CN116725640A

CN116725640A - Construction method of body puncture printing template

Info

Publication number: CN116725640A
Application number: CN202310731116.1A
Authority: CN
Inventors: 姜冠群; 王宁宁
Original assignee: Shandong Zhuoye Medical Technology Co ltd
Current assignee: Shandong Zhuoye Medical Technology Co ltd
Priority date: 2023-06-20
Filing date: 2023-06-20
Publication date: 2023-09-12
Anticipated expiration: 2043-06-20
Also published as: CN116725640B

Abstract

The invention belongs to the technical field of medical clinic, and particularly relates to a construction method of a body puncture printing template. The method comprises the following steps: performing human body image scanning processing on a user by using medical image equipment to obtain DICOM image data; preprocessing the DICOM image data to generate standard image data; performing target area modeling processing on the standard image data by using a three-dimensional modeling technology to generate a target area three-dimensional model; acquiring a historical body puncture needle tract area of a medical database; and establishing a mathematical model of the optimal positioning area of the body puncture needle tract by using a convolutional neural network algorithm and the historical body puncture needle tract area so as to generate a needle tract positioning area model. According to the invention, the body puncture printing template is constructed, so that the size of the inner diameter of the needle track and the penetration mode of each intersection point in the body puncture process are more accurate.

Description

Construction method of body puncture printing template

Technical Field

The invention belongs to the technical field of medical clinic, and particularly relates to a construction method of a body puncture printing template.

Background

Often, the manual work is not overly elaborate to perform the body piercing procedure, and may cause difficulties due to a number of intricate factors during which a body piercing template is required to increase the efficiency of the procedure. However, the conventional construction method of the body piercing print template cannot accurately determine the specific position and angle of each needle track intersection point, and the calculation of the needle track inner diameter does not accurately meet the actual requirements, and the untreated template may cause poor visual effect when the template is printed.

Disclosure of Invention

Based on the above, the present invention provides a method for constructing a body piercing print template, so as to solve at least one of the above technical problems.

In order to achieve the above object, a method for constructing a body piercing print template includes the steps of:

step S1: performing human body image scanning processing on a user by using medical image equipment to obtain DICOM image data; preprocessing the DICOM image data to generate standard image data; performing target area modeling processing on the standard image data by using a three-dimensional modeling technology to generate a target area three-dimensional model;

step S2: acquiring a historical body puncture needle tract area of a medical database; establishing a mathematical model of an optimal positioning area of the body puncture needle tract by utilizing a convolutional neural network algorithm and a historical body puncture needle tract area so as to generate a needle tract positioning area model; transmitting the target area three-dimensional model to a needle track positioning area model to perform needle track positioning area prediction processing, and generating a needle track positioning area of the target area three-dimensional model; performing needle track contour extraction processing on the needle track positioning region by using a region growing algorithm to generate needle track contour data;

step S3: acquiring needle track demand starting point data; calculating the needle track radial direction vector by using curve fitting algorithm to the needle track demand starting point data and the needle track contour data to generate a needle track direction vector; performing intersection point calculation on the needle track direction vector and the target area three-dimensional model according to a ray-boundary body phase separation algorithm to generate needle track intersection point coordinate information; performing accurate positioning calculation on the needle track position by using a minimum distance matching algorithm to the needle track intersection point coordinate information to generate accurate positioning data of the needle track;

Step S4: carrying out needle track inner diameter prediction processing on the needle track contour data and the needle track accurate positioning data by utilizing a random forest algorithm to generate needle track inner diameter data; performing a bool operation optimization treatment on the needle track inner diameter data and the target area three-dimensional model to obtain optimized needle track inner diameter data;

step S5: and performing puncture matrix printing and optimizing processing on the three-dimensional model of the target area, the accurate positioning data of the needle track and the optimized needle track inner diameter data by using a hierarchical printing technology so as to obtain an optimized puncture printing template.

According to the invention, through the human body image scanning processing, the medical image equipment can provide detailed human body internal structure information, so that doctors can know the illness state and anatomical structure of patients comprehensively. The preprocessing, such as image denoising and enhancement, of the DICOM image data can improve image quality, enable doctors to observe and analyze images more clearly, improve diagnosis accuracy, and achieve consistency and comparability of the image data under different equipment and scanning parameters, so that the doctors can compare and analyze multiple groups of image data conveniently. The three-dimensional modeling technology is utilized to process the standard image data, so that a more visual and vivid visual effect is provided, doctors are helped to better understand and analyze the human body structure, and accurate positioning and reference are provided for operation planning and treatment operation. By training the convolutional neural network algorithm with the data of the historical body puncture area, a mathematical model of the body puncture optimal positioning area can be established, which helps to provide a reliable reference and helps the doctor to select the optimal puncture position and angle. The three-dimensional model of the target area is transmitted to the needle track positioning area model for prediction processing, and the needle track positioning area of the three-dimensional model of the target area can be generated, so that a doctor can determine a specific area of puncture operation, and accurate positioning information is provided. The needle track contour data can be extracted by processing the needle track positioning area by using an area growing algorithm, so that visual information of doctors is provided, and the doctors are helped to better understand the shape and the path of the needle track. The curve fitting algorithm is utilized to calculate the path direction vector of the needle track according to the needle track demand starting point data and the needle track contour data, which provides the direction information of the needle track in the three-dimensional space and is helpful for determining the path and direction of the puncture operation. The intersection point coordinate information of the needle track and the target area can be obtained by carrying out intersection calculation on the path direction vector of the needle track and the three-dimensional model of the target area through a ray-boundary body phase intersection algorithm, so that the spatial relationship between the needle track and the target area is provided, and the position of the puncture point is helped to be determined. And the minimum distance matching algorithm is utilized to perform accurate positioning calculation on the intersection point coordinate information of the needle track, so that accurate position data of the needle track is obtained, a doctor can accurately determine the position of the puncture point, and the success rate and the safety of the puncture operation are improved. The needle track contour data and the needle track accurate positioning data are processed by utilizing a random forest algorithm, so that the inner diameter of the needle track can be predicted, information about the needle track size is provided, and doctors can know the space limitation in the puncturing operation and the selection of puncturing equipment. The optimized needle track inner diameter data is obtained by performing a bool operation optimization process on the needle track inner diameter data and the three-dimensional model of the target area, so that good adaptability of the needle track and the target area is ensured, and unnecessary interference and conflict are avoided. By combining the accurate positioning data of the needle track and the optimized inner diameter data of the needle track, a puncture matrix is generated, specific guiding information is provided, a doctor is helped to determine key parameters such as puncture points, puncture angles, depth and the like, and accuracy and safety of puncture operation are improved. Therefore, the construction method of the body puncture printing template can accurately determine the specific position and angle of each needle track intersection point, and the size of the needle track inner diameter is accurate to meet the actual requirement through the bool operation, and when the template is printed, the template is processed to bring a high-quality visual effect.

Preferably, step S1 comprises the steps of:

step S11: performing human body image scanning processing on a user by using medical image equipment to obtain DICOM image data;

step S12: carrying out multi-mode image fusion on the DICOM image data by utilizing a multi-mode image fusion algorithm to generate optimized DICOM image data;

step S13: performing image noise reduction and enhancement processing on the optimized DICOM image data to generate standard image data;

step S14: carrying out three-dimensional image modeling on the standard image data by utilizing a three-dimensional modeling technology to generate a three-dimensional image model;

step S15: acquiring a puncture area to be treated;

step S16: and carrying out target area model extraction processing on the three-dimensional image model according to the puncture area to be processed, and generating a target area three-dimensional model.

According to the invention, through the human body image scanning processing, the medical image equipment can provide detailed human body internal structure information, so that doctors can know the illness state and anatomical structure of patients comprehensively. And the optimized DICOM image data is obtained by multi-mode image fusion, so that the image quality and definition are improved, and a more reliable data base is provided for subsequent processing. Noise reduction and enhancement processing are carried out on the optimized DICOM image data, so that noise interference is reduced, a target structure is highlighted, and contrast and detail information of an image are improved. The three-dimensional modeling processing is carried out on the standard image data, so that more comprehensive and three-dimensional target area information is provided, and visual analysis and positioning of doctors before operation are facilitated. The extraction of the target region model from the region of puncture to be treated provides a specific model for a specific region, enabling a physician to more accurately assess and treat the region.

Preferably, step S13 comprises the steps of:

step S131: performing noise reduction processing on the optimized DICOM image data by using Gaussian filtering to generate noise-reduced DICOM image data;

step S132: performing image correction on the noise-reduced DICOM image data by using an artifact correction algorithm to generate corrected DICOM image data;

step S133: and performing image enhancement on the corrected DICOM image data by using an image enhancement calculation formula to generate standard image data.

According to the invention, the Gaussian filter is utilized to perform noise reduction processing on the optimized DICOM image data, so that noise interference in the image can be reduced, the definition and quality of the image are improved, and the noise-reduced data are more beneficial to doctors to accurately observe and analyze the target area. The method eliminates the artifacts and bad characteristics in the images, so that the images reflect the human tissue structure more truly and accurately, and the corrected DICOM image data is helpful for doctors to diagnose and evaluate the illness state of patients more accurately. The image enhancement calculation formula is utilized to carry out image enhancement on the corrected DICOM image data, so that the details and contrast of a target area can be highlighted, and the enhanced standard image data enables doctors to observe important information such as lesions, anatomical structures and the like more clearly, thereby being beneficial to improving diagnosis accuracy.

Preferably, the image enhancement calculation formula in step S133 is as follows:

where G (x, y) is represented as a gradation value after image enhancement, x is represented as an abscissa of an image, y is represented as an ordinate of an image, q is represented as an amplitude controlling an output gradation level, w is represented as a degree controlling a nonlinear gain adjustment effect, F (x, y) is represented as a gradation value of an unprocessed original image, r is represented as an amplitude controlling a nonlinear variation, u is represented as a directionality of an output signal, p is represented as a linear range threshold value of pixel brightness, o is represented as weight information of directionality of an output signal, and τ is represented as an abnormal adjustment value of the gradation value after image enhancement.

The invention utilizes an image enhancement calculation formula which fully considers the interaction relationship among the abscissa x of an image, the ordinate y of the image, the amplitude q of the control output gray level, the degree w of the control nonlinear gain adjustment effect, the gray value F (x, y) of an unprocessed original image, the amplitude r of the control nonlinear variation, the directivity u of an output signal, the linear range threshold p of pixel brightness, the weight information o of the directivity of the output signal and a function to form a functional relation q By q [1+wln (F (x, y))] ^r The degree of the nonlinear gain adjustment effect is controlled, the amplitude of the output gray level is controlled by adjusting parameters, the degree of the nonlinear gain adjustment effect is controlled, the amplitude of the nonlinear change is controlled, different degrees of gains can be carried out on different gray levels, the details in the image are more prominent, the nonlinear gain adjustment is beneficial to enhancing the contrast of the image, a doctor can observe fine structural changes more clearly, and the diagnosis accuracy is improved. By->The directionality of the output signals is controlled, the directionality and shape of the output signals can be controlled by adjusting the directionality of the parameter output signals, the linear range threshold of the pixel brightness and the weight information of the directionality of the output signals so as to adapt to the characteristics of different target areas, and the directionality adjustment of the output signals can highlight the details in the specific direction, thereby being beneficial to a doctor to better analyze the structural positioning and morphological characteristics in the image. By nonlinear gain adjustment, directivity adjustment of output signals and abnormal adjustment of gray values, contrast of images can be enhanced, details are highlighted, gray levels are balanced, and accordingly observation and analysis capabilities of the images are improved, accuracy and reliability of diagnosis are improved, and more accurate diagnosis and treatment decision making are supported. The function relation is adjusted and corrected by utilizing the abnormal adjustment value tau of the gray value after image enhancement, so that the error influence caused by abnormal data or error items is reduced, the gray value G (x, y) after image enhancement is generated more accurately, and the accuracy and the reliability of image enhancement processing on the corrected DICOM image data are improved. Meanwhile, the weight information and the adjustment value in the formula can be adjusted according to actual conditions and are applied to different image data, so that the flexibility and applicability of the algorithm are improved.

