characterized in that: The image acquisition module includes a preoperative acquisition unit and an intraoperative acquisition unit. The preoperative acquisition unit extracts the features, key features, and emphasis of the surgical site of the preoperative image data. The intraoperative acquisition unit acquires the features, key features, and noise level of the intraoperative image data in real time and stores them.

3. A model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 2, characterized in that: The calculation formula for realizing the preliminary fusion unit of preoperative and intraoperative image data is as follows: RT1=QT×(1+(CZ / 10))-|TCY / TCY max |×QT avg ; QT=SQRT(QT x 2 +QT y 2 +QT z 2 ); TCY=|QT key -HT key |; QT avg =(QT1+QT2+QT3+......+QT n ) / n; in: Initial fusion eigenvalues ​​of RT1 image data; QT is the characteristic value of preoperative image data; QT x , QT y and QT z They are the characteristic values ​​of pixels in three dimensions (x, y, z) of the preoperative image data; CZ is the emphasis weight value, which is flexibly adjusted according to the surgical site, and its value range is {0-1}, 0 means that the surgical emphasis is not considered, and 1 means that the surgical emphasis is important; TCY is the key feature difference value, which reflects the degree of difference between preoperative and intraoperative image data in key features; QT key is the key characteristic value before operation, HT key is the key characteristic value during the operation, and the key characteristic value before the operation is QT key and intraoperative key feature value HT key The calculation formula is the same as that of the characteristic value QT of preoperative image data; TCY max is the maximum difference value of the key feature; QT avg is the average eigenvalue of preoperative image data; n is the total amount of preoperative image data; QT1 is the first image data characteristic value, QT2 is the second image data characteristic value, QT3 is the third image data characteristic value, QT n is the nth image data feature value.

4. The model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 3, characterized in that: The calculation formula of the key feature unit in the preliminary fusion optimization is as follows: RT2=(RT1+QT key ×SQRT(CZ))-[(RT1×(1-QT key / QT key,max )) / 2]-[(ZS / 10)×SQRT(RT1 2 / (QT key 2 +1))]; ZS=YHT key -HT key ; in: RT2 is the optimized image data fusion feature value; QT key,max is the maximum critical characteristic value before surgery; ZS is the noise level of intraoperative image data; YHT key Expected values ​​of key intraoperative features.

5. The model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 4, characterized in that: The calculation formula of the preoperative image data adjustment guidance unit is as follows: QT new =[RT2+(RT1×QT) / QT max ]-[(RT2×(1-RT1 / RT1 max )) / CZ]-SQRT[(RT2 2 -RT1 2 )+TCY 2 ]; in: QT new is the adjusted feature value of preoperative image data; QT max is the maximum eigenvalue of the preoperative image data; RT1 max Initially fuse the maximum eigenvalue for the image data.

6. The model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 5, characterized in that: Based on the adjusted preoperative image data characteristic value QT new The fusion analysis of the preliminary fusion feature value RT1 of the image data is as follows: If QT new =RT1, which reflects that the fusion effect of preoperative and intraoperative image data is good, and the preoperative and intraoperative image data have achieved good matching in key features; If QT new ≠RT1, it reflects that there is a deviation in the fusion process of preoperative and intraoperative image data, and the preoperative and intraoperative image data do not achieve a good match in key features. When the feature value QT of the preoperative image data after adjustment is new , adjust the preoperative image data.

7. The model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 3, characterized in that: The key characteristic value before surgery is QT key The calculation formula is as follows: QT key =SQRT(QT key，x 2 +QT key，y 2 +QT key，z 2 ); QT key，x , QT key，y and QT key，z They are the characteristic values ​​of pixels in the key areas of the preoperative image data in three dimensions (key,x,key, y, key,z); The intraoperative key characteristic value HT key The calculation formula is as follows: HT key =SQRT(HT key，x 2 +HT key，y 2 +HT key，z 2 ); HT key，x , HT key，y and HT key，z They are the feature values ​​of pixels in the key areas of the intraoperative image data in three dimensions (key, x, key, y, key, z).

8. The model fusion system for preoperative and intraoperative image data of cardiac surgery according to claim 2, characterized in that: The image acquisition module includes an imaging device and a storage device; The fusion and processing module includes a computer processing device; The result output module includes input and output devices.

Images