Organ-like quantitative experiment analysis method based on deep learning and related device
By employing a deep learning-based approach, utilizing high-throughput microscopy and ensemble learning algorithms for quantitative experimental analysis of organoids, the problem of low accuracy in manual assessment in existing technologies is solved, achieving efficient and accurate quantitative experimental analysis of organoids.
Patent Information
- Application Number
- CN202511067141.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-31
- Publication Date
- 2025-11-07
AI Technical Summary
Existing quantitative experiments on organoids lack a systematic analysis framework, and manual evaluation is inaccurate, time-consuming, and labor-intensive, making it difficult to meet the needs of large-scale experiments.
We employed a deep learning-based approach to acquire pore images using a high-throughput microscope, obtained organoid data, and conducted quantitative experiments. We then used an ensemble learning algorithm model for data mapping analysis.
It reduces human evaluation errors, improves the accuracy and efficiency of analysis, reduces experimental costs, and meets the needs of large-scale organoid experimental analysis.
Smart Images
Figure CN120912563A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of experimental analysis, and in particular to a method for quantitative analysis of organoids based on deep learning and related devices. BACKGROUND
[0002] Organoids belong to three-dimensional (3D) cell cultures, and contain some key characteristics of their representative organs, and have a wide range of applications in quantitative experiments such as pathological experiments, drug sensitivity experiments, DNA and RNA sequencing experiments, etc.
[0003] However, the growth of each organoid in the same culture well is different, and the size, number, volume, etc. are different, and the number of organoids cannot be simply used to evaluate the number of experiments available. In the existing quantitative experiments of organoids, the number of experiments available for quantitative experiments of organoids is generally predicted manually, and there is a lack of a systematic analysis system. This manual evaluation method is greatly affected by subjective factors, and different evaluators may obtain different results, resulting in low evaluation accuracy. Moreover, manual operation is time-consuming and laborious, and the efficiency is low, which is difficult to meet the demand of large-scale organoid experimental analysis. SUMMARY
[0004] The embodiments of the present application provide a method for quantitative analysis of organoids based on deep learning and related devices, which have improved the above problems.
[0005] To achieve the above purpose, the technical scheme adopted by the present application is as follows:
[0006] In a first aspect, the present application provides a method for quantitative analysis of organoids based on deep learning, comprising:
[0007] obtaining a well site image based on a high-throughput microscope;
[0008] obtaining organoid data of the culture well, wherein the organoid data includes the number of organoids, the number of organoids under different culture well diameters, and the volume ratio;
[0009] obtaining quantitative experimental data of the culture well, and counting the quantitative experimental data after the quantitative experiment in the culture well;
[0010] organizing the obtained organoid data and quantitative experimental data into a data set, and using an ensemble learning algorithm model to map the two groups of data to realize quantitative analysis of organoids.
[0011] In combination with the first aspect, in some embodiments, the well site image is obtained based on a high-throughput microscope, comprising:
[0012] The high-throughput microscope scans the culture well in a layered scanning manner to obtain the organoid image of the well site.
[0013] In combination with the first aspect, in some embodiments, organoid data of the culture wells is obtained, wherein the organoid data comprises the number of organoids, the number of organoids under different culture well diameters, and the volume ratio, including:
[0014] The diameter interval corresponding to the number of organoids under different culture well diameters comprises [20, 30 pm], [30, 50 pm], [50, 100 pm], [100, 200 pm], and more than 200 pm.
[0015] In combination with the first aspect, in some embodiments, quantitative experimental data of the culture wells is obtained, and after the quantitative experiment is performed in the culture wells, the quantitative experimental data is counted, wherein the quantitative experiment comprises a pathology experiment, a drug sensitivity experiment, and DNA and RNA sequencing experiments, and the quantitative experimental data comprises the number of times of the pathology experiment, the number of times of the drug sensitivity experiment, and the number of times of the DNA and RNA sequencing experiments.
[0016] In combination with the first aspect, in some embodiments, quantitative experimental data of the culture wells is obtained, and after the quantitative experiment is performed in the culture wells, the quantitative experimental data is counted, including:
[0017] Based on Y Organoid The quantitative experimental data is defined and satisfies:
[0018]
[0019] Wherein, X Organoid is the total input of the organoid data, X num is the total number of organoids, X V1num is the number of organoids in the diameter interval [20, 30 pm], X V2num is the number of organoids in the diameter interval [30, 50 pm], X V3num is the number of organoids in the diameter interval [50, 100 pm], X V4num is the number of organoids in the diameter interval [100, 200 pm], X V5num is the number of organoids in the diameter interval more than 200 pm, and X vp is the volume ratio of organoids in the well site.
[0020] In combination with the first aspect, in some embodiments, the obtained organoid data and quantitative experimental data are arranged into a data set, and an integrated learning algorithm model is used to map the two groups of data to realize organoid quantitative experiment analysis, and satisfy:
[0021]
[0022] Wherein, VotingRegressor(Decision Tree Xorganoid , Random ForestXorganoid ) is a voting regression model, implementing an ensemble of decision tree models and random forest models, Decision Tree Xorganoid is the output of the decision tree model under the total input of organoid data, Random Forest Xorganoid is the output of the random forest model under the total input of organoid data;
[0023] The total input of organoid data, quantitative experimental data and ensemble learning algorithm model are sent to a graphics processing unit (GPU) for further iterative training.
[0024] In combination with the first aspect, in some embodiments, the total input of organoid data, quantitative experimental data and ensemble learning algorithm model are sent to a graphics processing unit (GPU) for further iterative training, including:
[0025] When the number of iterations reaches the preset target number, the iteration is stopped.
[0026] Secondly, the present application proposes an organoid quantitative experiment analysis system based on deep learning, which is configured to:
[0027] Obtain the well site image based on a high-throughput microscope;
[0028] Obtain the organoid data of the culture well, wherein the organoid data includes the number of organoids, the number of organoids under different culture well diameters, and the volume ratio;
[0029] Obtain the quantitative experimental data of the culture well, and count the quantitative experimental data after the quantitative experiment in the culture well;
[0030] Organize the obtained organoid data and quantitative experimental data into a data set, and use an ensemble learning algorithm model to map the two groups of data to realize organoid quantitative experiment analysis.
[0031] In combination with the second aspect, in some embodiments, the system is configured to:
[0032] Obtain the well site image based on a high-throughput microscope, including:
[0033] The high-throughput microscope scans the culture well in a layered scanning manner to obtain the organoid image of the well site.
[0034] In combination with the second aspect, in some embodiments, the system is configured to:
[0035] Obtain the organoid data of the culture well, wherein the organoid data includes the number of organoids, the number of organoids under different culture well diameters, and the volume ratio, including:
[0036] The diameter intervals corresponding to the number of organoids in different culture well diameters include [20, 30 pm], [30, 50 pm], [50, 100 pm], [100, 200 pm], and more than 200 pm.
[0037] In combination with the second aspect, in some embodiments, the system is configured to:
[0038] Obtaining quantitative experimental data of the culture well, after the quantitative experiment in the culture well, the quantitative experimental data is counted, the quantitative experiment includes pathological experiment, drug sensitivity experiment, DNA and RNA sequencing experiment, and the quantitative experimental data includes the number of pathological experiments, the number of drug sensitivity experiments, and the number of DNA and RNA sequencing experiments.
[0039] In combination with the first aspect, in some embodiments, the quantitative experimental data of the culture well is obtained, after the quantitative experiment in the culture well, the quantitative experimental data is counted, including:
[0040] Based on Y Organoid The quantitative experimental data is defined and satisfies:
[0041]
[0042] Wherein, X Organoid is the total input of organoid data, X num is the total number of organoids, X V1num is the number of organoids in the diameter interval [20, 30 pm], X V2num is the number of organoids in the diameter interval [30, 50 pm], X V3num is the number of organoids in the diameter interval [50, 100 pm], X V4num is the number of organoids in the diameter interval [100, 200 pm], X V5num is the number of organoids in the diameter interval more than 200 pm, and X vp is the volume ratio of organoids in the well site.
[0043] In combination with the first aspect, in some embodiments, the obtained organoid data and quantitative experimental data are sorted into a data set, and the two groups of data are mapped by using an integrated learning algorithm model to realize quantitative experimental analysis of organoids, and satisfy:
[0044]
[0045] Wherein, VotingRegressor(Decision Tree Xorganoid , Random Forest Xorganoid ) is a voting regression model, which realizes the collection of decision tree model and random forest model, Decision Tree XorganoidRandom Forest Xorganoid is the output of a random forest model under the total input of organoid data;
[0046] The total input of organoid data, quantitative experimental data and an ensemble learning algorithm model are input into a graphics processing unit (GPU) for iterative training again.
[0047] In combination with the second aspect, in some embodiments, the system is configured to:
[0048] The total input of organoid data, quantitative experimental data and an ensemble learning algorithm model are input into a graphics processing unit (GPU) for iterative training again, including:
[0049] The iteration is stopped when the number of iterations reaches a preset target number.
[0050] The third aspect of the embodiment of the application provides an electronic device, which comprises:
[0051] at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method provided in the first aspect of the embodiment of the application.
[0052] The fourth aspect of the embodiment of the application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the method provided in the first aspect of the embodiment of the application.
[0053] In summary, the above method and system have the following technical effects:
[0054] The method for analyzing quantitative experiments of organoids based on deep learning provided in the application first acquires well site images based on a high-throughput microscope, then acquires organoid data of culture wells, then acquires quantitative experimental data of the culture wells, then counts the quantitative experimental data after quantitative experiments are performed on the culture wells, and finally organizes the acquired organoid data and quantitative experimental data into a data set, maps the two groups of data by using an ensemble learning algorithm model, so as to realize quantitative experiment analysis of organoids. The method for analyzing quantitative experiments of organoids based on deep learning provided in the application takes organoid image analysis results as data, combines actual quantitative experimental results, avoids errors in manual evaluation and analysis errors caused by differences in the number and size of organoids, and is conducive to reducing experimental costs and better realizing quantitative experiment requirements of organoids. BRIEF DESCRIPTION OF DRAWINGS
[0055] Figure 1 FIG. 1 is a flowchart of a method for analyzing quantitative experiments of organoids based on deep learning provided in the embodiment of the application. DETAILED DESCRIPTION
[0056] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the protection scope of the present application.
[0057] The present application provides a kind of organoid quantitative experimental analysis method based on deep learning, please refer to Figure 1 , comprising the following steps:
[0058] S101: based on high-throughput microscope obtains well site image.
[0059] It can be understood that high-throughput microscope can be scanned in a layered scanning manner to obtain the organoid image of the well site. High-throughput microscope can perform detailed scanning operation on the culture well by layer-by-layer scanning. This scanning method not only can accurately cover each layer of each culture well, but also can effectively capture detailed images of organoids in the well site.
[0060] S102: obtain organoid data of culture well, wherein the organoid data includes the total number of organoids, the number of organoids under different culture well diameters, and the volume ratio.
[0061] Specifically, these organoid data cover detailed information in multiple aspects: first, the total number of organoids is recorded; second, the number of organoids under each specific diameter condition is counted for different diameters of culture wells; and finally, the volume ratio data of organoids are included to comprehensively analyze and evaluate the growth conditions and characteristics of organoids under different culture conditions.
[0062] In the present embodiment, the diameter intervals corresponding to the number of organoids under different culture well diameters include [20, 30 μm], [30, 50 μm], [50, 100 μm], [100, 200 μm] and more than 200 μm.
[0063] S103: obtain quantitative experimental data of culture well, and count the quantitative experimental data after quantitative experiment in the culture well.
[0064] It can be understood that quantitative experiment includes pathological experiment, drug sensitivity experiment, DNA and RNA sequencing experiment, and quantitative experimental data includes the number of pathological experiments, the number of drug sensitivity experiments, and the number of DNA and RNA sequencing experiments.
[0065] These data not only record the number of experiments of various types, but also reflect in detail the response of organoids to various experimental treatments under different culture conditions. By comparing the performance of organoids in different diameter culture wells in different experiments, researchers can deeply analyze the growth characteristics, pathological reactions, drug sensitivity, and changes in genetic material of organoids, thereby providing strong data support for disease research, drug screening, and personalized medicine. In addition, these data can also be used to train deep learning models to further improve the accuracy and efficiency of quantitative experimental analysis of organoids.
[0066] Specifically, as an embodiment, Y Organoid The quantitative experimental data is defined and satisfies:
[0067]
[0068] where X Organoid is the total input of organoid data, X num is the total number of organoids, X V1num is the number of organoids in the diameter interval [20, 30 μm], X V2num is the number of organoids in the diameter interval [30, 50 μm], X V3num is the number of organoids in the diameter interval [50, 100 μm], X V4num is the number of organoids in the diameter interval [100, 200 μm], X V5num is the number of organoids in the diameter interval exceeding 200 μm, and X vp is the volume proportion of organoids in the well site.
[0069] S104: Organoid data obtained and quantitative experimental data are sorted into a data set, and integrated learning algorithm model is used to map the two groups of data to realize quantitative experimental analysis of organoids.
[0070] It can be understood that the obtained organoid related data and the experimental data obtained by quantitative experimental means are systematically sorted and summarized to form a structured data set. On this basis, advanced integrated learning algorithm model is used to accurately map and correlate the two groups of data, aiming to effectively realize the in-depth analysis and research of organoids at the quantitative experimental level, thereby providing strong data support and analysis tools for scientific exploration in related fields.
[0071] Specifically, it satisfies:
[0072]
[0073] where VotingRegressor(Decision Tree Xorganoid , Random ForestXorganoid ) is a voting regression model, implementing an ensemble of decision tree models and random forest models, Decision Tree Xorganoid is the output of the decision tree model under the total input of organoid data, Random Forest Xorganoid is the output of the random forest model under the total input of organoid data;
[0074] The total input of organoid data, quantitative experimental data and ensemble learning algorithm model are sent to a graphics processing unit (GPU) for iterative training again. The iteration is stopped when the number of iterations reaches a preset target number, until the number of iterations reaches the preset target number, at which time the iteration process is automatically stopped, to ensure that the training effect of the model reaches the expected standard.
[0075] The application provides an organoid quantitative experiment analysis method based on deep learning. First, the well site image is obtained based on a high-throughput microscope. Then, the organoid data of the culture well is obtained. Then, the quantitative experimental data of the culture well is obtained. After the quantitative experiment in the culture well, the quantitative experimental data is counted. Finally, the obtained organoid data and quantitative experimental data are sorted into a data set. The two groups of data are mapped using an ensemble learning algorithm model to realize organoid quantitative experiment analysis. The organoid quantitative experiment analysis method based on deep learning provided by the application uses organoid image analysis results as data, combines actual quantitative experimental results, avoids errors in manual evaluation, and reduces analysis errors caused by differences in the number and size of organoids, which is beneficial to reducing experimental costs and better realizing the quantitative experiment demand of organoids.
[0076] Based on the same inventive concept, the embodiments of the application also provide an organoid quantitative experiment analysis system based on deep learning. The system is configured to:
[0077] Obtain the well site image based on the high-throughput microscope;
[0078] Obtain the organoid data of the culture well, wherein the organoid data includes the total number of organoids, the number of organoids under different culture well diameters, and the volume ratio;
[0079] Obtain the quantitative experimental data of the culture well, and count the quantitative experimental data after the quantitative experiment in the culture well;
[0080] Sort the obtained organoid data and quantitative experimental data into a data set, and map the two groups of data using an ensemble learning algorithm model to realize organoid quantitative experiment analysis.
[0081] In combination with the second aspect, in some embodiments, the system is configured to:
[0082] Obtain the well site image based on the high-throughput microscope, including:
[0083] The high-throughput microscope scans the culture wells in a layered scanning manner to obtain organoid images of the well positions.
[0084] In combination with the second aspect, in some embodiments, the system is configured to:
[0085] Obtain organoid data of the culture wells, wherein the organoid data includes the number of organoids, the number of organoids under different culture well diameters, and the volume ratio, including:
[0086] The diameter intervals corresponding to the number of organoids under different culture well diameters include [20, 30 μm], [30, 50 μm], [50, 100 μm], [100, 200 μm], and more than 200 μm.
[0087] In combination with the second aspect, in some embodiments, the system is configured to:
[0088] Obtain quantitative experimental data of the culture wells, and after the quantitative experiment is performed on the culture wells, the quantitative experimental data is counted, wherein the quantitative experiment includes a pathology experiment, a drug sensitivity experiment, and DNA and RNA sequencing experiments, and the quantitative experimental data includes the number of times of the pathology experiment, the number of times of the drug sensitivity experiment, and the number of times of the DNA and RNA sequencing experiments.
[0089] In combination with the first aspect, in some embodiments, the quantitative experimental data of the culture wells is obtained, and after the quantitative experiment is performed on the culture wells, the quantitative experimental data is counted, including:
[0090] Based on Y Organoid The quantitative experimental data is defined and satisfies:
[0091]
[0092] Wherein, X Organoid is the total input of the organoid data, X num is the total number of organoids, X V1num is the number of organoids in the diameter interval [20, 30 μm], X V2num is the number of organoids in the diameter interval [30, 50 μm], X V3num is the number of organoids in the diameter interval [50, 100 μm], X V4num is the number of organoids in the diameter interval [100, 200 μm], X V5num is the number of organoids in the diameter interval more than 200 μm, and X vp is the volume ratio of organoids in the well position.
[0093] In combination with the first aspect, in some embodiments, the obtained organoid data and quantitative experimental data are arranged into a data set, and the two groups of data are mapped by using an ensemble learning algorithm model to realize quantitative experimental analysis of the organoid and satisfy:
[0094]
[0095] wherein the VotingRegressor(Decision Tree Xorganoid , Random Forest Xorganoid ) is a voting regression model, a combination of a decision tree model and a random forest model, the Decision Tree Xorganoid is the output of the decision tree model under the total input of the organoid data, and the Random ForestXorganoid is the output of the random forest model under the total input of the organoid data.
[0096] The total input of the organoid data, the quantitative experimental data, and the ensemble learning algorithm model are sent to a graphics processing unit (GPU) for iterative training again.
[0097] In combination with the second aspect, in some embodiments, the system is configured to:
[0098] The total input of the organoid data, the quantitative experimental data, and the ensemble learning algorithm model are sent to a graphics processing unit (GPU) for iterative training again, including:
[0099] When the number of iterations reaches a preset target number, the iteration is stopped.
[0100] The organoid quantitative experimental analysis system based on deep learning provided in the present application first obtains a well site image based on a high-throughput microscope, then obtains organoid data of a culture well, then obtains quantitative experimental data of the culture well, then counts the quantitative experimental data after quantitative experiments are performed on the culture well, and finally arranges the obtained organoid data and quantitative experimental data into a data set, and maps the two groups of data by using an ensemble learning algorithm model to realize quantitative experimental analysis of the organoid. The organoid quantitative experimental analysis system based on deep learning provided in the present application takes the organoid image analysis result as data, combines the actual quantitative experimental result, avoids errors in manual evaluation and analysis errors caused by differences in the number and size of organoids, is conducive to reducing experimental costs, and better realizes the quantitative experimental needs of organoids.
[0101] Based on the same inventive concept, the embodiments of the present application also provide an electronic device, which comprises:
[0102] At least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the deep learning-based organoid quantitative experiment analysis method of the embodiments of the present application.
[0103] The electronic device provided in the present application first acquires a well site image based on a high-throughput microscope, then acquires organoid data of a culture well, then obtains quantitative experiment data of the culture well, after quantitative experiments are performed on the culture well, the quantitative experiment data is counted, finally the acquired organoid data and quantitative experiment data are sorted into a data set, and the two groups of data are mapped by using an integrated learning algorithm model, so as to realize organoid quantitative experiment analysis. The electronic device provided in the present application takes the organoid image analysis result as data, combines the actual quantitative experiment result, avoids the error of manual evaluation and the analysis error caused by the difference in the number and size of organoids, is beneficial to reducing the experimental cost, and better realizes the quantitative experiment demand of organoids.
[0104] In addition, to achieve the above-mentioned purpose, the embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the deep learning-based organoid quantitative experiment analysis method of the embodiments of the present application.
[0105] The various constituent components of the electronic device will be specifically introduced as follows:
[0106] The processor is the control center of the electronic device, and can be one processor or a plurality of processing elements. For example, the processor is one or more central processing units (CPU), can be an application specific integrated circuit (ASIC), or be configured as one or more integrated circuits to implement the embodiments of the present application, such as one or more microprocessors (digital signal processors, DSP), or one or more field programmable gate arrays (FPGA).
[0107] Optionally, the processor can execute various functions of the electronic device by running or executing a software program stored in the memory and calling data stored in the memory.
[0108] The memory is used to store a software program for executing the scheme of the present application, and is controlled by the processor to execute, and the specific implementation manner can refer to the above-mentioned method embodiments, which will not be described here again.
[0109] Optionally, the memory can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, and can be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory can be integrated with the processor or exist independently and be coupled to the processor through the interface circuit of the electronic device, and the embodiments of the present application do not make a specific limitation in this regard.
[0110] The transceiver is configured to communicate with the network device or the terminal device.
[0111] Optionally, the transceiver can include a receiver and a transmitter. The receiver is configured to implement the receiving function, and the transmitter is configured to implement the transmitting function.
[0112] Optionally, the transceiver can be integrated with the processor or exist independently and be coupled to the processor through the interface circuit of the router, and the embodiments of the present application do not make a specific limitation in this regard.
[0113] In addition, the technical effects of the electronic device can refer to the technical effects of the data transmission method of the above-mentioned method embodiments, and will not be repeated here.
[0114] It should be understood that the processor in the embodiments of the present application can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.
[0115] It should also be understood that the memory in the embodiments of the present application can be volatile or nonvolatile memory, or can include both volatile and nonvolatile memory. The nonvolatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory, among others. The volatile memory can be random access memory (RAM), which is used as external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM), among others.
[0116] The above-described embodiments can be implemented in part or in whole through software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When loaded and executed by a computer, the computer instructions or computer programs can generate the flow or function according to the embodiments of the present application in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, such as from one website site, computer, server, or data center to another website site, computer, server, or data center through a wired (e.g., infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing a set of one or more available media. The available media can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.
[0117] It should be understood that the term "and / or" used herein is merely an association relationship between the associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents an "or" relationship between the associated objects before and after it, but it can also represent an "and / or" relationship, which can be understood in the context before and after it.
[0118] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of the items, including any combination of single or multiple items. For example, at least one of a, b, or c can mean a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.
[0119] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0120] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on specific applications and design constraints. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.
Claims
1. A deep learning-based organoid quantitative experiment analysis method, characterized by, The method comprises: acquiring well site images based on a high-throughput microscope; acquiring organoid data of the culture wells, wherein the organoid data comprises a total number of organoids, a number of organoids under different diameters of the culture wells, and a volume ratio; obtaining quantitative experimental data of the culture wells, and counting the quantitative experimental data after quantitative experiments are performed on the culture wells; organizing the acquired organoid data and quantitative experimental data into a data set, and mapping the two groups of data by using an ensemble learning algorithm model to realize quantitative experimental analysis of organoids.
2. The deep learning-based organoid quantitative experimental analysis method according to claim 1, characterized in that, The method of acquiring well site images based on a high-throughput microscope comprises: scanning the culture wells in a layered scanning manner by using the high-throughput microscope to acquire organoid images of the well sites. 3.The deep learning-based organoid quantitative experimental analysis method of claim 1, wherein, The method of acquiring organoid data of the culture wells comprises: The diameter intervals corresponding to the number of organoids under different diameters of the culture wells include [20, 30 μm], [30, 50 μm], [50, 100 μm], [100, 200 μm], and more than 200 μm.
4. The deep learning-based organoid quantitative experimental analysis method according to claim 1, characterized in that, The method of obtaining quantitative experimental data of the culture wells comprises:
5. The deep learning-based organoid quantitative experimental analysis method according to claim 4, characterized in that, The quantitative experiments include pathological experiments, drug sensitivity experiments, and DNA and RNA sequencing experiments, and the quantitative experimental data includes the number of times of pathological experiments, the number of times of drug sensitivity experiments, and the number of times of DNA and RNA sequencing experiments. Based on Y Organoid The quantitative experimental data are defined and satisfy: wherein X Organoid is the total input of the organoid data, X num is the total number of the organoids, X V1num is the number of the organoids in the diameter interval [20, 30 pm], X V2num is the number of the organoids in the diameter interval [30, 50 pm], X V3num is the number of the organoids in the diameter interval [50, 100 pm], X V4num is the number of the organoids in the diameter interval [100, 200 pm], X V5num is the number of the organoids in the diameter interval over 200 pm, X vp is the volume fraction of the organoids in the well.
6. The deep learning-based organoid quantitative experimental analysis method according to claim 1, characterized in that, The method of obtaining quantitative experimental data of the culture wells comprises: wherein VotingRegressor(Decision Tree Xorganoid , Random Forest Xorganoid ) is a voting regressor model, implementing an ensemble of a decision tree model and a random forest model, Decision Tree Xorganoid is the output of the decision tree model under the total input of the organoid data, and Random Forest Xorganoid is the output of the random forest model under the total input of the organoid data; organizing the acquired organoid data and quantitative experimental data into a data set, and mapping the two groups of data by using an ensemble learning algorithm model to realize quantitative experimental analysis of organoids.
7. The deep learning-based organoid quantitative experimental analysis method according to claim 6, characterized in that, The method of sending the total input of the organoid data, the quantitative experimental data, and the ensemble learning algorithm model into a graphics processing unit (GPU) for iterative training again comprises: stopping iteration when the number of iterations reaches a preset target number.
8. A deep learning-based organoid quantitative experiment analysis system, characterized by, The system is configured to: acquire well site images based on a high-throughput microscope; acquire organoid data of the culture wells, wherein the organoid data comprises a number of organoids, a number of organoids under different diameters of the culture wells, and a volume ratio; obtain quantitative experimental data of the culture wells, and count the quantitative experimental data after quantitative experiments are performed on the culture wells; organize the acquired organoid data and quantitative experimental data into a data set, and map the two groups of data by using an ensemble learning algorithm model to realize quantitative experimental analysis of organoids.
9. An electronic device, comprising: The system comprises: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method as claimed in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, A computer program product comprising a computer readable medium having stored thereon a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 7.