[go: up one dir, main page]

CN116681685B - Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data - Google Patents

Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data

Info

Publication number
CN116681685B
CN116681685B CN202310722477.XA CN202310722477A CN116681685B CN 116681685 B CN116681685 B CN 116681685B CN 202310722477 A CN202310722477 A CN 202310722477A CN 116681685 B CN116681685 B CN 116681685B
Authority
CN
China
Prior art keywords
image
cell
calcium imaging
photon
video
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310722477.XA
Other languages
Chinese (zh)
Other versions
CN116681685A (en
Inventor
陈岗
李珍
徐斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN202310722477.XA priority Critical patent/CN116681685B/en
Publication of CN116681685A publication Critical patent/CN116681685A/en
Application granted granted Critical
Publication of CN116681685B publication Critical patent/CN116681685B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30024—Cell structures in vitro; Tissue sections in vitro

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

本发明公开了一种基于双光子钙成像数据的细胞胞体自动识别与分割方法及装置，首先获取细胞的双光子钙成像视频，对每一帧视频图像进行图像去噪与图像增强。然后将双光子钙成像视频降维成局部相关摘要图像，在局部相关摘要图像上应用基于多尺度圆点增强的细胞识别算法，计算成像区域内不同尺寸的近圆结构的中心坐标，即细胞胞体的种子点。最后对于每一个种子点，将双光子钙成像视频裁剪为以种子点为中心大小固定的区块视频，在区块视频上应用基于椭圆形状约束活动轮廓模型的细胞分割算法，得到细胞胞体的轮廓。本发明能够提高运算效率、降低数据存储量。

The present invention discloses a method and device for automatic identification and segmentation of cell soma based on two-photon calcium imaging data. First, a two-photon calcium imaging video of the cell is acquired, and image denoising and image enhancement are performed on each frame of the video image. The two-photon calcium imaging video is then reduced in dimension into a local correlation summary image. A cell recognition algorithm based on multi-scale dot enhancement is applied to the local correlation summary image to calculate the center coordinates of near-circular structures of different sizes within the imaging area, i.e., the seed points of the cell soma. Finally, for each seed point, the two-photon calcium imaging video is cropped into a block video of a fixed size centered on the seed point. A cell segmentation algorithm based on an elliptical shape constrained active contour model is applied to the block video to obtain the outline of the cell soma. The present invention can improve computational efficiency and reduce data storage volume.

Description

Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data
Technical Field
The invention relates to the technical field of image processing and neurobiological data analysis, in particular to a method and a device for automatically identifying and dividing cell bodies based on two-photon calcium imaging data.
Background
The brain is a very important and complex organ, and many countries have already listed brain science and brain-like research as strategic level science and technology of the country, and the development of brain science has not been separated from the efficient and high-resolution neural activity recording technology. Two-photon calcium imaging technology is currently the standard tool for monitoring large numbers of neurons, recording the activity of large numbers of neurons in vivo at single cell resolution levels. However, due to the huge data volume generated by the two-photon calcium imaging technology, low signal-to-noise ratio of the data and the characteristic of simultaneously imaging hundreds or thousands of neurons, the difficulty in data analysis is brought about by the fact that the identification of cells manually is time-consuming and the standard is difficult to unify, and a plurality of algorithms are currently used for automatically, accurately and rapidly identifying the positions and the outlines of single nerve cells.
Aiming at the difficulty that the data volume of calcium imaging is huge, some cell identification and segmentation algorithms compress three-dimensional calcium imaging video into a two-dimensional abstract image by a method related to time sequence mean value, time sequence maximum value or time sequence, so that the data volume is only related to the size of an imaging area and is irrelevant to imaging time, and then cells are identified or segmented on the abstract image by using a traditional mode identification and image segmentation algorithm, but the method has the defects that neurons in the abstract image overlap due to the fact that the time dimension is abandoned, and the difficulty of cell identification and segmentation is increased.
Some cell identification and segmentation algorithms are directly based on calcium imaging video data, and the calcium imaging video data are regarded as spatial combinations of calcium fluorescence signals with different components, signals are clustered or calcium imaging videos are subjected to matrix decomposition, so that cell identification and segmentation are realized, but the method has the defects of large data storage amount and low operation efficiency, and the result of identification and segmentation is not the cell body of a neuron which is more interesting in research, and may also comprise structures such as axons, dendrites and the like of the neuron.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides an automatic cell body recognition and segmentation method based on two-photon calcium imaging data, which can improve the operation efficiency and reduce the data storage capacity, and solves the problems that cells are easy to leak out in cell recognition and the cell body recognition and segmentation result contains structures such as axons, dendrites and the like besides cells.
The invention aims at realizing the following technical scheme that in a first aspect, the invention provides a cell body automatic identification and segmentation method based on two-photon calcium imaging data, which comprises the following steps:
step 1, acquiring a two-photon calcium imaging video of a cell, and carrying out image denoising and image enhancement on each frame of video image;
Step 2, calculating time sequence correlation coefficients of fluorescent calcium signals of coordinates of each pixel point in the two-photon calcium imaging video and fluorescent calcium signals of a plurality of coordinate points in a neighborhood around the fluorescent calcium signals, solving an average value to serve as gray values of the pixel points, reducing the dimension of the two-photon calcium imaging video into a locally relevant abstract image, enhancing a circular structure with diameters in a certain range in the image by constructing a function related to characteristic values of hessian matrix on the locally relevant abstract image based on a cell recognition algorithm enhanced by multi-scale dots, and calculating center coordinates of circular structures with different dimensions in an imaging area, namely seed points of cell bodies;
Step 3, according to the seed points obtained in the step 2, cutting the two-photon calcium imaging video into block videos with fixed sizes taking the seed points as centers for each seed point, and applying a cell segmentation algorithm based on an elliptic shape constraint movable contour model on the block videos to obtain the contour of cell bodies, wherein the method comprises the steps of constructing a function related to a level set Energy function of (2)In the process of minimizingIs (are) simultaneously evolvedWherein the function isCorresponding to the contour of the cell, in order to construct an energy functionFirstly, constructing a data driving item according to the characteristics of two-photon calcium imaging data, measuring the difference of correlation vectors inside and outside the outline, and obtaining the correlation vector I (u) of each coordinate point u by calculating the correlation values of the fluorescent calcium signal of the position and the fluorescent calcium signals of all coordinate points v in the block:
wherein the pixel points u= (x, y) in the image, I (u) is a one-dimensional correlation vector, As a new feature of the pixel, N is the total number of pixel points in the block, v is all coordinate points in the block, Ω is defined as an image domain, and an image I is defined on the image domain, Ω -R N, secondly, an elliptical shape constraint term is added in an energy function by using a priori value of elliptical shape of a cell body, so that the finally obtained cell outline tends to be elliptical, and the method specifically comprises the following steps:
in the level set evolution process, fitting the contour into a parameterized ellipse, and limiting the level set evolution by using the ellipse, so that the contour of the final level set is also close to an ellipse, and alternately evolving as follows:
(1) Initializing level set functions And an elliptical shape phi (x, y);
(2) At the position of In the determined case, the evolution equation of phi (x, y) is obtained by minimizing the energy function by the gradient descent method, and since phi (x, y) is a parameterized ellipse, the evolution equation is determined by the center coordinate (x 0,y0) of the ellipse, the semi-major axis a of the ellipse, the semi-minor axis b of the ellipse, and the angle θ of rotation of the ellipse:
thus evolving phi (x, y), i.e., the update parameters (x 0,y0, a, b, θ);
(3) Determining phi (x, y), minimizing energy function by gradient descent method, and updating
(4) Repeating the steps (2) and (3) until reaching convergence condition, namely level set functionIs less than a set value or the iteration number is greater than the set value.
Further, the image denoising adopts an anisotropic filtering algorithm.
Further, the image enhancement adopts top-hat transformation to obtain a region with brighter gray scale in the original image.
Further, in step 2, based on the multi-scale dot enhanced cell recognition algorithm, the specific steps of calculating the center coordinates of the near-circular structures with different sizes in the imaging region are as follows:
(2.1) setting an upper limit sigma min and a lower limit sigma max of the scale parameter according to the upper limit and the lower limit of the cell radius in the locally relevant abstract image;
(2.2) calculating a eigenvalue lambda 1、λ2 of the Hessian matrix of each pixel point (x, y) in the single-scale gaussian space for each scale parameter sigma in the upper and lower limits of the scale parameter;
(2.3) introducing eigenvalues lambda 3, wherein the eigenvalues of the Hessian matrix corresponding to the spherical structure in the three-dimensional space meet lambda 1≈λ2≈λ3 <0, the eigenvalues of the Hessian matrix corresponding to the tubular structure meet |lambda 2|≈|λ3|≥λ1, and setting a parameter tau to regularize lambda ρ;
(2.4) calculating a spherical Structure similarity function
(2.5) Repeating the steps (2.2) to (2.4), and calculating VR (x, y, sigma) corresponding to all scale parameters sigma;
(2.6) taking the maximum value of VR (x, y, sigma) of each pixel point (x, y) under all scale parameters to obtain a multiscale spherical structure similarity function VR (x, y), and obtaining a spherical structure enhanced image;
(2.7) setting a threshold value TH according to a3 sigma principle of normal distribution in a probability theory, and binarizing the spherical structure enhanced image, wherein the pixel points belonging to the spherical structure have values of 1 and the other pixel points have values of 0;
(2.8) removing the structure with the area which does not meet the requirement in the figure by using a morphological filtering method;
(2.9) calculating the gravity center of each spherical structure in the graph to obtain the coordinates of the seed points.
In a second aspect, the invention further provides an automatic cell body recognition and segmentation device based on two-photon calcium imaging data, which comprises a memory and one or more processors, wherein executable codes are stored in the memory, and the automatic cell body recognition and segmentation method based on the two-photon calcium imaging data is realized when the executable codes are executed by the processors.
In a third aspect, the present invention further provides a computer readable storage medium having a program stored thereon, which when executed by a processor, implements the method for automatically identifying and dividing cell bodies based on two-photon calcium imaging data.
The invention has the beneficial effects that:
1. The automatic processing of the visual cortex neuron fluorescence activity data acquired by the two-photon fluorescence microscope is realized, the positions of hundreds or thousands of neurons in an imaging area are automatically identified, the outline of a cell body is segmented, and the fluorescence calcium signals of the neurons are subsequently extracted for analyzing the neural mechanism of the cognitive function by combining behaviors, so that the method has important value for promoting the development of brain science.
2. Firstly, compressing a three-dimensional calcium imaging video into a two-dimensional abstract image, then identifying seed points of cells on the abstract image, making the data size of a cell identification algorithm only related to the size of an imaging area and not related to imaging time, then cutting the calcium imaging video into a block video by taking the seed points as the center, and then dividing the outline of the cells on the block video, so that the data size and the operation amount of a cell division algorithm are greatly reduced.
3. The cell recognition algorithm based on multi-scale dot enhancement is provided, so that the cell bodies of the pericytes with different sizes in an imaging area can be recognized, and structures such as axons, dendrites and the like of neurons are not recognized.
4. The cell segmentation algorithm based on the elliptic constraint movable contour model is provided, and the data driving term based on the correlation vector structure enables the segmented contour to be more accurate, the introduced elliptic constraint term enables the zero level set contour to be close to an ellipse, and therefore the segmented contour only comprises cell bodies, and structures such as axons, dendrites and the like are excluded.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an automatic cell body recognition and segmentation method based on two-photon calcium imaging data;
FIG. 2 is a graph showing the results of cell recognition;
FIG. 3 is a graph showing the results of cell segmentation;
FIG. 4 is a schematic diagram showing the action of elliptic constraint terms in a cell segmentation method;
fig. 5 is a block diagram of an automatic cell body recognition and segmentation device based on two-photon calcium imaging data.
Detailed Description
The following describes the embodiments of the present invention in further detail with reference to the drawings.
As shown in fig. 1, the method for automatically identifying and dividing cell bodies based on two-photon calcium imaging data provided by the invention comprises the following specific steps:
Step 1, acquiring two-photon calcium imaging video of cells, and carrying out image denoising and image enhancement on each frame of video image.
And 1.1, denoising the image, namely an anisotropic filtering algorithm.
And 1.2, enhancing the image, namely changing the top cap to obtain a region with brighter gray scale in the original image.
Step 2, calculating time sequence correlation coefficients of fluorescent calcium signals of coordinates of each pixel point in a two-photon calcium imaging video and fluorescent calcium signals of a plurality of coordinate points in a neighborhood around the fluorescent calcium signals, solving an average value to serve as gray values of the pixel points, reducing the dimension of the two-photon calcium imaging video into a locally relevant abstract image, enhancing a circular structure with diameters in a certain range in the image on the locally relevant abstract image based on a cell recognition algorithm enhanced by multi-scale dots by constructing a function (Frangi A F,Niessen W J,Vincken K L,et al.Multiscale vessel enhancement filtering[C].International conference on medical image computing and computer-assisted intervention.Springer,Berlin,Heidelberg,1998:130-137.) related to characteristic values of hessian matrixes, and calculating center coordinates of the circular structures with different dimensions in an imaging area, namely seed points of cell bodies, wherein the method comprises the following steps:
And 2.1, for all coordinate points in a two-photon calcium imaging video imaging region, respectively calculating time sequence correlation coefficients of fluorescent calcium signals of each coordinate point (x, y) and fluorescent calcium signals of 8 coordinate points in the surrounding vicinity thereof, taking the average value of the 8 time sequence correlation coefficients as the gray value of the pixel point, and finally obtaining a local correlation abstract image.
Step 2.2, a cell identification algorithm based on multi-scale dot enhancement enhances a circular structure with a diameter within a certain range in an image by constructing a function related to a characteristic value of hessian matrix, and calculates the center of the circular structure as a position coordinate of a cell body, wherein the specific steps are as follows:
(2.1) setting an upper limit sigma min and a lower limit sigma max of the scale parameter according to the upper limit and the lower limit of the cell radius in the locally relevant abstract image;
(2.2) calculating a eigenvalue lambda 1、λ2 of the Hessian matrix of each pixel point (x, y) in the single-scale gaussian space for each scale parameter sigma in the upper and lower limits of the scale parameter;
(2.3) introducing eigenvalues lambda 3, wherein the eigenvalues of the Hessian matrix corresponding to the spherical structure in the three-dimensional space meet lambda 1≈λ2≈λ3 <0, the eigenvalues of the Hessian matrix corresponding to the tubular structure meet |lambda 2|≈|λ3|≥λ1, and setting a parameter tau to regularize lambda ρ;
(2.4) calculating a spherical Structure similarity function
(2.5) Repeating the steps (2) to (4), and calculating VR (x, y, sigma) corresponding to all scale parameters sigma;
(2.6) taking the maximum value of VR (x, y, sigma) of each pixel point (x, y) under all scale parameters to obtain a multiscale spherical structure similarity function VR (x, y), and obtaining a spherical structure enhanced image;
(2.7) setting a proper threshold value TH according to the 3 sigma principle of normal distribution in the probability theory, and binarizing the spherical structure enhanced image, wherein the pixel points belonging to the spherical structure have values of 1 and the other pixel points have values of 0;
(2.8) removing the structure with the area smaller than 176um 2 in the figure by using a morphological filtering method;
(2.9) calculating the gravity center of each spherical structure in the graph to obtain the coordinates of the seed points.
And 3, according to the seed points obtained in the step 2, cutting the two-photon calcium imaging video into block videos with fixed sizes by taking the seed points as the center, and applying a cell segmentation algorithm based on an elliptic shape constraint movable contour model on the block videos to obtain the contour of a cell body, wherein the contour is shown in fig. 2 and 3.
Step 3.1, the length and width of the block video are determined by the diameter of cell bodies in the imaging area of the two-photon calcium imaging video, and 41 multiplied by 41.
And 3.2, a cell segmentation algorithm based on an elliptic shape constraint active contour model is an improvement on a distance regularized level set evolution model, and after the level set evolution is finished, a zero level set contour converges to the edge of a cell body, wherein the contour is the contour of the cell body. In particular, constructing a function about a level setEnergy function of (2)In the process of minimizingIs (are) simultaneously evolvedWherein the function isCorresponding to the contour of the cell, in order to construct an energy functionFirstly, constructing a data driving item according to the characteristics of two-photon calcium imaging data, secondly, because the neuron consists of structures such as a cell body, an axon, a dendrite and the like, the invention only wants to divide the outline of the cell body to calculate the fluorescence signal of the cell body in a subsequent step, and the invention adds an elliptic shape constraint item in an energy function by utilizing the priori value of the elliptical shape of the cell body, so that the finally obtained cell outline tends to be elliptical. The improvement of the level set evolution model of distance regularization comprises the following two points:
The improvement 1 is that the data driving item is not based on gray value any more, but the difference of correlation vectors inside and outside the outline is measured, the correlation vector I (u) of each coordinate point u is obtained by calculating the correlation value of the fluorescent calcium signal of the position and the fluorescent calcium signals of all coordinate points v in the block:
the improvement 2 is that an ellipse constraint term is added in the energy function, in the process of level set evolution, the contour is fitted into a parameterized ellipse, and the ellipse is used for limiting the evolution of the level set, so that the contour of the final level set is also close to an ellipse, and the steps of alternating evolution are as follows:
(1) Initializing level set functions And an elliptical shape phi (x, y);
(2) At the position of In the determined case, the evolution equation of phi (x, y) is obtained by minimizing the energy function by the gradient descent method, and since phi (x, y) is a parameterized ellipse, the evolution equation is determined by the center coordinate (x 0,y0) of the ellipse, the semi-major axis a of the ellipse, the semi-minor axis b of the ellipse, and the angle θ of rotation of the ellipse:
Thus evolving phi (x, y), i.e., the update parameters (x 0,y0, a, b, θ);
(3) Determining phi (x, y), minimizing energy function by gradient descent method, and updating
(4) Repeating the steps (2) and (3) until reaching convergence condition, namely level set functionIs less than a set value or the iteration number is greater than the set value.
The role of the elliptic constraint term in the cell segmentation method, as shown in a in fig. 4, b in fig. 4, and c in fig. 4, 3 columns show the results of the cell segmentation algorithm with and without elliptic constraint terms applied to 3 sets of data, respectively. The left side is where no elliptical constraint is used, while the right side uses elliptical constraint. The results of the three rows above show that the introduction of the elliptic constraint term can limit the segmentation result to the neuron cell body without considering the axons and dendrites of the neurons, while the results of the last row show that for overlapping neurons, the introduction of the elliptic constraint term can also prevent the segmentation result from overflowing the boundary of the real cells.
Corresponding to the embodiment of the automatic cell body identification and segmentation method based on the two-photon calcium imaging data, the invention also provides an embodiment of the automatic cell body identification and segmentation device based on the two-photon calcium imaging data.
Referring to fig. 5, the device for automatically identifying and dividing a cell body based on two-photon calcium imaging data according to the embodiment of the present invention includes a memory and one or more processors, wherein executable codes are stored in the memory, and when the processor executes the executable codes, the processor is configured to implement the method for automatically identifying and dividing a cell body based on two-photon calcium imaging data according to the above embodiment.
The embodiment of the automatic cell body identification and segmentation device based on the two-photon calcium imaging data can be applied to any equipment with data processing capability, and the equipment with data processing capability can be equipment or a device such as a computer. The apparatus embodiments may be implemented by software, or may be implemented by hardware or a combination of hardware and software. Taking software implementation as an example, the device in a logic sense is formed by reading corresponding computer program instructions in a nonvolatile memory into a memory by a processor of any device with data processing capability. From the hardware level, as shown in fig. 5, a hardware structure diagram of an apparatus with optional data processing capability, where the automatic cell body recognition and segmentation apparatus based on two-photon calcium imaging data is located, is provided in the present invention, and in addition to the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 5, the apparatus with optional data processing capability in the embodiment generally includes other hardware according to the actual function of the apparatus with optional data processing capability, which is not described herein again.
The implementation process of the functions and roles of each unit in the above device is specifically shown in the implementation process of the corresponding steps in the above method, and will not be described herein again.
For the device embodiments, reference is made to the description of the method embodiments for the relevant points, since they essentially correspond to the method embodiments. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purposes of the present invention. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The embodiment of the invention also provides a computer readable storage medium, on which a program is stored, which when executed by a processor, implements the automatic cell body recognition and segmentation method based on two-photon calcium imaging data in the above embodiment.
The computer readable storage medium may be an internal storage unit, such as a hard disk or a memory, of any of the data processing enabled devices described in any of the previous embodiments. The computer readable storage medium may also be an external storage device of any device having data processing capabilities, such as a plug-in hard disk, a smart memory card (SMART MEDIA CARD, SMC), an SD card, a flash memory card (FLASH CARD), etc. provided on the device. Further, the computer readable storage medium may include both internal storage units and external storage devices of any data processing device. The computer readable storage medium is used for storing the computer program and other programs and data required by the arbitrary data processing apparatus, and may also be used for temporarily storing data that has been output or is to be output.
The above-described embodiments are intended to illustrate the present invention, not to limit it, and any modifications and variations made thereto are within the spirit of the invention and the scope of the appended claims.

Claims (6)

1.一种基于双光子钙成像数据的细胞胞体自动识别与分割方法，其特征在于，该方法包括以下步骤：1. A method for automatic identification and segmentation of cell bodies based on two-photon calcium imaging data, characterized in that the method comprises the following steps: 步骤1、获取细胞的双光子钙成像视频，对每一帧视频图像进行图像去噪与图像增强；Step 1: Obtain two-photon calcium imaging video of cells and perform image denoising and image enhancement on each frame of video image; 步骤2、计算双光子钙成像视频中每个像素点坐标的荧光钙信号与其周围邻域内若干个坐标点的荧光钙信号的时序相关系数并求均值作为像素点的灰度值，将双光子钙成像视频降维成局部相关摘要图像，在局部相关摘要图像上基于多尺度圆点增强的细胞识别算法，通过构造与hessian矩阵的特征值有关的函数，增强图像中直径在一定范围内的圆形结构，计算成像区域内不同尺寸的圆形结构的中心坐标，即细胞胞体的种子点；Step 2: Calculate the temporal correlation coefficient between the fluorescent calcium signal of each pixel coordinate in the two-photon calcium imaging video and the fluorescent calcium signals of several coordinate points in its surrounding neighborhood, and calculate the average value as the grayscale value of the pixel. Reduce the dimension of the two-photon calcium imaging video into a local correlation summary image. Based on the local correlation summary image, a cell recognition algorithm based on multi-scale dot enhancement is used to enhance circular structures with a certain diameter range in the image by constructing a function related to the eigenvalues of the Hessian matrix. Calculate the center coordinates of circular structures of different sizes in the imaging area, i.e., the seed points of the cell body. 步骤3、根据步骤2得到的种子点，对于每一个种子点，将双光子钙成像视频裁剪为以种子点为中心大小固定的区块视频，在区块视频上应用基于椭圆形状约束活动轮廓模型的细胞分割算法，得到细胞胞体的轮廓；具体为：构造关于水平集函数的能量函数在最小化的同时演化其中函数对应于细胞的轮廓，为了构造能量函数首先根据双光子钙成像数据的特征构造数据驱动项，衡量轮廓内外相关性向量的差异，每个坐标点u的相关性向量I(u)，通过计算该位置的荧光钙信号与区块内所有坐标点v的荧光钙信号的相关值后得到：Step 3: Based on the seed points obtained in step 2, for each seed point, the two-photon calcium imaging video is cropped into a block video with a fixed size centered on the seed point. The cell segmentation algorithm based on the elliptical shape constrained active contour model is applied to the block video to obtain the outline of the cell body; specifically: construct a level set function Energy function In minimization Simultaneous evolution The function Corresponding to the outline of the cell, in order to construct the energy function First, a data-driven term is constructed based on the characteristics of two-photon calcium imaging data to measure the difference in correlation vectors inside and outside the contour. The correlation vector I(u) of each coordinate point u is obtained by calculating the correlation value of the fluorescent calcium signal at that position with the fluorescent calcium signals of all coordinate points v in the block: 其中图像中的像素点u＝(x,y)，I(u)为一维的相关性向量，作为该像素的新特征，N是区块内像素点的总数，v是区块内所有坐标点，Ω定义为图像域，图像I:Ω→RN被定义在图像域上；其次，利用胞体呈椭圆形状的先验值，在能量函数中增加一个椭圆形状约束项，使得最终获得的细胞轮廓趋于椭圆；具体如下：Wherein the pixel point u in the image is (x, y), I(u) is a one-dimensional correlation vector, As the new feature of the pixel, N is the total number of pixels in the block, v is the coordinates of all points in the block, Ω is defined as the image domain, and the image I:Ω→ RN is defined in the image domain. Secondly, using the prior value that the cell body is elliptical, an elliptical shape constraint term is added to the energy function, so that the final cell contour tends to be elliptical. The details are as follows: 在水平集演化的过程中，将轮廓拟合为一个参数化的椭圆，并用这个椭圆限制水平集的演化，使得最终水平集的轮廓也接近一个椭圆，交替演化的步骤如下：During the level set evolution process, the contour is fitted into a parameterized ellipse, and the ellipse is used to constrain the evolution of the level set so that the contour of the final level set is also close to an ellipse. The steps of alternating evolution are as follows: (1)初始化水平集函数与椭圆形状φ(x,y)；(1) Initialize the level set function with elliptical shape φ(x,y); (2)在确定的情况下，通过梯度下降法最小化能量函数，得到φ(x,y)的演化方程，由于φ(x,y)是参数化的椭圆，由椭圆的中心坐标(x0,y0)、椭圆的半长轴a、椭圆的半短轴b、椭圆旋转的角度θ确定：(2) When the energy function is determined, the gradient descent method is used to minimize the energy function and obtain the evolution equation of φ(x,y). Since φ(x,y) is a parameterized ellipse, it is determined by the center coordinates of the ellipse (x 0 ,y 0 ), the semi-major axis a of the ellipse, the semi-minor axis b of the ellipse, and the angle θ of the ellipse rotation: 因此演化φ(x,y)即更新参数(x0,y0,a,b,θ)；Therefore, evolving φ(x,y) means updating the parameters (x 0 ,y 0 ,a,b,θ); (3)确定φ(x,y)后，再通过梯度下降法最小化能量函数，更新 (3) After determining φ(x,y), the energy function is minimized by gradient descent and updated (4)重复步骤(2)、步骤(3)直至达到收敛条件，即水平集函数的变化小于设定值或迭代次数大于设定值。(4) Repeat steps (2) and (3) until the convergence condition is reached, that is, the level set function The change is less than the set value or the number of iterations is greater than the set value. 2.根据权利要求1所述的一种基于双光子钙成像数据的细胞胞体自动识别与分割方法，其特征在于，所述图像去噪采用各向异性滤波算法。2. The method for automatic cell body recognition and segmentation based on two-photon calcium imaging data according to claim 1, wherein the image denoising adopts an anisotropic filtering algorithm. 3.根据权利要求1所述的一种基于双光子钙成像数据的细胞胞体自动识别与分割方法，其特征在于，所述图像增强采用顶帽变换，得到原图中灰度较亮的区域。3. A method for automatic cell body recognition and segmentation based on two-photon calcium imaging data according to claim 1, characterized in that the image enhancement adopts top hat transformation to obtain a brighter grayscale area in the original image. 4.根据权利要求1所述的一种基于双光子钙成像数据的细胞胞体自动识别与分割方法，其特征在于，步骤2中，基于多尺度圆点增强的细胞识别算法，计算成像区域内不同尺寸的近圆结构的中心坐标的具体步骤如下：4. The method for automatic cell body recognition and segmentation based on two-photon calcium imaging data according to claim 1, wherein in step 2, the specific steps of calculating the center coordinates of near-circular structures of different sizes within the imaging area based on a multi-scale dot-enhanced cell recognition algorithm are as follows: (2.1)根据局部相关摘要图像中细胞半径的上下限设置尺度参数的上限σmin与下限σmax；(2.1) Setting the upper limit σ min and lower limit σ max of the scale parameter according to the upper and lower limits of the cell radius in the local correlation summary image; (2.2)对于尺度参数上下限内的每个尺度参数σ，计算单尺度高斯空间内每个像素点(x,y)的Hessian矩阵的特征值λ1、λ2；(2.2) For each scale parameter σ within the upper and lower limits of the scale parameter, calculate the eigenvalues λ 1 and λ 2 of the Hessian matrix of each pixel point (x, y) in the single-scale Gaussian space; (2.3)引入特征值λ3，其中三维空间中球状结构对应的Hessian矩阵特征值满足λ1≈λ2≈λ3＜0,管状结构对应的Hessian矩阵特征值满足|λ2|≈|λ3|≥λ1，并设置参数τ将其正则化为λρ；(2.3) Introduce the eigenvalue λ 3 , where the eigenvalue of the Hessian matrix corresponding to the spherical structure in three-dimensional space satisfies λ 1 ≈λ 2 ≈λ 3 ＜0, and the eigenvalue of the Hessian matrix corresponding to the tubular structure satisfies |λ 2 |≈|λ 3 |≥λ 1 , and set the parameter τ to regularize it to λ ρ ; (2.4)计算球状结构相似性函数 (2.4) Calculate the spherical structure similarity function (2.5)重复步骤(2.2)到步骤(2.4)，计算所有尺度参数σ对应的VR(x,y,σ)；(2.5) Repeat steps (2.2) to (2.4) to calculate VR(x, y, σ) corresponding to all scale parameters σ; (2.6)对于每个像素点(x,y)，取其在所有尺度参数下的VR(x,y,σ)的最大值，得到多尺度球状结构相似性函数VR(x,y)，即得到球状结构增强图像；(2.6) For each pixel (x, y), take the maximum value of VR(x, y, σ) under all scale parameters to obtain the multi-scale spherical structure similarity function VR(x, y), that is, the spherical structure enhanced image; (2.7)根据概率学理论中正态分布的3σ原则设置阈值TH，将球状结构增强图像二值化，其中属于球状结构的像素点值为1，其他像素点的值为0；(2.7) According to the 3σ principle of normal distribution in probability theory, the threshold TH is set to binarize the spherical structure enhanced image, where the pixel value of the spherical structure is 1 and the value of other pixels is 0; (2.8)使用形态学滤波方法去除图中面积不符合需求的结构；(2.8) Use morphological filtering to remove structures whose area does not meet the requirements; (2.9)计算图中每个圆球状结构的重心，得到种子点坐标。(2.9) Calculate the center of gravity of each spherical structure in the figure and obtain the coordinates of the seed point. 5.一种基于双光子钙成像数据的细胞胞体自动识别与分割装置，包括存储器和一个或多个处理器，所述存储器中存储有可执行代码，其特征在于，所述处理器执行所述可执行代码时，实现如权利要求1-4中任一项所述的一种基于双光子钙成像数据的细胞胞体自动识别与分割方法。5. A device for automatic identification and segmentation of cell bodies based on two-photon calcium imaging data, comprising a memory and one or more processors, wherein the memory stores executable code, and wherein when the processor executes the executable code, it implements a method for automatic identification and segmentation of cell bodies based on two-photon calcium imaging data as described in any one of claims 1 to 4. 6.一种计算机可读存储介质，其上存储有程序，其特征在于，所述程序被处理器执行时，实现如权利要求1-4中任一项所述的一种基于双光子钙成像数据的细胞胞体自动识别与分割方法。6. A computer-readable storage medium having a program stored thereon, wherein when the program is executed by a processor, the method for automatically identifying and segmenting cell bodies based on two-photon calcium imaging data according to any one of claims 1 to 4 is implemented.
CN202310722477.XA 2023-06-19 2023-06-19 Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data Active CN116681685B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310722477.XA CN116681685B (en) 2023-06-19 2023-06-19 Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310722477.XA CN116681685B (en) 2023-06-19 2023-06-19 Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data

Publications (2)

Publication Number Publication Date
CN116681685A CN116681685A (en) 2023-09-01
CN116681685B true CN116681685B (en) 2025-07-29

Family

ID=87779003

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310722477.XA Active CN116681685B (en) 2023-06-19 2023-06-19 Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data

Country Status (1)

Country Link
CN (1) CN116681685B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN120580564B (en) * 2025-08-01 2025-11-25 上海微谱检测科技集团股份有限公司 Training method, prediction method, device and equipment for two-photon image prediction model

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107346545A (en) * 2017-05-22 2017-11-14 沈阳工业大学 Improved confinement growing method for the segmentation of optic cup image
CN108154513A (en) * 2017-11-21 2018-06-12 中国人民解放军第三军医大学 Cell based on two photon imaging data detects automatically and dividing method
EP3836020B1 (en) * 2019-12-12 2024-08-28 National Cheng Kung University A method for determining cellular nuclear-to-cytoplasmic ratio

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
双光子钙成像数据自动化分析研究及应用;李珍;中国优秀硕士学位论文全文数据库基础科学辑;20250515;A006-580 *

Also Published As

Publication number Publication date
CN116681685A (en) 2023-09-01

Similar Documents

Publication Publication Date Title
CN113313234B (en) Neural network system and method for image segmentation
JP7760515B2 (en) Automated tumor detection based on image processing
Rupprecht et al. Deep active contours
US20250285265A1 (en) Systems and methods for detecting structures in 3d images
de Brebisson et al. Deep neural networks for anatomical brain segmentation
CN105574859B (en) A kind of liver neoplasm dividing method and device based on CT images
Su et al. Robust cell detection and segmentation in histopathological images using sparse reconstruction and stacked denoising autoencoders
CN108537102B (en) High-resolution SAR image classification method based on sparse features and conditional random field
CN106296653A (en) Brain CT image hemorrhagic areas dividing method based on semi-supervised learning and system
Goceri Intensity normalization in brain MR images using spatially varying distribution matching
CN108122221B (en) Segmentation method and device for cerebral ischemia area in diffusion weighted imaging image
CN108038513A (en) A kind of tagsort method of liver ultrasonic
WO2022127500A1 (en) Multiple neural networks-based mri image segmentation method and apparatus, and device
CN114494174A (en) Method and device for detecting chip bonding line defects
CN116681685B (en) Automatic cell body identification and segmentation method and device based on two-photon calcium imaging data
CN111080658A (en) Cervical MRI image segmentation method based on deformable registration and DCNN
WO2020140380A1 (en) Method and device for quickly dividing optical coherence tomography image
CN108520261A (en) A method and device for identifying the number of peanut kernels
CN116433976A (en) Image processing method, device, device and storage medium
Berger et al. An adaptive sampling scheme to efficiently train fully convolutional networks for semantic segmentation
CN120259690A (en) A microscopic cell image tracking method and device based on graph neural network
CN113705632A (en) Rectal cancer MRI image classification method and device, electronic equipment and storage medium
Valverde et al. Multiple sclerosis lesion detection and segmentation using a convolutional neural network of 3d patches
CN119417988A (en) Breast nodule three-dimensional reconstruction method and system
Mayerich et al. Fast cell detection in high-throughput imagery using GPU-accelerated machine learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant