Intelligent control system of three-dimensional platform for THz spectrum detection of tumor tissue slices

Abstract

The present invention provides a three-dimensional platform intelligent control system suitable for THz spectrum detection of tumor tissue slices, wherein the correction coefficient acquisition module is used to perform detection scanning according to preset tissue slices, and determine the correction coefficient according to the result of the detection scanning; the positioning scanning module is used to perform positioning scanning on the slice tissue slide to obtain the initial position information of the slice tissue slide; the point-by-point scanning module is used to control the working state of the motor structure according to the initial position information and the correction coefficient to move the slice tissue slide point by point; the image preprocessing module is used to smooth and denoise the slice tissue image to obtain a denoised image; the three-dimensional model reconstruction module is used to perform three-dimensional reconstruction according to the denoised image to obtain a three-dimensional data model; the tumor prediction module is used to perform tumor judgment according to the three-dimensional data model to obtain the detection result. The present invention can realize tumor boundary recognition and three-dimensional reconstruction, improve the accuracy of detection, and provide a theoretical basis for tumor resection schemes.

Description

Technical Field

[0001] The invention relates to the technical field of medical detection, and in particular to an intelligent control system of a three-dimensional platform for THz spectrum detection of tumor tissue slices. Background Art

[0002] Gliomas are in a highly heterogeneous and transcellular state. Their communication with the surrounding microenvironmental components will affect the expression of various glioma markers and change the phenotype of non-tumor astrocytes through the microenvironment. With the development of molecular pathology, it has been found that gliomas with different molecular classifications (such as IDH mutations and wild-type) have significantly different treatment responsiveness and final prognosis. However, in the current diagnosis and treatment process (especially during the first surgery), the results of molecular pathology research have not been applied to the diagnosis and treatment process in a timely manner to achieve differentiated surgical strategies and guide intraoperative chemoradiotherapy and ultra-early postoperative chemotherapy; the core reason is that it is impossible to obtain molecular pathology results containing position information in real time or quasi-real time during surgery. The rapid response of THz waves overcomes the sampling bias caused by the spatial heterogeneity of the tumor, and is expected to achieve non-destructive and non-labeled dynamic identification of the spatial distribution of tumor molecular classifications during surgery, so as to carry out precise treatment for molecular classification at an early stage.

[0003] The current terahertz spectroscopy system (TAS7500SP) is a terahertz (THz) analysis system composed of a fiber laser module and a data acquisition module. THz irradiates biological tissue samples and interacts with the tissue to be tested. Subsequently, the terahertz waves carrying sample information are collected by a photoconductive detector and transmitted to a computer data analysis module. This system is popular among clinical researchers for its advantages such as integration and simple operation. However, the existing modules have the most complete information about tumors in fresh tissues, and the procedures of fixation, dehydration, and embedding are cumbersome and damage the molecular structure information of some tumors; however, fresh tissue slices are easily damaged and cannot be moved.

[0004] Therefore, the system can currently only collect information on a single light spot in a tissue slice, and there is an urgent need to find a module that can quickly scan biological tissue slices point by point to achieve information collection and analysis of spatial heterogeneity within the tissue. Summary of the invention

[0005] In order to overcome the deficiencies of the prior art, the object of the present invention is to provide an intelligent control system for a three-dimensional platform for THz spectrum detection of tumor tissue slices.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] An intelligent control system for a three-dimensional platform for THz spectrum detection of tumor tissue slices, the three-dimensional platform comprising a movable table, a motor structure and a detector; a slice tissue slide to be detected is arranged on the movable table; the motor structure is used to control the movable table to move, so as to realize the detection of the slice tissue slide by the detector, and the intelligent control system comprises:

[0008] A correction coefficient acquisition module is used to perform a detection scan according to a preset tissue slice and determine the correction coefficient according to the result of the detection scan;

[0009] A positioning scanning module, used for performing positioning scanning on the sliced ​​tissue slide to obtain initial position information of the sliced ​​tissue slide;

[0010] A point-by-point scanning module, used for controlling the working state of the motor structure according to the initial position information and the correction coefficient, so as to move the slice tissue slide point by point to obtain a plurality of slice tissue images;

[0011] An image preprocessing module, used for smoothing and denoising the slice tissue image to obtain a denoised image;

[0012] A three-dimensional model reconstruction module, used to perform three-dimensional reconstruction based on the denoised image to obtain a three-dimensional data model;

[0013] The tumor prediction module is used to make tumor judgment based on the three-dimensional data model to obtain detection results.

[0014] Preferably, the correction coefficient acquisition module includes:

[0015] A sample scanning unit, used for placing the glass slide with the tissue slice embedded therein on the movable table, and performing a comprehensive initial scan to record the theoretical position of each scanning feature point of the tissue slice and the actual position obtained by scanning;

[0016] The correction coefficient calculation unit is used to calculate a set of correction coefficients according to the deviation between the theoretical position and the actual position by using the least square method or other optimization algorithms.

[0017] Preferably, the image preprocessing module includes:

[0018] A neighborhood determination unit, used for taking a neighborhood with each pixel point in the slice tissue image as the center;

[0019] A pixel feature value determination unit, used to calculate the pixel feature value of each pixel point in the corresponding neighborhood;

[0020] A smoothing model building unit, used to build a smoothing denoising model according to the pixel feature value;

[0021] A denoising unit is used to denoise the slice tissue image using the smoothing denoising model to obtain the denoised image.

[0022] Preferably, the formula of the smoothing denoising model is:

[0023]

[0024] Wherein, f(x) is the smoothing denoising model, Represents the gradient value of pixel point x in the corresponding neighborhood, is the pixel mean of the neighborhood centered on pixel point x, u(x) represents the pixel value of pixel point x, and u median (x) represents the pixel median of the neighborhood centered on pixel x.

[0025] Preferably, the denoising unit comprises:

[0026] A smoothing value calculation subunit, used to calculate the smoothing value of each pixel point using the smoothing denoising model;

[0027] The mean filter subunit is used to denoise all pixel points whose smoothing values ​​are greater than a denoising threshold using a mean filter method to obtain the denoised image.

[0028] Preferably, the three-dimensional model reconstruction module includes:

[0029] A first feature extraction unit, used for performing edge detection and key point detection according to the denoised image to obtain image feature points;

[0030] An image registration unit, used for performing image registration on the denoised image according to the image feature points to obtain a registered image;

[0031] A three-dimensional reconstruction unit is used to perform depth estimation and image reconstruction according to the registered image to obtain the three-dimensional data model.

[0032] Preferably, the tumor prediction module comprises:

[0033] A second feature extraction unit, configured to extract tumor features according to the three-dimensional data model to obtain extracted feature parameters;

[0034] The comparison unit is used to compare the extracted characteristic parameters according to a preset tumor characteristic parameter model to determine whether the slice tissue to be detected is tumor tissue.

[0035] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0036] The present invention provides an intelligent control system of a three-dimensional platform for THz spectrum detection of tumor tissue slices. The three-dimensional platform comprises a movable table, a motor structure and a detector; a slice tissue slide to be detected is arranged on the movable table; the motor structure is used to control the movable table to move, so as to realize the detection of the slice tissue slide by the detector, and the intelligent control system comprises: a correction coefficient acquisition module, which is used to perform detection scanning according to preset tissue slices and determine the correction coefficient according to the result of the detection scanning; a positioning scanning module, which is used to perform positioning scanning on the slice tissue slide to obtain initial position information of the slice tissue slide; a point-by-point scanning module, which is used to control the working state of the motor structure according to the initial position information and the correction coefficient, so as to move the slice tissue slide point by point to obtain multiple slice tissue images; an image preprocessing module, which is used to perform smoothing and denoising on the slice tissue image to obtain a denoised image; a three-dimensional model reconstruction module, which is used to perform three-dimensional reconstruction according to the denoised image to obtain a three-dimensional data model; and a tumor prediction module, which is used to perform tumor judgment according to the three-dimensional data model to obtain a detection result. The present invention can realize high-precision three-dimensional detection of tumor slices, which not only improves the accuracy and efficiency of detection, but also can provide more detailed data support for subsequent tumor research and treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a cover glass provided in an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of a glass slide provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] 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.

[0042] The purpose of the present invention is to provide an intelligent control system for a three-dimensional platform for THz spectrum detection of tumor tissue slices, which can achieve high-precision three-dimensional detection of tumor slices, not only improving the accuracy and efficiency of detection, but also providing more detailed data support for subsequent tumor research and treatment.

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides an intelligent control system of a three-dimensional platform for THz spectrum detection of tumor tissue slices, wherein the three-dimensional platform comprises a movable table, a motor structure and a detector; a slice tissue slide to be detected is arranged on the movable table; the motor structure is used to control the movable table to move so as to realize the detection of the slice tissue slide by the detector, and the intelligent control system comprises:

[0045] A correction coefficient acquisition module is used to perform a detection scan according to a preset tissue slice and determine the correction coefficient according to the result of the detection scan;

[0046] A positioning scanning module, used for performing positioning scanning on the sliced ​​tissue slide to obtain initial position information of the sliced ​​tissue slide;

[0047] A point-by-point scanning module, used for controlling the working state of the motor structure according to the initial position information and the correction coefficient, so as to move the slice tissue slide point by point to obtain a plurality of slice tissue images;

[0048] An image preprocessing module, used for smoothing and denoising the slice tissue image to obtain a denoised image;

[0049] A three-dimensional model reconstruction module, used to perform three-dimensional reconstruction based on the denoised image to obtain a three-dimensional data model;

[0050] The tumor prediction module is used to make tumor judgment based on the three-dimensional data model to obtain detection results.

[0051] Preferably, the correction coefficient acquisition module includes:

[0052] A sample scanning unit, used for placing the glass slide with the tissue slice embedded therein on the movable table, and performing a comprehensive initial scan to record the theoretical position of each scanning feature point of the tissue slice and the actual position obtained by scanning;

[0053] The correction coefficient calculation unit is used to calculate a set of correction coefficients according to the deviation between the theoretical position and the actual position by using the least square method or other optimization algorithms.

[0054] Specifically, in this embodiment, in order to achieve precise adjustment of the spatial position when scanning the tumor slide point by point, we can introduce the concept of a correction coefficient, which is used to adjust the position of the three-dimensional mobile platform to compensate for the non-ideal factors of the system (such as mechanical errors, deformation caused by temperature changes, etc.).

[0055] Exemplarily, the present embodiment first selects or prepares a group of tissue slices with known geometric features. These samples should have feature points that are easy to identify and measure for subsequent correction coefficient calculations. Secondly, the tissue slices are placed on a three-dimensional mobile platform and a comprehensive initial scan is performed. The theoretical position of each feature point and the actual position obtained by the scan are recorded. Then, based on the deviation between the theoretical position and the actual position, one or a group of correction coefficients are calculated by the least squares method or other optimization algorithms. The correction coefficients of this embodiment will be used to adjust the moving path of the three-dimensional mobile platform to reduce the deviation.

[0056] Optionally, the feature points in this step are those points on the tissue slice that are easy to identify and accurately measure, and they are used as reference points in image processing and analysis to evaluate and correct the performance of the scanning system. The selection of these points is critical to the subsequent correction process because their positional accuracy directly affects the accuracy of the correction coefficients and the accuracy of the final tumor slide scan.

[0057] Furthermore, before starting point-by-point scanning, this embodiment applies the previously calculated correction coefficient to the control system of the mobile platform. In this way, each time the platform moves, it automatically adjusts its position according to the correction coefficient to compensate for possible errors. The three-dimensional mobile platform is controlled to move the tumor slide point by point according to a predetermined scanning path and step length. For each point, a high-resolution scanner will perform image acquisition. Since the correction coefficient has been applied, it can be ensured that the acquired image data has a high spatial position accuracy. Then the image data of each point is collected and recorded. These data can be further processed through image processing and analysis software, such as performing subsequent three-dimensional reconstruction, feature extraction and other steps.

[0058] Furthermore, by analyzing the acquired image data and combining it with the spatial positioning accuracy after the correction coefficient is optimized, the cell structure and tumor characteristics in the tumor slide can be more accurately identified and analyzed, thereby improving the accuracy and efficiency of tumor detection.

[0059] In this embodiment, the correction coefficient is introduced and applied to the control system of the three-dimensional mobile platform, which can significantly improve the spatial position accuracy of the tumor slide during point-by-point scanning. This not only helps to improve the accuracy of tumor detection, but also provides more reliable data support for subsequent research and analysis.

[0060] Preferably, the image preprocessing module includes:

[0061] A neighborhood determination unit, used for taking a neighborhood with each pixel point in the slice tissue image as the center;

[0062] A pixel feature value determination unit, used to calculate the pixel feature value of each pixel point in the corresponding neighborhood;

[0063] A smoothing model building unit, used to build a smoothing denoising model according to the pixel feature value;

[0064] A denoising unit is used to denoise the slice tissue image using the smoothing denoising model to obtain the denoised image.

[0065] Preferably, the formula of the smoothing denoising model is:

[0066]

[0067] Wherein, f(x) is the smoothing denoising model, Represents the gradient value of pixel point x in the corresponding neighborhood, is the pixel mean of the neighborhood centered on pixel point x, u(x) represents the pixel value of pixel point x, and u median (x) represents the pixel median of the neighborhood centered on pixel x.

[0068] Preferably, the denoising unit comprises:

[0069] A smoothing value calculation subunit, used to calculate the smoothing value of each pixel point using the smoothing denoising model;

[0070] The mean filter subunit is used to denoise all pixel points whose smoothing values ​​are greater than a denoising threshold using a mean filter method to obtain the denoised image.

[0071] In practical applications, the present invention can set a denoising threshold according to the smoothing value of the pixel point, and at the same time, perform mean filtering on all the pixels in the noise interval to obtain the median pixel point, and then use the median pixel point to replace the values ​​of all the pixels in the neighborhood, so as to obtain a denoised image. The original filtering algorithm, such as the mean filtering algorithm, performs mean processing on the pixels in each neighborhood on the slice tissue image, so the processed image will become blurred, and the invention can find the noise points on the image by using the smoothing denoising model, and then perform mean filtering on the corresponding noise points, which can smooth out the noise points in the image while maximally maintaining the original information of the target image.

[0072] Preferably, the three-dimensional model reconstruction module includes:

[0073] A first feature extraction unit, used for performing edge detection and key point detection according to the denoised image to obtain image feature points;

[0074] An image registration unit, used for performing image registration on the denoised image according to the image feature points to obtain a registered image;

[0075] A three-dimensional reconstruction unit is used to perform depth estimation and image reconstruction according to the registered image to obtain the three-dimensional data model.

[0076] Optionally, before the three-dimensional reconstruction process, this embodiment also performs preprocessing processes such as contrast enhancement and normalization on the image data. By adjusting the contrast of the image, the tumor cells and other structures in the image are made clearer, and by normalizing the image, it is ensured that all images have the same brightness and contrast levels.

[0077] Furthermore, this embodiment extracts useful features from the preprocessed image through edge detection and key point detection, wherein edge detection is used to identify the edges of tumor cells in the image, and key point detection is used to find key points in the image, which are highly identifiable between different images.

[0078] Furthermore, this embodiment aligns images from multiple perspectives into the same coordinate system, specifically by matching feature points detected in one image with corresponding feature points in another image, and calculating the transformations (such as rotation, translation, and scaling) required to map one image to another.

[0079] Furthermore, this embodiment uses the registered image data to generate a three-dimensional data model through the following steps:

[0080] Depth estimation: Estimate the depth information of each pixel based on the image registration results.

[0081] Voxel reconstruction: converts depth information into voxels (three-dimensional pixels), each voxel represents a point in three-dimensional space.

[0082] Surface reconstruction: Based on the voxel model, reconstruct the 3D surface of the tumor cells. This usually involves some smoothing to generate a more realistic 3D model.

[0083] Through the above steps, the collected image data is used to perform three-dimensional reconstruction of tumor slices to form a three-dimensional data model, which can provide doctors with more comprehensive information to better understand the characteristics of the tumor and plan treatment.

[0084] Preferably, the tumor prediction module comprises:

[0085] A second feature extraction unit, configured to extract tumor features according to the three-dimensional data model to obtain extracted feature parameters;

[0086] The comparison unit is used to compare the extracted characteristic parameters according to a preset tumor characteristic parameter model to determine whether the slice tissue to be detected is tumor tissue.

[0087] Specifically, the steps for determining a tumor in this embodiment are as follows:

[0088] Feature extraction: Based on the three-dimensional data model, extract the characteristic parameters of the tumor, such as volume, morphology, edge characteristics, etc.

[0089] Tumor judgment: Based on the preset tumor characteristic parameter model, the three-dimensional data model in the slice is compared and analyzed to determine whether it is tumor tissue, as well as the specific type and grade of the tumor.

[0090] like Figure 2 and Figure 3 As shown, the glass slide in this embodiment is a quartz glass slide, and its size is 25.4*76.2*0.5mm. The black rectangular block in the figure is a rectangular groove with a width of 1mm and a depth of 0.5mm, and the gray block is a rectangular groove with a width of 20*40mm and a depth of 0.1mm. The gray area is used to place tissue slices, which can meet (50-100um thickness, this thickness has been tested in the early stage and is suitable for optical detection of this instrument); the black groove is used to apply vaseline. After the early experiments and the effect of vaseline, the hydration of the tissue slice can be avoided or slowed down. At the same time, the interference phenomenon in the detection process can be avoided.

[0091] The beneficial effects of the present invention are as follows:

[0092] The present invention can realize high-precision three-dimensional detection of tumor slices, which not only improves the accuracy and efficiency of detection, but also can provide more detailed data support for subsequent tumor research and treatment.

[0093] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0094] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. An intelligent control system for a three-dimensional platform for THz spectrum detection of tumor tissue slices, the three-dimensional platform comprising a movable table, a motor structure and a sample stage for a THz spectrum system; a tissue slice to be detected is arranged on the movable table; the motor structure is used to control the movable table to move so that the movable table can detect the slice tissue, characterized in that: The intelligent control system comprises: A correction coefficient acquisition module is used to perform a detection scan according to a preset tissue slice and determine the correction coefficient according to the result of the detection scan; A positioning scanning module, used for performing positioning scanning on the tissue slice to obtain initial position information of the tissue slice slide; A point-by-point scanning module, used for controlling the working state of the motor structure according to the initial position information and the correction coefficient, so as to move the sliced ​​tissue slide point by point to obtain a tissue slice point array image; An image preprocessing module, used for smoothing and denoising the tissue slice image to obtain a denoised image; A three-dimensional model reconstruction module, used to perform three-dimensional reconstruction based on the denoised image to obtain a three-dimensional data model; A tumor prediction module, used to make tumor judgment based on the three-dimensional data model and obtain a detection result; The correction coefficient acquisition module comprises: A sample scanning unit, used for placing the glass slide with the tissue slice embedded therein on the movable table, and performing a comprehensive initial scan to record the theoretical position of each scanning feature point of the tissue slice and the actual position obtained by scanning; A correction coefficient calculation unit, used to calculate a set of correction coefficients according to the deviation between the theoretical position and the actual position by using a least square method or other optimization algorithm; The image preprocessing module comprises: A neighborhood determination unit, used for taking a neighborhood with each pixel point in the slice tissue image as the center; A pixel feature value determination unit, used to calculate the pixel feature value of each pixel point in the corresponding neighborhood; A smoothing model building unit, used to build a smoothing denoising model according to the pixel feature value; A denoising unit, configured to denoise the slice tissue image using the smoothing denoising model to obtain the denoised image; The formula of the smoothing denoising model is: ； in, The smoothing denoising model, Represents pixel x The gradient value in the corresponding neighborhood, Pixel x The pixel mean of the neighborhood centered at Represents pixel x The pixel value of Indicated in pixels x The pixel median of the neighborhood centered at is; the denoising unit comprises: A smoothing value calculation subunit, used to calculate the smoothing value of each pixel point using the smoothing denoising model; A mean filter subunit is used to denoise all pixels whose smoothing values ​​are greater than a denoising threshold using a mean filter method to obtain the denoised image; The three-dimensional model reconstruction module comprises: A first feature extraction unit, used for performing edge detection and key point detection according to the denoised image to obtain image feature points; An image registration unit, used for performing image registration on the denoised image according to the image feature points to obtain a registered image; A three-dimensional reconstruction unit, used for performing depth estimation and image reconstruction according to the registered image to obtain the three-dimensional data model; The tumor prediction module comprises: A second feature extraction unit, configured to extract tumor features according to the three-dimensional data model to obtain extracted feature parameters; The comparison unit is used to compare the extracted characteristic parameters according to a preset tumor characteristic parameter model to determine whether the slice tissue to be detected is tumor tissue.

Images