TW202120906A - Artificial intelligence based cell detection method by using optical kinetics and system thereof - Google Patents

Abstract

An artificial intelligence based cell detection method by using optical kinetics includes acquiring a plurality of cells, sampling N images of the cells during a first time point and a second time point, determining an analytic region of each cell, executing an image processing process to the analytic region of the each cell according to the N images of the cells for generating a plurality of optical kinetics vector parameters, generating deformation vector information of the cells during the first time point and the second time point according to the plurality of optical kinetics vector parameters, acquiring at least one key deformation vector characteristic value of the cells according to the deformation vector information, and inputting the deformation vector information and the at least one key deformation vector characteristic value to a neural network for detecting qualities of the cells.

Description

Cell detection method and system using artificial intelligence of photodynamic technology

The present invention describes a cell detection method and system using artificial intelligence of photodynamic technology, especially a cell detection method and system with artificial intelligence capable of detecting cell quality and identifying cell functions.

With the rapid development of technology, women of childbearing age often suffer from infertility due to work pressure, eating habits, civilization diseases, abnormal ovulation function, hormone imbalance or some chronic diseases. Nowadays, infertility is a high-paying treatment course, which is in huge demand in Taiwan, China and the international market, and its demand is growing rapidly every year. Many women choose In Vitro Fertilization (IVF) to treat infertility. In vitro artificial insemination is to take out the eggs and sperm, perform in vitro fertilization under manual operation, and cultivate them into embryos, and then implant the embryos back into the mother's body. However, the success rate of existing infertility treatments is only 30%. The focus of infertility treatment is the choice of embryos. However, the existing methods for selecting embryos are mainly used by embryologists to use embryo photos or time-lapse video data to judge the quality of embryos in a subjective manner. In terms of current technology, due to the lack of a systematic and automated way of judging the quality of embryos, in the course of infertility treatment, the physician will subjectively select the implantation of embryos, and the implantation success rate is still low, so it is also current The bottleneck of infertility treatment.

In other words, under the current course of infertility treatment, doctors can only observe the condition of the embryo when it divides in a subjective way. For example, doctors will use a subjective way to classify embryos into multiple levels based on the number of cells in the embryo, the uniformity of cell division, and the degree of fragmentation during division. For example, embryos that divide evenly into even-numbered cells are better, while embryos that produce incomplete odd-numbered cell divisions and more fragments have poor growth potential. However, as mentioned above, since the current judgment of the quality of embryos is mainly based on the experience of doctors, the best embryos are selected by subjective judgment. Therefore, it is difficult to improve the success rate of the current technology in the treatment of infertility, and it is easy to be affected by the subjective opinions of different doctors (for example, misjudgment).

An embodiment of the present invention provides a cell detection method using artificial intelligence of photodynamic technology. The cell detection method using artificial intelligence of photodynamic technology includes obtaining a plurality of cells, sampling N images of the cells between the first time point and the second time point, and determining the analysis area of each cell, according to the For N images of these cells, the image processing procedure is performed on the analysis area of each cell to generate a plurality of photodynamic vector parameters. According to the photodynamic vector parameters, the cells at the first time point and the second time point are generated according to the photodynamic vector parameters. For the deformation vector information between time points, obtain at least one key deformation vector characteristic value of the cells according to the deformation vector information, and input the deformation vector information and at least one key deformation vector characteristic value into the neural network to train the neural network Network, and the use of neural networks to establish cell quality detection models with artificial intelligence procedures to detect cell quality and/or identify cells. The first time point is before the second time point, and N is a positive integer greater than 2.

Another embodiment of the present invention provides a cell detection system using artificial intelligence of photodynamic technology. The cell detection system using artificial intelligence of photodynamic technology includes a carrier, a lens module, an image capture device, a processor, and a memory. The carrier has a containing slot for placing a plurality of cells. The lens module faces the carrier for magnifying the details of the cells. The image capturing device faces the lens module and is used to obtain the images of the cells through the lens module. The processor is coupled to the lens module and the image capturing device for adjusting the magnification of the lens module and processing the images of the cells. After the cells are placed in the accommodating slot of the carrier, the processor controls the image capturing device to sample N images of the cells between the first time point and the second time point through the lens module, and the processor determines each For the analysis area of a cell, perform image processing on the analysis area of each cell based on the N images of the cells to generate a plurality of photodynamic vector parameters, and generate the cells according to the photodynamic vector parameters According to the deformation vector information between the first time point and the second time point, at least one key deformation vector characteristic value of the cells is obtained according to the deformation vector information. The processor contains a neural network-like. The deformation vector information and at least one key deformation vector feature value are used to train the neural network. The processor uses a neural network to establish a cell quality detection model with an artificial intelligence process to detect cell quality and/or identify cells. The first time point is before the second time point, and N is a positive integer greater than 2.

FIG. 1 is a block diagram of an embodiment of the cell detection system 100 using artificial intelligence of photodynamic technology according to the present invention. In order to simplify the description, the cell detection system 100 using artificial intelligence of photodynamic technology is hereinafter referred to as the "cell detection system 100". The cell detection system 100 includes a carrier 10, a lens module 11, an image capturing device 12, a processor 13 and a memory 14. The carrier 10 has a containing slot for placing a plurality of cells. For example, the carrier 10 may be a petri dish, and the containing tank inside may contain some culture liquid. Multiple cells can develop in the culture medium. The plurality of cells can be a plurality of germ cells, a plurality of embryos, or any cells that can be divided and to be observed. The lens module 11 faces the carrier 10 for magnifying the details of the cells. The lens module 11 can be any lens module with optical or digital zoom capabilities, such as a microscope module. The image capturing device 12 faces the lens module 11 to obtain images of the cells through the lens module 11. In the cell detection system 100, the image capturing device 12 can be a camera with a photosensitive element or a hyperspectrometer. In other words, the image capturing device 12 obtains the image of the cells through the lens module 11, which can be a grayscale image, a hyperspectral image, or a depth-of-field composite image. If the image capturing device 12 is a hyperspectrometer, the images of these cells can be: (A) a cell image corresponding to any wavelength in the hyperspectrometer image, (B) a gray scale synthesized by the cell images of each wavelength in the hyperspectral image Cell imaging. Any reasonable image format belongs to the scope disclosed by the present invention. The processor 13 is coupled to the lens module 11 and the image capturing device 12 for adjusting the magnification of the lens module 11 and processing the images of the cells. The processor 13 can be a central processing unit, a microprocessor, or any programmable processing unit. The processor 13 has a neural network, such as a deep neural network (Deep Neural Networks, DNN), and can perform machine learning and deep learning functions. Therefore, the neural network of the processor 13 can be trained and can be regarded as the processing core of artificial intelligence. The memory 14 is coupled to the processor 13 for storing training data and analysis data during image processing.

In the cell detection system 100, after the cells of the carrier 10 are placed, the processor 13 controls the image capturing device 12 to sample through the lens module 11 between the first time point and the second time point N images of these cells. Next, the processor 13 determines the analysis area of each cell, and performs image processing on the analysis area of each cell based on the N images of the cells to generate a plurality of photodynamic vector parameters. The processor 13 generates deformation vector information of the cells between the first time point and the second time point according to the photodynamic vector parameters, and obtains at least one key deformation vector feature of the cells according to the deformation vector information Numerical value. As mentioned above, the processor 13 includes a trainable neural network. Therefore, the deformation vector information and at least one key deformation vector feature value can be used to train the neural network. After the neural network-like training is completed, the processor 13 can use the neural network to establish a cell quality detection model with artificial intelligence procedures to detect cell quality and/or identify cells. In the cell detection system 100, the first time point is before the second time point, and N is a positive integer greater than 2. In other words, the cell detection system 100 can train the neural network with the time series of cell image information between two different time points. After the neural network-like training is completed, the cell detection system 100 has the cell detection function of artificial intelligence, and has the ability to automatically detect cell quality and/or identify cells. The details of how the cell detection system 100 trains a neural network to perform the cell detection function of artificial intelligence will be described later.

FIG. 2 is a flowchart of a cell detection method performed by the cell detection system 100 using artificial intelligence of photodynamic technology. The cell detection method may include step S201 to step S208. Any reasonable step changes or technical changes belong to the scope disclosed by the present invention. The contents of step S201 to step S208 are described as follows: Step S201: Obtain a plurality of cells; Step S202: Sample N images of the cells between the first time point and the second time point; Step S203: Determine the analysis area of each cell; Step S204: According to the N images of the cells, perform an image processing procedure on the analysis area of each cell to generate a plurality of photodynamic vector parameters; Step S205: Generate the deformation vector information of the cells between the first time point and the second time point according to the photodynamic vector parameters; Step S206: Obtaining at least one key deformation vector characteristic value of the cells according to the deformation vector information; Step S207: Input the deformation vector information and at least one key deformation vector characteristic value into the neural network-like to train the neural network-like; Step S208: A neural network is used to establish a cell quality detection model with artificial intelligence procedures to detect cell quality and/or identify cells.

In order to simplify the description, the following "cell" only uses "embryo" as an example for illustration. However, the present invention is not limited to this. The definition of cell can be germ cells, nerve cells, tissue cells, animal and plant cells, or anything that needs to be studied. And observed cells. In step S201, the researcher or medical staff may first obtain multiple embryos. In step S202, the processor 13 may control the image capturing device 12 to sample N images of the embryos between the first time point and the second time point. The image capturing device 12 can obtain N images between two different time points in any manner. For example, the image capturing device 12 can obtain an image collection composed of multiple frames (Frames) in a period of time (between the first time point and the second time point) by means of recording (such as 30fps or 60fps) . Alternatively, the image capturing device 12 may periodically take pictures of a plurality of embryos within a period of time (between the first time point and the second time point) to obtain N images of the embryos. In addition, the first time point and the second time point may be any two time points in the observation cycle when the embryos develop and divide in the culture medium. For example, the first time point and the second time point can be selected as the first day and the fifth day, respectively, to observe the development and division of the embryos. In other words, the image capturing device 12 can continuously take pictures of the embryos from the 0th hour to the 120th hour to obtain N images. In step S203, the processor 13 may determine the analysis area of each embryo. Medical staff or researchers can also manually determine the analysis area of each embryo. Here, the analysis area of each embryo can be selected as the area with the greatest difference between the two embryos, or the area with the greatest difference between the superior embryo and the inferior embryo. For example, the analysis area may be a blastocyst area or a custom area of each embryo in these embryos. Moreover, the range of the analysis area is smaller than the size of a single embryo.

Next, in step S204, the processor 13 may perform an image processing procedure on the analysis area of each embryo according to the N images of the embryos to generate a plurality of photodynamic vector parameters. For example, the processor 13 may perform Digital Image Correlation (DIC) analysis on the analysis area of each embryo based on the N images of the embryos. Digital Image Correlation (DIC) technology is a technology that analyzes the changes in the characteristics of an object in a time interval before and after the time point to calculate the shape of the object. In other words, under the DIC technology, the photodynamic vector parameters can be based on the changes in the deformation of the embryos in the M-1th image (front) and the Mth image (rear) in N images. Obtained, and M≤N. Moreover, in the cell detection system 100, the photodynamic vector parameters may include displacement vector information, strain vector information, and deformation speed vector information, as described below. In step S204, the processor 13 can perform DIC analysis on the analysis area of each embryo based on the N images of the embryos. After analyzing the N images using DIC technology, the processor 13 can obtain the vertical displacement information, horizontal displacement information, strain vector distribution information, and deformation speed of each embryo in the front and rear images (the M-1th image and the Mth image). Vector information. The vertical displacement information can be defined as the distance and coordinates of the vertical displacement of the embryo in the analysis area between the two images. The horizontal displacement information can be defined as the distance and coordinates of the horizontal displacement of the analysis area of the embryo between the two images. The strain vector distribution information can be defined as the embryo between two images, quantifying the extent of expansion or contraction of the analysis area. The deformation speed vector information can be derived from the displacement distance and the amount of strain divided by the time difference. For example, if the time difference between the M-1th image and the Mth image is t and the horizontal displacement distance is p, the value of the horizontal component of the deformation velocity vector is p/t. The time difference between the M-1th image and the Mth image is t, the vertical displacement distance is q, and the value of the vertical component of the deformation velocity vector is q/t. The time difference between the M-1th image and the Mth image is t, and the amount of strain is r, so the value of the strain component of the deformation velocity vector is r/t. Then, in step S205, the processor 13 can generate the deformation vector information of the embryos between the first time point and the second time point according to the photodynamic vector parameters. The deformation vector information can be generated by the aforementioned photodynamic vector parameters using linear or non-linear formulas.

Then, in step S206, the processor 13 can obtain at least one key deformation vector characteristic value of the embryos according to the deformation vector information. For example, the processor 13 may perform DIC analysis on the analysis area of each embryo based on the N images of the embryos, so as to obtain at least one local maximum value (Local Maximum) from the deformation vector information of the embryos. ) Of the dependent variable. Similarly, the processor 13 can perform DIC analysis on the analysis area of each embryo based on the N images of the embryos, so as to obtain at least one local minimum in the deformation vector information of the embryos. strain. In addition, the at least one key deformation vector characteristic value may include at least one local maximum strain and/or at least one local minimum strain. For ease of understanding, the strain at different time points is described in Table T1, as shown below. time 0-00 0-01 0-02 0-03 Maximum strain 0.0795 0.227 0.113 0.336 Minimum strain 0.002 -0.013 -0.006 -0.01 Strain distribution Strain Strain Embryo Behavior Link No split No split No split No split time 0-04 0-05 0-06 0-07 Maximum strain 0.104 0.147 0.061 0.138 Minimum strain -0.001 -0.016 -0.0065 -0.014 Strain distribution Embryo Behavior Link No split No split No split No split Table T1

According to Table T1, at time 0-01, the maximum strain is 0.227. 0.227 is significantly higher than the maximum strain 0.0795 at time 0-00 and the maximum strain 0.113 at time 0-02. Therefore, the maximum strain of 0.227 at the time point 0-01 can be defined as the strain of the local maximum (Local Maximum). Therefore, the strain distribution corresponding to time point 0-01 points to the note of "strain large". Similarly, at time 0-03, the maximum strain is 0.336. 0.336 is significantly higher than the maximum strain at time 0-02, 0.113, the maximum strain at time 0-04, and 0.104, and the maximum at time 0-05. The strain is 0.147, the maximum strain at time 0-06 is 0.061, and the maximum strain at time 0-07 is 0.138. Therefore, the maximum strain of 0.336 at time point 0-03 can be defined as the strain of the local maximum. Therefore, the strain distribution corresponding to time point 0-03 points to the note of "strain large". Similarly, the strain of at least one local minimum (Local Minimum) can also be obtained by statistics. The processor 13 can store the data of the table T1, the strain of at least one local maximum value, and the strain of at least one local minimum value in the memory 14.

As mentioned above, the processor 13 has a neural network, such as a deep neural network (DNN), which can perform machine learning and deep learning functions. Therefore, the neural network of the processor 13 can be trained. In order to train the neural network of the processor 13, in step S207, the cell detection system 100 can input the deformation vector information and at least one key deformation vector characteristic value to the neural network in the processor 13 to train the neural network. road. After the quasi-neural network is trained, according to step S208, the processor 13 can use the quasi-neural network to establish a cell quality detection model with artificial intelligence procedures to detect embryo quality and/or identify embryos. In other words, the cell detection system 100 using artificial intelligence to detect embryo quality and/or identify embryos may include two stages. The first stage is the training stage. The cell detection system 100 can input the entire data volume of the time series or the feature value of the reduced time data volume into the neural network, and can further input at least one key deformation vector feature value Into a neural network to build a cell quality detection model. The second stage is the detection stage of artificial intelligence. After the neural network-like training is completed, the processor 13 can use the trained neural network-like cell quality detection model to determine embryo quality and identify embryos. Therefore, the cell detection system 100 of the present invention can prevent medical personnel or researchers from judging the quality of embryos in a subjective manner, so that the pregnancy success rate can be greatly improved for the course of infertility treatment.

FIG. 3 is a schematic diagram of the cell detection system 100 using artificial intelligence of photodynamic technology, adding additional steps to enhance the accuracy of cell detection. In order to further enhance the accuracy of judging embryo quality and identifying embryos, the cell detection system 100 can also introduce morphological detection technology to enhance the neural network-like cell detection capability, as shown below. Step S201: Obtain a plurality of cells; Step S301: Detect the edge feature data of the cells between the first time point and the second time point using morphological edge detection technology, and input the edge feature data to the neural network; Step S302: Using the ellipse detection method of typology, according to the edge feature data, detect the division time point and the number of each cell between the first time point and the second time point of the cells, and compare each cell The split time point and the number of splits are input to the neural network to train the neural network.

Similarly, in order to simplify the description, the following "cell" only uses "embryo" as an example for illustration, but the present invention is not limited to this. The definition of cell can be germ cell, nerve cell, tissue cell, animal or plant cell or Any cell that needs to be studied and observed. In order to further strengthen the accuracy of judging embryo quality and identifying embryos, after the cell detection system 100 obtains a plurality of cells (embryos) in the aforementioned step S201, the processor 13 can use the edge detection technology of morphology to detect the The edge feature data of these embryos between the first time point and the second time point are input to the neural network. The edge feature data can be the outline of the entire embryo or a specific part (for example, a blastocyst), and the data format can be expressed in multiple coordinates. For example, on a two-dimensional plane, the outline of a single embryo can be a closed line segment, which can be represented by (X ₁ , Y ₁ ) to (X _L , Y _L ), where L is a positive integer and the larger the L, the higher the resolution. Then, in step S302, the processor 13 can use the morphological ellipse detection method to detect the division time of each embryo between the first time point and the second time point based on the edge feature data. The points and the number of splits, and the split time point and the number of splits of each embryo are input into the neural network to train the neural network. It is explained here that the processor 13 may preset at least one ellipse curve fitting function. After the aforementioned step S301 detects the coordinates (X ₁ , Y ₁ ) to (X _L , Y _L ) corresponding to at least one closed line segment, the processor 13 may use at least one elliptical curve fitting function to match its coordinates (X ₁ ,Y ₁ ) to (X _L ,Y _L ) whether it conforms to the elliptical closed curve. Therefore, the processor 13 can obtain the number of elliptical closed curves that are successfully fitted in the image at a certain point in time. The processor 13 may regard the number of successfully fitted elliptical closed curves as the number of divisions of the embryo. Therefore, based on the information of the N images at different time points, the processor 13 can detect the split time point and the number of splits of each embryo. In other words, compared to the cell detection method from step S201 to step S208 described in Figure 2, the cell detection system 100 can introduce additional steps S301 to S302 to obtain more information (such as edge feature data, each The split time point and number of an embryo) to train the neural network. Therefore, the training of the similar neural network will be more optimized, thereby increasing the accuracy of artificial intelligence in detecting the quality of cells.

FIG. 4 is a schematic diagram of the input data and output data of the neural network-like processor 13 in the cell detection system 100 using artificial intelligence of photodynamic technology. As mentioned above, the image capturing device 12 can take pictures of multiple embryos between two different time points to generate N images. The cell detection system 100 can use DIC technology to analyze N images to generate the photodynamic vector parameters. Moreover, the photodynamic vector parameters can be used to generate deformation vector information D1 and at least one key deformation vector characteristic value D2. The generated deformation vector information D1 and at least one key deformation vector characteristic value D2 can be used to train the neural network in the processor 13. After the neural network-like training is completed, the processor 13 can use artificial intelligence procedures to judge embryo quality and/or identify embryos. The processor 13 can output cell quality output data D6. It should be understood that each of the N images can be a two-dimensional image or a three-dimensional image. If each of the N images is a two-dimensional image, the photodynamic vector parameters, deformation vector information, and at least one key deformation vector characteristic value are in a K-dimensional data format. For example, at the time point T and the specific wavelength λ of the hyperspectrometer, the optical signal of the pixel S1 at the coordinate (x, y) can be expressed as S1(λ, T, x, y). The optical signal S1(λ, T, x, y) of the pixel S1 is a four-dimensional signal format. Similarly, if each of the N images is a three-dimensional image, the photodynamic vector parameters, the deformation vector information, and the at least one key deformation vector feature value are in the format of (K+1)-dimensional data . For example, at the time point T and the specific wavelength λ of the hyperspectrometer, the optical signal of the pixel S2 at the coordinates (x, y, z) can be expressed as S2(λ, T, x, y, z). The optical signal S2 (λ, T, x, y, z) of the pixel S2 is a signal format of five dimensions. K is a positive integer greater than 2. Therefore, it can be expected that when the data format of the cell detection system 100 has a higher dimension, the computational complexity will become higher. When the data format of the cell detection system 100 is of a lower dimension, the computational complexity will be lower.

And, as mentioned above, the neural network of the processor 13 can use edge feature data D3 and the split time point D5 of each embryo and the number of divisions D4 for training. The neural network can use artificial intelligence programs. Optimize the cell quality detection model. Therefore, as shown in Figure 4, the neural network in the processor can receive deformation vector information D1, at least one key deformation vector feature value D2, edge feature data D3, the number of splits D4, and the split time point D5. In addition, after the neural network training in the processor is completed, the processor can use artificial intelligence procedures to select the egg retrieval/embryo of infertile women to output cell quality output data D6. The cell quality output data D6 can be in any form of data format, such as outputting grading data of embryo quality, outputting the percentage ranking of at least one embryo, or outputting the cell quality of at least one embryo. The quality of a cell can be determined by the numerical content of the detailed chemical composition of the cell, the pros and cons of the genetic gene, the pros and cons of the cell's developmental state in a specific time, or whether the cell has a disease. In the case of germ cells, it can also be determined by pregnancy, newborn health, and gender.

In summary, the present invention describes a cell detection method and system using artificial intelligence of photodynamic technology. The application group of the cell detection system can be women with infertility. Medical staff can first establish an artificial intelligence cell quality detection model based on a large amount of cell data, and then they can treat women with infertility. In addition, the artificial intelligence-like neural network can receive various parameters related to photodynamics and morphology, such as deformation vector information, at least one key deformation vector feature value, edge feature data, number of splits, and split time points. Therefore, the use of similar neural networks can prevent medical staff or researchers from judging the quality of embryos in a subjective way. Women with infertility can culture multiple embryos first. The cell detection system uses artificial intelligence to determine the best embryos based on the inter-cell quality detection model, and then artificially implants them into the mother's uterus to increase the success rate of conception. The foregoing descriptions are only preferred embodiments of the present invention, and all equivalent changes and modifications made in accordance with the scope of the patent application of the present invention shall fall within the scope of the present invention.

100: Cell detection system using artificial intelligence of photodynamic technology 10: Vehicle 11: Lens module 12: Image capture device 13: processor 14: Memory S201 to S208: steps S301 to S302: steps D1: Deformation vector information D2: At least one key deformation vector eigenvalue D3: Edge feature data D4: Number of splits D5: Split time point D6: Cell quality output data

FIG. 1 is a block diagram of an embodiment of the cell detection system using artificial intelligence using photodynamic technology of the present invention. Figure 2 is a flow chart of the method of cell detection performed by the cell detection system using artificial intelligence based on photodynamic technology in Figure 1. Figure 3 is a schematic diagram of the artificial intelligence cell detection system using photodynamic technology in Figure 1 with additional steps to enhance the accuracy of cell detection. Figure 4 is a schematic diagram of the input data and output data of the neural network-like processor in the cell detection system using artificial intelligence using photodynamic technology in Figure 1.

100: Cell detection system using artificial intelligence of photodynamic technology

10: Vehicle

11: Lens module

12: Image capture device

13: processor

14: Memory

Claims (12)

A cell detection method using artificial intelligence of photodynamic technology, including: Obtain a plurality of cells; Sample N images of the cells between the first time point and the second time point; Determine an analysis area of each cell; According to the N images of the cells, an image processing procedure is performed on the analysis area of each cell to generate a plurality of photodynamic vector parameters; Generating the deformation vector information of the cells between the first time point and the second time point according to the photodynamic vector parameters; Obtaining at least one key deformation vector characteristic value of the cells according to the deformation vector information; Inputting the deformation vector information and the at least one key deformation vector characteristic value to a type of neural network to train the type of neural network; and Use this type of neural network to establish a cell quality detection model with an artificial intelligence process to detect cell quality and/or identify cells; The first time point is before the second time point, and N is a positive integer greater than 2. The method according to claim 1, wherein the cell lines are a plurality of germ cells, and the first time point and the second time point are during an observation cycle of the germ cells developing and dividing in the culture medium Any two points in time. The method according to claim 1, wherein the cell lines are a plurality of embryos, and the analysis region is a blastocyst region or a custom region of each embryo in the embryos, and the analysis region A range of is smaller than the size of a single cell. The method according to claim 1, wherein the image analysis is performed on the analysis area of each cell to generate the photodynamic vector parameters, which is to perform a digital image correlation on the analysis area of each cell (Digital Image Correlation, DIC) technology analysis, and the photodynamic vector parameters are obtained based on the changes in the deformation of the cells in the M-1 image and the M image in the N images, M≤N. The method according to claim 4, wherein the photodynamic vector parameters include displacement vector information, strain vector information, and deformation speed vector information. The method described in claim 1, which additionally includes: According to the N images of the cells, perform a Digital Image Correlation (DIC) analysis on the analysis area of each cell to obtain at least one part of the deformation vector information of the cells Local Maximum (Local Maximum) dependent variable; The at least one key deformation vector characteristic value includes the at least one local maximum value of the strain. The method described in claim 1, which additionally includes: According to the N images of the cells, the analysis area of each cell is analyzed by the digital image-related technology to obtain at least one local minimum in the deformation vector information of the cells strain; The at least one key deformation vector characteristic value includes the at least one local minimum value of the strain. The method according to claim 1, wherein each of the N images is a two-dimensional image or a three-dimensional image, if each of the N images is the two-dimensional image , The photodynamic vector parameters, the deformation vector information, and the at least one key deformation vector characteristic value are in a K-dimensional data format, and if each of the N images is the three-dimensional image, The photodynamic vector parameters, the deformation vector information, and the at least one key deformation vector characteristic value are in the format of (K+1)-dimensional data, and K is a positive integer greater than 2. The method according to claim 1, wherein each of the N images is a gray-scale image, a hyperspectral image, or a depth-of-field composite image. The method described in claim 1, which additionally includes: Use a typological edge detection technology to detect the edge feature data of the cells between the first time point and the second time point, and input the edge feature data into the neural network to Train this type of neural network; and Using an ellipse detection method of the morphology, according to the edge feature data, the cells are detected between the first time point and the second time point, a division time point and a division time point of each cell And input the division time point and the division number of each cell into the neural network to train the neural network. The method according to claim 10, wherein after the neural network uses the edge feature data and the split time point of each cell and the number of splits for training, the neural network uses the artificial intelligence program Optimize the cell quality detection model. A cell detection system using artificial intelligence of photodynamic technology, including: A carrier with a accommodating slot for placing a plurality of cells; A lens module facing the carrier for magnifying the details of the cells; An image capturing device facing the lens module for obtaining images of the cells through the lens module; A processor coupled to the lens module and the image capturing device for adjusting a magnification of the lens module and processing the images of the cells; and A memory, coupled to the processor, for storing training data and image processing analysis data; After the cells are placed in the accommodating slot of the carrier, the processor controls the image capturing device to sample N images of the cells between the first time point and the second time point through the lens module , The processor determines an analysis area of each cell, and performs an image processing procedure on the analysis area of each cell based on the N images of the cells to generate a plurality of photodynamic vector parameters, according to the light The dynamic vector parameter generates the deformation vector information of the cells between the first time point and the second time point, and obtains at least one key deformation vector characteristic value of the cells according to the deformation vector information, and the processing The device includes a type of neural network. The deformation vector information and the at least one key deformation vector feature value are used to train the type of neural network. The processor uses the type of neural network to establish a cell quality test by an artificial intelligence process. Model to detect cell quality and/or identify cells, the first time point is before the second time point, and N is a positive integer greater than 2.

Images