Preferably, step S2 comprises the steps of:

step S21: acquiring a historical body puncture needle tract area of a medical database;

step S22: establishing a mapping relation of an optimal positioning area of the body puncture needle tract by using a convolutional neural network algorithm so as to generate an initial needle tract positioning area model;

step S23: performing model training on the initial needle track positioning area model by utilizing the historical body puncture needle track area to generate a needle track positioning area model;

step S24: transmitting the target area three-dimensional model to a needle track positioning area model to perform needle track positioning area prediction processing, and generating a needle track positioning area of the target area three-dimensional model;

step S25: and performing needle track contour extraction processing on the needle track positioning region by using a region growing algorithm to generate needle track contour data.

According to the invention, by acquiring the historical body puncture needle tract area of the medical database, the puncture needle tract of the three-dimensional image model of the target area can be known to move in a certain area, and the three-dimensional image of the target area of the area is targeted by using what needle tract to provide reference and basis for the subsequent needle tract positioning area; by using a convolutional neural network algorithm, the mapping relationship of the optimal positioning region of the body puncture needle tract can be learned and established, so that an initial needle tract positioning region model is generated. By using the data of the historical body puncture needle tract area for model training, the initial needle tract positioning area model can be further optimized and adjusted to be more accurate and reliable. The target area three-dimensional model is transmitted to the needle track positioning area model, so that the prediction processing of the needle track positioning area can be performed, the needle track positioning area of the target area three-dimensional model is determined, and accurate positioning information is provided for subsequent needle track operation. By applying the region growing algorithm, the needle track contour data can be generated based on the needle track positioning region, the shape and contour information of the needle track can be extracted, the length information of the needle track can be determined, and detailed guidance and visual information can be provided for the subsequent puncture operation aiming at what needle track type should be used for the three-dimensional image of the target region.

Preferably, step S3 comprises the steps of:

step S31: acquiring needle track demand starting point data;

step S32: performing curve path fitting processing on the needle track demand starting point data by using a curve fitting algorithm to generate a simulated needle track curve path;

step S33: performing curve path optimization processing on the simulated needle track curve path by using the needle track contour data to generate an optimal simulated needle track curve path;

step S34: calculating tangential direction vectors of the optimal simulated needle track curve path by using a center difference method to generate needle track direction vectors;

step S35: performing intersection point calculation on the needle track direction vector and the target area three-dimensional model according to a ray-boundary body phase separation algorithm to generate needle track intersection point coordinate information;

step S36: and carrying out accurate positioning calculation on the needle track position by utilizing a minimum distance matching algorithm to the needle track intersection point coordinate information, and generating accurate positioning data of the needle track.

According to the method, the starting point data of the needle track demand is obtained, so that the starting point of the puncture operation can be determined, and basic data is provided for subsequent needle track calculation. By means of a curve fitting algorithm, the needle track demand starting point data can be fitted into a smooth curve path, and the expected track of the needle track is simulated. By optimizing the simulated needle track curve path and the actual needle track contour data, the curve path can be adjusted and corrected so as to be more in line with the real form of the needle track and the structural characteristics of the target area. By the center difference method, tangential direction vectors of each point on the curve path of the optimal simulated needle track can be calculated, and the direction and trend of the needle track at the point are indicated. And calculating an intersection point between the needle track direction vector and the three-dimensional model of the target area by using a ray-boundary body phase correlation algorithm to obtain intersection point coordinate information of the needle track and the target area. Through a minimum distance matching algorithm, the coordinate of the intersection point of the needle track can be accurately matched with the target area, the accurate position of the needle track in the target area is calculated, and accurate puncture positioning information is provided.

Preferably, step S33 includes the steps of:

step S331: performing curve path parameter optimization calculation on the needle track contour data by using a curve path parameter optimization calculation formula to generate curve path optimization parameters;

step S332: and carrying out parameter optimization on the simulated needle track curve path according to the curve path optimization parameters to generate an optimal simulated needle track curve path.

According to the invention, by applying the curve path parameter optimization calculation formula, the curve path optimization parameters which can better describe the needle track shape and characteristics can be obtained by carrying out parameter optimization on the needle track contour data. By utilizing the curve path optimization parameters, the simulated needle track curve path can be adjusted and optimized to be more matched with the actual needle track contour data, the optimal simulated needle track curve path is generated, the needle track contour data and the simulated needle track curve path can be optimized, and the simulated needle track path can more accurately reflect the actual needle track shape and characteristics.

Preferably, the curve path parameter optimization calculation formula in step S331 is as follows:

wherein P is expressed as a curve path optimization parameter, N is expressed as the number of path points of the data of the demand start point of the needle track, a is expressed as a position in the horizontal direction for adjusting the curve path, b is expressed as a curve degree affecting the curve, θ _i Represented as azimuth angle, k of current point in polar coordinate system _i Represented as the distance of the current point in the polar coordinate system,deformation data expressed as track profile data, ω expressed as an abnormal adjustment value of the curve path optimization parameter.

The invention utilizes a curve path parameter optimization calculation formula which fully considers the number N of the path points of the needle track demand starting point data, the position a of the curve path in the horizontal direction, the bending degree b of the curve, the azimuth angle theta of the current point in the polar coordinate system _i Distance k of current point in polar coordinate system _i Deformation data of needle track profile dataAnd the interaction relationship between the functions to form a functional relationship +.>The position of the curve path in the horizontal direction is adjusted to enable the position of the curve path relative to the starting point in the horizontal direction to deviate, so that certain anatomical structures are avoided or the puncture path is adjusted according to individual differences of patients; the curve-influencing bending degree parameter is considered in different surgical operations or lesionsIn order to adapt to different surgical operations or pathological changes; the azimuth angle in the polar coordinate system and the distance in the polar coordinate system describe the shape of the curve path in the polar coordinate system, so that the optimal simulated needle path curve path meeting the actual requirements can be obtained by only analyzing the azimuth angle and the distance of the needle path point number on the polar coordinate, the result is not influenced while the data calculation amount is reduced, and the calculation force is reduced; and the influence of different needle track lengths, materials and other deformation information on the curve path is considered through the deformation data of the needle track profile data, so that the curve can be smoother or more curved in the areas so as to adapt to specific puncture requirements. The puncture path can be accurately controlled and optimized, the puncture accuracy is improved, the risk is reduced, the success rate of the operation is increased, and better operation guidance and decision basis are provided. And the function relationship is adjusted and corrected by using the abnormal adjustment value omega of the curve path optimization parameter, so that the error influence caused by abnormal data or error items is reduced, the curve path optimization parameter P is generated more accurately, and the accuracy and the reliability of curve path parameter optimization calculation processing on the needle track profile data are improved. Meanwhile, the adjustment value in the formula can be adjusted according to actual conditions and is applied to body puncture templates of different parts, so that the flexibility and applicability of the algorithm are improved.

Preferably, step S4 comprises the steps of:

step S41: carrying out mathematical model parameter marking processing on the needle track contour data and the needle track accurate positioning data to mark as model prediction parameters;

step S42: performing feature extraction processing on the model prediction parameters by using a principal component analysis method to generate model prediction feature parameters;

step S43: establishing a mapping relation between puncture conditions and the needle track inner diameter by using a random forest algorithm to generate an initial needle track inner diameter prediction model;

step S44: performing model training treatment on the initial needle track inner diameter prediction model by using model prediction parameters to generate a needle track inner diameter prediction model;

step S45: performing needle track inner diameter prediction processing on the model prediction characteristic parameters by using a needle track inner diameter prediction model to generate needle track inner diameter data;

step S46: performing a bool operation on the needle track inner diameter data and the target area three-dimensional model to obtain intersection data of the needle track inner diameter data and the target area three-dimensional model;

step S47: the inner diameter of the needle track is adjusted according to the intersection data, and the needle track inner diameter adjusting data is generated;

step S48: repeating steps S46 and S47 when the intersection data is greater than a preset needle track inner diameter optimization threshold; and when the intersection data is not more than a preset needle track inner diameter optimization threshold value, generating optimized needle track inner diameter data.

The invention obtains model prediction parameters by carrying out mathematical model parameter marking processing on the needle track contour data and the needle track precise positioning data, wherein the marking parameters contain important information about the needle track shape, positioning and other characteristics. The feature extraction process is performed on model prediction parameters using principal component analysis, and the most representative and discriminative feature parameters are extracted, which helps to reduce the dimensionality of the data, capture key features, and provide a more efficient representation of the data. The mapping relation between the puncture condition and the needle track inner diameter is established by using a random forest algorithm, and an initial needle track inner diameter prediction model is generated, and the model can predict the proper inner diameter range of the needle track according to the puncture condition and other parameters. The initial needle track inner diameter prediction model is trained by using model prediction parameters, so that the needle track inner diameter prediction model is generated, the model can accurately predict the inner diameter data of the needle track according to the model prediction parameters, and a proper puncture needle diameter selection reference is provided for the operation. The needle track inner diameter prediction model is utilized to conduct needle track inner diameter prediction processing on the model prediction characteristic parameters, the inner diameter data of the needle track is obtained, the prediction process is based on model learning and training, and the inner diameter of the needle track can be estimated rapidly and accurately under the condition of lacking practical measurement. After the needle track inner diameter data is obtained, the needle track inner diameter data and the target area three-dimensional model are subjected to Boolean operation, so that intersection data of the needle track inner diameter data and the target area three-dimensional model are obtained, whether the needle track intersects with the target area or not is facilitated to be determined, and reference is provided for subsequent inner diameter adjustment. The inner diameter of the needle track is adjusted according to the intersection data, and adjusted needle track inner diameter data are generated. Finally, when the intersection data is greater than the preset needle track inner diameter optimization threshold, the steps S46 and S47 are repeatedly executed to further optimize the needle track inner diameter data, so that the interaction between the needle track and the target area can be ensured to be minimized, and the accuracy and safety of the operation are improved.

Preferably, step S5 comprises the steps of:

step S51: performing plane printing processing on the three-dimensional model of the target area, the accurate positioning data of the needle track and the optimized inner diameter data of the needle track by using a hierarchical printing technology to generate a puncture printing template;

step S52: performing effective part matrix extraction processing on the puncture printing template to generate an effective puncture printing template;

step S53: and carrying out smoothing treatment on the effective puncture printing template by using a smoothing algorithm to generate an optimized puncture printing template.

The invention performs plane printing processing on the three-dimensional model of the target area, the accurate positioning data of the needle track and the optimized inner diameter data of the needle track by using a hierarchical printing technology. This step converts the three-dimensional data into a flat-panel print template that can be printed out layer by a 3D printer. And (3) carrying out effective part matrix extraction processing on the puncture printing template to generate an effective puncture printing template, so that effective parts related to puncture in the template are extracted, unnecessary parts are removed, the structure of the printing template is simplified, and the printing time and material consumption are reduced. And smoothing the effective puncture printing template by using a smoothing algorithm to generate an optimized puncture printing template, and smoothing the surface of the printing template to be smoother and smoother, so that the discontinuity and the surface roughness in the printing process are reduced, and the printing quality and the accuracy of puncture operation are improved.

The method has the advantages that the quality and the definition of medical images can be improved, the interference of noise and artifacts can be reduced, and medical workers can know the anatomical structure and pathological changes of patients more accurately through the image processing steps. This helps the healthcare worker in making treatment regimens and making surgical decisions more accurately and reliably. Through the steps of model construction and needle track positioning, the method provides an auxiliary tool for positioning for puncture operation. By establishing a needle track positioning region model and extracting needle track profile data, a medical worker can accurately position a puncture target and determine an optimal puncture path. This helps to improve puncture accuracy and success rate, and reduce trauma and injury to the patient. The needle track inner diameter prediction step enables medical workers to predict the inner diameter of the puncture needle track, so that proper puncture tools and materials are selected, and the successful operation of the puncture process is ensured. This helps to reduce pain and discomfort during lancing, and to improve patient comfort and experience. By generating the puncture printing template, a practical visual tool is provided for medical workers. The puncture printing template presents the anatomical structure and the puncture path of the target area in the form of a three-dimensional model, so that a medical worker can more intuitively understand the key points and key steps of the puncture operation. This helps to improve the skill of the healthcare worker, reduce the risk of handling, and also increases the communication and collaboration effort between the patient and the medical team. Through the comprehensive application of the steps of image processing, model construction, needle path planning, needle path inner diameter prediction and the like, more accurate, reliable and safe guidance is provided for puncture operation, the improvement of operation effect is facilitated, complications and operation risks are reduced, and better medical services are provided for patients.

Drawings

FIG. 1 is a schematic flow chart of steps of a method for automatically extracting and analyzing data information of a medical examination sheet according to the present invention;

FIG. 2 is a detailed flowchart illustrating the implementation of step S1 in FIG. 1;

FIG. 3 is a detailed flowchart illustrating the implementation of step S2 in FIG. 1;

FIG. 4 is a flowchart illustrating the detailed implementation of step S3 in FIG. 1;

FIG. 5 is a flowchart illustrating the detailed implementation of step S4 in FIG. 1;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

The embodiment of the application provides a method for constructing a body puncture printing template, and medical imaging equipment comprises the following steps: at least one of computed tomography, magnetic resonance imaging, and the like. The three-dimensional model of the target region includes, but is not limited to: at least one of an arm region three-dimensional model, a leg region three-dimensional model, and the like.

To achieve the above object, referring to fig. 1 to 5, a method for constructing a body piercing print template includes the steps of:

In the embodiment of the present invention, as described with reference to fig. 1, a schematic flow chart of steps of a method for constructing a body piercing print template according to the present invention is provided, and in this example, the method for constructing a body piercing print template includes the following steps:

in embodiments of the present invention, the patient is scanned using computed tomography techniques, which will generate a series of DICOM image data. Preprocessing the acquired DICOM image data may include removing noise, adjusting contrast and brightness of the image, and performing image normalization, for example, an image filtering algorithm may be applied to remove noise and enhance the contrast of the image using a histogram equalization method. After the standardized image data is obtained, the standard image data may be processed using a three-dimensional modeling technique to generate a three-dimensional model of the target region, for example, a head, which may be reconstructed from the continuous tomographic images using a volume rendering technique. This process involves techniques such as voxel reconstruction, volume rendering, and volume rendering of the image to generate a three-dimensional head model having a spatial structure.

in the embodiment of the invention, the historical body puncture needle tract area data in the medical database needs to be acquired, for example, the puncture record of the previous patient can be extracted from the electronic medical record system of the hospital, and the puncture record comprises known puncture needle tract positions and related information. A mathematical model is established by utilizing a Convolutional Neural Network (CNN) algorithm to generate a needle track positioning area model, and in the training process, known puncture needle track data can be used as a training sample and input into the CNN model for learning and feature extraction, so that the mathematical model of the needle track positioning area is established. Transmitting the three-dimensional model of the target area to a needle track positioning area model to conduct prediction processing of the needle track positioning area, generating the needle track positioning area of the three-dimensional model of the target area, inputting the three-dimensional model of the target area to a trained CNN model, and obtaining the needle track positioning area by utilizing the prediction capability of the CNN model. The needle track positioning area is processed by using an area growing algorithm, the needle track contour data is extracted, and the area growing algorithm can gradually expand and identify pixel points belonging to the needle track according to the predicted needle track positioning area, so that the needle track contour data is obtained.

in the embodiment of the invention, the data of the needle track demand starting point is obtained. This may be achieved by the doctor or operator marking or entering the desired start and end positions of the needle track. And processing the needle track demand starting point data and the needle track contour data by using a curve fitting algorithm, calculating the direction vector of the needle track path, and fitting the starting point and the contour data point into a smooth curve by using a least square method or spline curve fitting technology to obtain the direction vector of the needle track. And according to a ray-boundary body phase intersection algorithm, performing intersection point calculation on the needle track direction vector and the three-dimensional model of the target area to determine an intersection point of the needle track and the target area, and performing intersection judgment on the three-dimensional model of the target area by tracking rays emitted from the needle track starting point along the needle track direction vector and obtaining intersection point coordinate information of the needle track. And processing the coordinate information of the needle track intersection point by using a minimum distance matching algorithm, and performing accurate positioning calculation on the needle track position, for example, the distance between the needle track intersection point and the nearest point on the surface of the target area can be calculated, and the point with the minimum distance is selected as the final position of the needle track, so that the accurate positioning data of the needle track can be obtained.

in the embodiment of the invention, the needle track contour data and the needle track accurate positioning data are processed by utilizing a random forest algorithm to predict the needle track inner diameter, for example, the needle track contour data and the accurate positioning data are used as input features, a random forest model is trained for predicting the needle track inner diameter, and the model can predict the inner diameter of a new needle track by learning the relation between the needle track contour and the positioning information in the historical data and the inner diameter. And (3) carrying out Boolean operation optimization processing on the needle track inner diameter data and the three-dimensional model of the target area to obtain optimized needle track inner diameter data, for example, carrying out mutual operation on the needle track inner diameter data and the three-dimensional model of the target area by using Boolean operation, so that the needle track inner diameter is ensured to be consistent with the structure of the target area, the non-intersecting part of the needle track inner diameter data and the target area is removed, and more accurate needle track inner diameter data is obtained.

In the embodiment of the invention, a hierarchical printing technology is used, a three-dimensional model of a target area, accurate positioning data of a needle track and a three-dimensional model of the target area after puncture generated by optimizing the inner diameter data of the needle track are printed layer by layer according to layer height, positioning is carried out on each layer according to the accurate positioning data of the needle track, and needle track puncture simulation is printed on a corresponding matrix position, so that an optimized puncture printing template is obtained.

Preferably, step S1 comprises the steps of:

Step S15: acquiring a puncture area to be treated;

As an example of the present invention, referring to fig. 2, a detailed implementation step flow diagram of step S1 in fig. 1 is shown, where step S1 includes:

in embodiments of the present invention, the patient is scanned using computed tomography techniques, which will generate a series of DICOM image data.

in the embodiment of the invention, the multi-mode image fusion algorithm is utilized to carry out multi-mode image fusion on the DICOM image data to generate optimized DICOM image data, for example, the DICOM data (such as X-rays, CT and MRI) from different image modes are fused to extract multi-mode information, so that more comprehensive and accurate image data is obtained.

in the embodiment of the invention, the optimized DICOM image data is subjected to image noise reduction and enhancement processing to generate standard image data, for example, an image processing algorithm is applied to perform noise reduction processing on the DICOM image, noise interference is removed, and image enhancement is performed to improve the contrast, definition and detail visibility of the image.

in the embodiment of the invention, the standard image data is subjected to three-dimensional image modeling by utilizing a three-dimensional modeling technology to generate a three-dimensional image model, for example, the three-dimensional image model is based on the standard image data, and the three-dimensional reconstruction algorithm and the three-dimensional modeling software are used for converting the image data into the three-dimensional model to present the three-dimensional shape of the anatomical structure and the organ in the human body.

Step S15: acquiring a puncture area to be treated;

in the embodiment of the invention, the puncture area to be processed is acquired, and the specific area needing to be punctured is determined according to specific medical requirements or operation plans, for example, the specific tumor position or anatomical structure is determined.

In the embodiment of the invention, the target region model extraction processing is performed on the three-dimensional image model according to the puncture region to be processed, so as to generate a target region three-dimensional model, for example, the target region corresponding to the puncture region to be processed is extracted from the three-dimensional image model by using an image segmentation algorithm or an interactive labeling tool, and an independent three-dimensional model containing the target region is generated.

Preferably, step S13 comprises the steps of:

In the embodiment of the invention, the optimized DICOM image data is subjected to noise reduction processing by using Gaussian filtering to generate the noise-reduced DICOM image data, for example, the DICOM image is subjected to smoothing processing by using a Gaussian filtering algorithm to remove high-frequency noise, so that granular or speckle noise in an image is reduced, and the noise-reduced DICOM image data is obtained. The noise-reduced DICOM image data is image-corrected using an artifact correction algorithm to generate corrected DICOM image data, for example, according to artifacts or artifacts that may be present in the image, the image is adjusted using the correction algorithm to eliminate the effects of the artifacts, thereby obtaining more accurate and true DICOM image data. And performing image enhancement on the corrected DICOM image data by using an image enhancement calculation formula to generate standard image data, for example, applying an image enhancement algorithm to adjust the gray value details of the image, thereby improving the visual effect of the image and obtaining the standard image data which is clearer and has good visual perception.

The invention utilizes an image enhancement calculation formula which fully considers the interaction relationship among the abscissa x of an image, the ordinate y of the image, the amplitude q of the control output gray level, the degree w of the control nonlinear gain adjustment effect, the gray value F (x, y) of an unprocessed original image, the amplitude r of the control nonlinear variation, the directivity u of an output signal, the linear range threshold p of pixel brightness, the weight information o of the directivity of the output signal and a function to form a functional relation qBy q [1+wln (F (x, y))] ^r The degree of the nonlinear gain adjustment effect is controlled, the amplitude of the output gray level is controlled by adjusting parameters, the degree of the nonlinear gain adjustment effect is controlled, the amplitude of the nonlinear change is controlled, different degrees of gains can be carried out on different gray levels, the details in the image are more prominent, the nonlinear gain adjustment is beneficial to enhancing the contrast of the image, a doctor can observe fine structural changes more clearly, and the diagnosis accuracy is improved. By->The directionality of the output signals is controlled, the directionality and shape of the output signals can be controlled by adjusting the directionality of the parameter output signals, the linear range threshold of the pixel brightness and the weight information of the directionality of the output signals so as to adapt to the characteristics of different target areas, and the directionality adjustment of the output signals can highlight the details in the specific direction, thereby being beneficial to a doctor to better analyze the structural positioning and morphological characteristics in the image. By nonlinear gain adjustment, transmission The directionality adjustment of the output signals and the abnormal adjustment of the gray values can enhance the contrast of the image, highlight details and balance gray levels, so that the observation and analysis capability of the image are improved, the accuracy and the reliability of diagnosis are improved, and more accurate diagnosis and treatment decision making are supported. The function relation is adjusted and corrected by utilizing the abnormal adjustment value tau of the gray value after image enhancement, so that the error influence caused by abnormal data or error items is reduced, the gray value G (x, y) after image enhancement is generated more accurately, and the accuracy and the reliability of image enhancement processing on the corrected DICOM image data are improved. Meanwhile, the weight information and the adjustment value in the formula can be adjusted according to actual conditions and are applied to different image data, so that the flexibility and applicability of the algorithm are improved.

Preferably, step S2 comprises the steps of:

As an example of the present invention, referring to fig. 3, a detailed implementation step flow diagram of step S2 in fig. 1 is shown, where step S2 includes:

in the embodiment of the invention, the historical body puncture needle tract area of the medical database is obtained, for example, the stored past puncture operation record or related image data, including the position, the shape and other information of the puncture needle tract, is obtained from the medical database.

in the embodiment of the invention, a convolutional neural network algorithm is used as an initial model for predicting the optimal positioning area of the body puncture needle tract in the later period so as to generate an initial needle tract positioning area model.

in the embodiment of the invention, the historical body puncture needle tract area is utilized to carry out model training on the initial needle tract positioning area model, so as to generate the needle tract positioning area model, for example, the historical puncture needle tract area is used as a training data set to be transmitted to the initial needle tract positioning area model for training, so that the model parameters of the initial needle tract positioning area model are optimized, and a more accurate needle tract positioning area model is generated.

in the embodiment of the invention, the three-dimensional model of the target area is transmitted to the needle track positioning area model to conduct needle track positioning area prediction processing, the needle track positioning area of the three-dimensional model of the target area is generated, for example, the three-dimensional model of the target area is input into the needle track positioning area model, the target area is predicted by using the model, and the needle track positioning area is obtained, namely, the needle track position in the three-dimensional model of the target area is predicted.

In the embodiment of the invention, a region growing algorithm is applied, and the boundary contour information of the needle track is automatically extracted according to a certain pixel similarity criterion based on the predicted needle track positioning region, so that the needle track contour data is obtained.

Preferably, step S3 comprises the steps of:

step S31: acquiring needle track demand starting point data;

As an example of the present invention, referring to fig. 4, a detailed implementation step flow diagram of step S3 in fig. 1 is shown, where step S3 includes:

step S31: acquiring needle track demand starting point data;

in the embodiment of the invention, the data of the needle track demand starting point is obtained. This may be achieved by the doctor or operator marking or entering the desired start and end positions of the needle track.

in the embodiment of the invention, curve-fitting processing is performed on the needle track demand starting point data by using a curve-fitting algorithm to generate a simulated needle track curve path, for example, a least square method or a Bezier curve-fitting algorithm is used to fit the needle track starting point data to obtain a simulated curve path.

in the embodiment of the invention, the curve path optimization processing is carried out on the simulated needle track curve path by using the needle track contour data to generate the optimal simulated needle track curve path, for example, the simulated curve path and the needle track contour data are subjected to cross verification, and the curve is adjusted and optimized so as to be more in line with the shape and constraint conditions of the needle track, thereby obtaining the optimal simulated needle track curve path.

in the embodiment of the invention, tangential direction vector calculation is performed on the optimal simulated needle track curve path by using a central difference method, so as to generate a needle track direction vector, for example, the tangential direction vector at each point on the curve is obtained by performing central difference calculation on the optimal simulated needle track curve path, and is used for representing the advancing direction of the needle track.

in the embodiment of the invention, the intersection point calculation is carried out on the needle track direction vector and the three-dimensional model of the target area according to the ray-boundary body phase separation algorithm, the needle track intersection point coordinate information is generated, for example, rays are sent out from the needle track starting point along the direction vector by utilizing the ray projection algorithm, the intersection calculation is carried out on the needle track starting point along the direction vector and the three-dimensional model of the target area, and the intersection point coordinate of the needle track and the target area is found.

In the embodiment of the invention, the needle track intersection point coordinate information is subjected to needle track position accurate positioning calculation by utilizing the minimum distance matching algorithm, the needle track accurate positioning data is generated, for example, the needle track intersection point coordinate is matched with the surface of the target area by utilizing the minimum distance matching algorithm, and the surface point closest to the intersection point is found, so that the accurate positioning of the needle track position is realized.

Preferably, step S33 includes the steps of:

In the embodiment of the invention, the curve path parameter optimization calculation formula is utilized to perform curve path parameter optimization calculation on the needle track profile data to generate curve path optimization parameters, for example, a gradient descent method can be used to minimize the gap between the curve path and the needle track profile data. And gradually adjusting the parameters by calculating the partial derivative of each curve path parameter and updating according to the optimization target until the effect of minimizing the gap is achieved. In this way, curve path optimization parameters suitable for the needle track profile data can be obtained. And performing parameter optimization on the simulated needle track curve path according to the curve path optimization parameters to generate an optimal simulated needle track curve path, for example, adjusting the simulated needle track curve path according to the calculated curve path optimization parameters. For example, the position, curvature or other related attributes of the control point of the curve path can be adjusted according to the limiting and optimizing targets of the value range of the parameter, so that the simulated needle track curve path meets the actual requirements better.

The invention utilizes a curve path parameter optimization calculation formula which fully considers the number N of the path points of the needle track demand starting point data, the position a of the curve path in the horizontal direction, the bending degree b of the curve, the azimuth angle theta of the current point in the polar coordinate system _i Distance k of current point in polar coordinate system _i Deformation data of needle track profile dataAnd the interaction relationship between the functions to form a functional relationship +.>The position of the curve path in the horizontal direction is adjusted to enable the position of the curve path relative to the starting point in the horizontal direction to deviate, so that certain anatomical structures are avoided or the puncture path is adjusted according to individual differences of patients; the bending degree parameters affecting the curve are considered under different operation or pathological conditions so as to adapt to the different operation or pathological conditions; orientation in polar coordinate systemThe distance between the angle and the polar coordinate system describes the shape of the curve path in the polar coordinate system, so that the optimal simulated needle track curve path meeting the actual requirements can be obtained by only analyzing the azimuth angle and the distance of the needle track path point number on the polar coordinate, the result is not influenced while the data calculation amount is reduced, and the calculation force is reduced; and the influence of different needle track lengths, materials and other deformation information on the curve path is considered through the deformation data of the needle track profile data, so that the curve can be smoother or more curved in the areas so as to adapt to specific puncture requirements. The puncture path can be accurately controlled and optimized, the puncture accuracy is improved, the risk is reduced, the success rate of the operation is increased, and better operation guidance and decision basis are provided. And the function relationship is adjusted and corrected by using the abnormal adjustment value omega of the curve path optimization parameter, so that the error influence caused by abnormal data or error items is reduced, the curve path optimization parameter P is generated more accurately, and the accuracy and the reliability of curve path parameter optimization calculation processing on the needle track profile data are improved. Meanwhile, the adjustment value in the formula can be adjusted according to actual conditions and is applied to body puncture templates of different parts, so that the flexibility and applicability of the algorithm are improved.

Preferably, step S4 comprises the steps of:

As an example of the present invention, referring to fig. 5, a detailed implementation step flow diagram of step S4 in fig. 1 is shown, where step S4 includes:

in the embodiment of the invention, the characteristic point positions of the needle track contour data and the related information of the needle track accurate positioning data are marked and used as input parameters of subsequent model prediction to obtain model prediction parameters.

in the embodiment of the invention, the marked model prediction parameters are subjected to dimension reduction processing through a principal component analysis algorithm, the most representative characteristic parameters are extracted, the data dimension is reduced, important information is reserved, and the model prediction characteristic parameters are generated.

in the embodiment of the invention, a random forest algorithm is used for modeling the relation between the puncture condition and the needle track inner diameter, and an initial needle track inner diameter prediction model is trained.

in the embodiment of the invention, the marked model prediction parameters are used as input data to train an initial needle track inner diameter prediction model, and the prediction capacity and accuracy of the model are optimized to generate the needle track inner diameter prediction model.

in the embodiment of the invention, the extracted model prediction characteristic parameters are input into a needle track inner diameter prediction model to perform needle track inner diameter prediction processing, so as to obtain a prediction result of the needle track inner diameter.

in the embodiment of the invention, the needle track inner diameter data is assumed to be expressed as a geometric body in a three-dimensional space, and the three-dimensional model of the target area is also expressed as another geometric body, and the overlapped part of the needle track inner diameter data and the target area, namely intersection data of the needle track inner diameter data and the target area, can be obtained by carrying out Boolean operation on the three-dimensional model and the geometric body.

in the embodiment of the invention, the needle track inner diameter data is adjusted according to the geometric information of the intersection data, for example, if the intersection data indicates that the needle track inner diameter is partially overlapped with the target area, the size of the needle track inner diameter may need to be reduced to ensure that the needle track inner diameter meets the requirement of the target area. Conversely, if the intersection data indicates that the needle track inner diameter does not overlap the target region, it may be necessary to enlarge the size of the needle track inner diameter to ensure the accuracy of the lancing operation.

In the embodiment of the invention, whether the steps S46 and S47 are required to be repeatedly executed is judged according to the preset needle track inner diameter optimization threshold, if the intersection data is larger than the preset needle track inner diameter optimization threshold, the adjusted needle track inner diameter still needs to be further optimized, at the moment, the adjusted needle track inner diameter data can be used as new needle track inner diameter data, boolean operation and inner diameter adjustment can be carried out again until the intersection data is not larger than the preset threshold any more, and once the intersection data is not larger than the preset needle track inner diameter optimization threshold, the final optimized needle track inner diameter data can be generated.

Preferably, step S5 comprises the steps of:

In the embodiment of the application, layering processing is carried out on the three-dimensional model of the target area, the accurate positioning data of the needle track and the optimized inner diameter data of the needle track, each layer is printed step by using a printing technology, and finally the puncture printing template is obtained. And (3) performing image processing on the puncture printing template, extracting effective parts in the puncture printing template, and removing irrelevant background and noise to obtain the effective puncture printing template only containing the information required by puncture. And (3) processing the effective puncturing printing template by using a smoothing algorithm to remove unnecessary noise and details, so that the surface of the printing template is smoother and more continuous, and the accuracy and stability of puncturing operation are improved.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the 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

1. A method of constructing a body piercing print template, comprising the steps of:

2. The method of constructing a body piercing print template according to claim 1, wherein step S1 comprises the steps of:

step S15: acquiring a puncture area to be treated;

3. The method of constructing a body piercing print template according to claim 2, wherein step S13 comprises the steps of:

4. The method of constructing a body piercing print template according to claim 3, wherein the image enhancement calculation formula in step S133 is as follows:

5. A method of constructing a body piercing print template according to claim 3, wherein step S2 comprises the steps of:

6. The method of constructing a body piercing print template according to claim 5, wherein step S3 comprises the steps of:

step S31: acquiring needle track demand starting point data;

7. The method of constructing a body piercing print template as claimed in claim 6, wherein step S33 comprises the steps of:

8. The method of constructing a body piercing print template according to claim 7, wherein the curve path parameter optimization calculation formula in step S331 is as follows:

9. The method of constructing a body piercing print template according to claim 7, wherein step S4 comprises the steps of:

10. The method of constructing a body piercing print template according to claim 9, wherein step S5 comprises the steps of: