An organoid abnormality monitoring system based on big data

Through the combination of big data and convolutional neural networks, an organoid abnormality monitoring system is built, which solves the challenge of abnormality monitoring in in vitro culture, real-time monitoring and resource allocation are achieved, and the safety and efficiency of organoid transplantation are improved.

CN118658614BActive Publication Date: 2024-11-19SHANDONG PROVINCIAL HOSPITAL AFFILIATED TO SHANDONG FIRST MEDICAL UNIVERSITY (SHANDONG PROVINCIAL HOSPITAL)
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411146784.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-19
Estimated Expiration
2044-08-21

AI Technical Summary

Technical Problem

Organoids cannot fully simulate the development and maturation process of organoids in vitro culture, resulting in challenging abnormality monitoring and difficult to achieve real-time and effective abnormality detection and resource allocation in the prior art.

Method used

The organoid abnormality monitoring system based on big data collects morphological, functional and gene expression information, builds an abnormality evaluation model, combines a convolutional neural network for real-time monitoring, and obtains an abnormal impact assessment index to ensure timely discovery and handling of abnormal situations.

Benefits of technology

Real-time monitoring of organoid abnormalities is realized, medical resources are allocated reasonably, abnormal situations are discovered and handled in a timely manner, and safety and efficiency after organoid transplantation are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118658614B_ABST
    Figure CN118658614B_ABST
Patent Text Reader

Abstract

The present invention discloses an organoid abnormality monitoring system based on big data, which specifically relates to the field of data monitoring technology, including a data acquisition module, an organoid abnormality assessment module, an abnormal impact assessment module and an algorithm processing module. The data acquisition module monitors the target organoid, collects morphological abnormality information, functional abnormality information and gene expression information as well as cell level information, clinical impact information and cost information and real-time data after transplantation. The present invention is based on analyzing and evaluating the morphological abnormality information, functional abnormality information and gene expression information of the target organoid, evaluating the abnormality degree of the target organoid, and when the abnormality degree is judged to be high, constructing an abnormal impact assessment model, obtaining an abnormal impact assessment index on the patient after organoid transplantation, and then constructing a convolutional neural network model, thereby realizing real-time monitoring of organoid abnormalities, rationally allocating medical resources, and ensuring timely discovery and handling of abnormal situations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data monitoring technology, and more specifically, to an organoid abnormality monitoring system based on big data. Background Art

[0002] Organoids are structures that spontaneously form under three-dimensional culture conditions in vitro from stem cells or induced pluripotent stem cells (iPSCs) and have functional and structural characteristics similar to those of organs in vivo. Organoid abnormality monitoring involves observing and analyzing these in vitro cultured organoids to detect abnormalities in their growth, development or function. This technology is of great significance in fields such as disease research, drug screening and regenerative medicine.

[0003] As a frontier field of biomedical research, organoid technology is developing rapidly and showing great potential. Through continuous optimization and innovation, organoids have achieved the successful cultivation of a variety of organoids, and will play an increasingly important role in disease research, drug screening, personalized medicine and regenerative medicine. With the maturity of technology and the expansion of applications, organoids are expected to bring revolutionary changes in future medical research and clinical applications.

[0004] In addition, although organoids can simulate the basic structure of some organs, they still cannot achieve the complex multicellular structure and function of organs in the body, and the lack of a complete vascular network limits their nutrient supply and waste elimination, making it impossible for organoids to fully simulate the development and maturation process of organs in the body during in vitro culture. The existence of these problems makes the abnormal monitoring of organoids quite valuable. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an organoid abnormality monitoring system based on big data, which analyzes and evaluates the morphological abnormality information, functional abnormality information and gene expression information of the target organoid to evaluate the abnormality degree of the target organoid. When the abnormality degree is judged to be high, an abnormal impact assessment model is constructed to obtain the abnormal impact assessment index of the patient after organoid transplantation, and then a convolutional neural network model is constructed, thereby realizing real-time monitoring of organoid abnormalities, rationally allocating medical resources, and ensuring timely detection and handling of abnormal situations.

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

[0007] A big data-based organoid abnormality monitoring system, including the following modules:

[0008] The data collection module monitors the target organoids and collects information on morphological abnormalities, functional abnormalities, gene expression, cell-level information, clinical impact information, cost information, and real-time data after transplantation;

[0009] The organoid abnormality assessment module builds an organoid abnormality assessment model to assess the degree of abnormality of the target organoid;

[0010] Abnormal impact assessment module: For areas with high abnormality in organoids, an abnormal impact assessment model is constructed by combining cell-level information, clinical impact information, and cost information to obtain the abnormal impact index of the target organoid on the patient;

[0011] The algorithm processing module establishes a convolutional neural network model using the organoid abnormality assessment index, abnormal impact assessment index and real-time data after transplantation as factors, monitors the convolutional neural network model for organoid abnormalities in real time, and decides whether to issue an alarm based on the monitoring results.

[0012] Specifically, in the data acquisition module, the ultimate goal of organoid research is to provide an important scientific basis and technical support for more effective and safer human organ transplantation. In actual application, due to the suddenness of the patient's disease and rejection reaction and considering the medical cost, it is difficult to always cultivate completely abnormal organoids. Therefore, it is a feasible and practical method to quickly screen out organoids with low abnormality and low adverse effects on patients, and build an abnormal monitoring system according to demand. In actual use, due to factors such as equipment cost and the complexity of real-time monitoring, the present invention is based on the long-term normal operation of organoid cultivation and the relative stability of the human body environment after transplantation, and the premise of continuously monitoring the real-time data of organoids, which may lead to the phenomenon of risk changes at a certain monitoring moment.

[0013] In the organoid abnormality assessment module, an organoid abnormality assessment model is constructed. The specific process is to collect the morphological abnormality information, functional abnormality information and gene expression information of the target organoid. The morphological abnormality information includes the morphological abnormality coefficient, which is calibrated as , the dysfunction information includes the dysfunction coefficient, calibrated as , gene expression information includes gene expression coefficients, calibrated as ;

[0014] Morphological abnormality coefficient By collecting the perimeter and area of ​​the target organoid, they are marked as and , then the circularity of the target organoid is , get the average roundness of the same type of organoids in the database and calibrate it as , then the morphological abnormality coefficient ;

[0015] Functional abnormality coefficient By collecting the secretions corresponding to the target organoid, including hormones, enzymes or metabolites, the secretion concentration in the organoid is calibrated as , obtain the concentration of the same type of secretion of the same type of organoid in the database and mark it as , x is the organoid number in the database, x=1, 2, 3, 4, ..., u, u is a positive integer, then the dysfunction coefficient ;

[0016] Gene expression coefficient The expression profile of genes in organoids was analyzed by qPCR, RNA sequencing or single-cell RNA sequencing technology. The gene overlap was obtained by comparing the gene expression of the same type of organoids in the database and calibrated as The expression level and modification status of proteins in organoids were detected by Western Blot, ELISA or mass spectrometry. The protein expression coincidence was obtained by comparing the expression level and modification status of proteins of the same type in the database, which was calibrated as , then the gene expression coefficient ;

[0017] The organoid abnormality assessment model is constructed by weighting the morphological abnormality information, functional abnormality information and gene expression information of the target organoid to generate the organoid abnormality assessment index. , and the corresponding coefficients are morphological anomaly coefficients , dysfunction coefficient , gene expression coefficient , the formula is ;

[0018] at the same time, All are greater than 0 and are set according to actual conditions. For example, the expert weighting method is adopted, that is, experts in related fields are invited to determine the weights of various indicators through professional opinion surveys and comprehensive evaluations to ensure that the weight coefficients can accurately reflect the importance of various indicators in the evaluation of organoid abnormalities. In addition, various methods such as hierarchical analysis method and fuzzy comprehensive evaluation method can also be considered to determine the weight coefficients to ensure the objectivity and scientificity of the weight coefficients. I will not go into details here.

[0019] In the organoid abnormality assessment module, the organoid abnormality assessment index obtained from the organoid abnormality assessment model is used to reflect the degree of abnormality of the target organoid. The larger its value is, the greater the difference between the target organoid and the normal organoid in morphology, the higher the degree of functional abnormality, the rarer the expression of genes and proteins, and the higher the overall degree of abnormality.

[0020] In the Organoid Abnormality Assessment Module, when the Organoid Abnormality Assessment Index When it is greater than the set abnormal threshold, it means that the target organoid is highly abnormal and has a large difference from the normal organoid, and a high abnormal data signal is output for subsequent evaluation;

[0021] When organoid abnormality is assessed When the value is less than or equal to the set abnormality threshold, it indicates that the target organoid has a low degree of abnormality and is not much different from normal organoids. The evaluation result is directly output without further steps.

[0022] In the abnormal impact assessment module, after receiving the high abnormal data signal, the abnormal impact assessment model is constructed. The specific process is to collect the cell level information, clinical impact information and cost information of the target organoid. The cell level information includes the cell level coefficient, which is calibrated as , clinical impact information includes clinical impact coefficient, calibrated as , the cost information includes the cost coefficient, denoted as ;

[0023] Cell level coefficient The number of cells surviving within one week of transplantation into the human body is measured as , obtain the number of cells surviving in the abnormal organoids of human transplantation target one week before, calibrated as , then the cell level coefficient ;

[0024] Clinical impact factor The time of normal life of the patient after the target organoid transplantation is collected and calibrated as , to obtain the normal life time of patients with the same disease who did not undergo organoid transplantation, calibrated as , w is the patient number who did not undergo organoid transplantation, w=1, 2, 3, 4, ..., v, v is a positive integer, then the clinical impact coefficient ;

[0025] Cost factor By collecting the target organoid culture and transplantation costs, the , the average annual income of the patient's family is calculated and calibrated as , then the cost coefficient ;

[0026] The abnormal impact assessment model is constructed by weighting the cell-level information, clinical impact information and cost information of the target organoid to generate the abnormal impact assessment index , and the corresponding coefficients are the cell level coefficients , Clinical Impact Factor , cost coefficient , the formula is ;

[0027] at the same time, All are greater than 0 and are set according to actual conditions. For example, the expert weighting method is adopted, that is, experts in related fields are invited to determine the weights of various indicators through professional opinion surveys and comprehensive evaluations to ensure that the weight coefficients can accurately reflect the importance of each indicator in the evaluation of organoid abnormalities. In addition, various methods such as hierarchical analysis method and fuzzy comprehensive evaluation method can be considered to determine the weight coefficients to ensure the objectivity and scientificity of the weight coefficients. I will not go into details here.

[0028] In the abnormal impact assessment module, the abnormal impact assessment index obtained from the abnormal impact assessment model is used to reflect the impact caused by the abnormality of the target organoid. The larger its value is, the lower the survival rate of the target organoid at the cellular level after transplantation, the less time the patient can live a normal life after transplantation, the greater the burden on the patient's family, and the greater the adverse impact.

[0029] In the abnormal impact assessment module, when the abnormal impact assessment index When it is greater than the set impact threshold, it indicates that the target organoid has a high degree of adverse impact on the patient, and a detection data signal is output. It is recommended to equip the organoid monitoring system after transplantation and conduct subsequent evaluation;

[0030] When abnormal impact assessment index When the value is less than or equal to the set impact threshold, it indicates that the impact of the target organoid is low, and the evaluation result is directly output without further steps.

[0031] In the algorithm processing module, the organoid abnormality assessment index, abnormal impact assessment index and organoid real-time data are used as construction factors of the convolutional neural network model. A convolutional neural network model is established to enable users to achieve real-time monitoring of organoid abnormalities after transplantation.

[0032] Furthermore, the method for constructing the convolutional neural network model of the present invention is as follows:

[0033] Step S4.1, collect real-time image data and related physiological parameters of the transplanted organoids through sensors, and preprocess the collected image data, including denoising, standardization, and data enhancement. The preprocessed data set is divided into a training set, a validation set, and a test set, with a ratio of 70% training set, 15% validation set, and 15% test set to ensure the generalization ability of the model and avoid overfitting;

[0034] Step S4.2, define the structure of the convolutional neural network model, including convolution layer, activation function, pooling layer, fully connected layer, input layer to determine the input image size, then add several convolution layers, extract the spatial features of the image, construct the convolution layer, mathematically expressed as , where x is the input image, w is the convolution kernel, b is the bias, and the activation function is defined as , then add a pooling layer to reduce the size of the feature map and retain important features, then add a fully connected layer to use the features extracted by the convolutional layer and the pooling layer for classification. The fully connected operation formula is , where W is the weight matrix, b is the bias, is the activation function, and finally add the output layer, setting the number of output nodes and activation function according to the specific task: ;

[0035] Step S4.3, train the model using the training set and calculate the model output through forward propagation: ,in, is the predicted output, x is the input image, is the model parameter, and then the loss is calculated based on the predicted results and the true label: , and then calculate the gradient of the loss with respect to the model parameters by backpropagation: ,in, is the learning rate, is the gradient of the loss function with respect to the parameters, and the optimizer is used to update the model parameters based on the obtained results;

[0036] Step S4.4, use the validation set to verify the model, adjust the model parameters and hyperparameters, use the test set to evaluate the performance of the model, calculate the accuracy, precision, recall, and F1 score indicators, deploy the trained model in the real-time monitoring system, continuously monitor the organoids, and issue an alarm in time when an abnormality is detected.

[0037] Technical effects and advantages of the present invention:

[0038] The present invention is based on analyzing and evaluating the morphological abnormality information, functional abnormality information and gene expression information of the target organoid, and evaluates the abnormality degree of the target organoid. When the abnormality degree is judged to be high, an abnormal impact assessment model is constructed to obtain the abnormal impact assessment index on the patient after organoid transplantation, and then a convolutional neural network model is constructed, thereby realizing real-time monitoring of organoid abnormalities, rationally allocating medical resources, and ensuring timely detection and handling of abnormal situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of the organoid abnormality monitoring system based on big data of the present invention.

[0040] Figure 2 This is a flow chart of the organoid abnormality monitoring method based on big data of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The present invention is based on analyzing and evaluating the morphological abnormality information, functional abnormality information and gene expression information of the target organoid, evaluating the degree of abnormality of the target organoid, and when the abnormality degree is judged to be high, constructing an abnormal impact assessment model to obtain the abnormal impact assessment index on the patient after organoid transplantation, and then constructing a convolutional neural network model, thereby realizing real-time monitoring of organoid abnormalities, rationally allocating medical resources, and ensuring timely detection and handling of abnormal situations. Example

[0043] like Figure 1 As shown, a module of an organoid abnormality monitoring system based on big data is as follows:

[0044] The data collection module monitors the target organoids and collects information on morphological abnormalities, functional abnormalities, gene expression, cell-level information, clinical impact information, cost information, and real-time data after transplantation;

[0045] The organoid abnormality assessment module builds an organoid abnormality assessment model to assess the degree of abnormality of the target organoid;

[0046] Abnormal impact assessment module: For areas with high degree of organoid abnormality, an abnormal impact assessment model is constructed by combining cell-level information, clinical impact information, and cost information to obtain the abnormal impact index of the target organoid on the patient;

[0047] The algorithm processing module establishes a convolutional neural network model using the organoid abnormality assessment index, abnormal impact assessment index and real-time data after transplantation as factors, monitors the convolutional neural network model for organoid abnormalities in real time, and decides whether to issue an alarm based on the monitoring results.

[0048] Specifically, in the data acquisition module, the ultimate goal of organoid research is to provide an important scientific basis and technical support for more effective and safer human organ transplantation. In actual application, due to the suddenness of the patient's disease and rejection reaction and considering the medical cost, it is difficult to always cultivate completely abnormal organoids. Therefore, it is a feasible and practical method to quickly screen out organoids with low abnormality and low adverse effects on patients, and build an abnormal monitoring system according to demand. In actual use, due to factors such as equipment cost and the complexity of real-time monitoring, the present invention is based on the long-term normal operation of organoid cultivation and the relative stability of the human body environment after transplantation, and the premise of continuously monitoring the real-time data of organoids, which may lead to the phenomenon of risk changes at a certain monitoring moment.

[0049] In the organoid abnormality assessment module, an organoid abnormality assessment model is constructed. The specific process is to collect the morphological abnormality information, functional abnormality information and gene expression information of the target organoid. The morphological abnormality information includes the morphological abnormality coefficient, which is calibrated as , the dysfunction information includes the dysfunction coefficient, calibrated as , gene expression information includes gene expression coefficients, calibrated as ;

[0050] Morphological abnormality coefficient By collecting the perimeter and area of ​​the target organoid, they are marked as and , then the circularity of the target organoid is , get the average roundness of the same type of organoids in the database and calibrate it as , then the morphological abnormality coefficient ;

[0051] Functional abnormality coefficient By collecting the secretions corresponding to the target organoid, including hormones, enzymes or metabolites, the secretion concentration in the organoid is calibrated as , obtain the concentration of the same type of secretion of the same type of organoid in the database and mark it as , x is the organoid number in the database, x=1, 2, 3, 4, ..., u, u is a positive integer, then the dysfunction coefficient ;

[0052] Gene expression coefficient The expression profile of genes in organoids was analyzed by qPCR, RNA sequencing or single-cell RNA sequencing technology. The gene overlap was obtained by comparing the gene expression of the same type of organoids in the database and calibrated as The expression level and modification status of proteins in organoids were detected by Western Blot, ELISA or mass spectrometry. The protein expression coincidence was obtained by comparing the expression level and modification status of proteins of the same type in the database, which was calibrated as , then the gene expression coefficient ;

[0053] The organoid abnormality assessment model is constructed by weighting the target organoid's morphological abnormality information, functional abnormality information, and gene expression information to generate an organoid abnormality assessment index. , and the corresponding coefficients are morphological anomaly coefficients , dysfunction coefficient , gene expression coefficient , the formula is ;

[0054] at the same time, All are greater than 0 and are set according to actual conditions. For example, the expert weighting method is adopted, that is, experts in related fields are invited to determine the weights of various indicators through professional opinion surveys and comprehensive evaluations to ensure that the weight coefficients can accurately reflect the importance of various indicators in the evaluation of organoid abnormalities. In addition, various methods such as hierarchical analysis method and fuzzy comprehensive evaluation method can also be considered to determine the weight coefficients to ensure the objectivity and scientificity of the weight coefficients. I will not go into details here.

[0055] In the organoid abnormality assessment module, the organoid abnormality assessment index obtained from the organoid abnormality assessment model is used to reflect the degree of abnormality of the target organoid. The larger its value is, the greater the difference between the target organoid and the normal organoid in morphology, the higher the degree of functional abnormality, the rarer the expression of genes and proteins, and the higher the overall degree of abnormality.

[0056] In the Organoid Abnormality Assessment Module, when the Organoid Abnormality Assessment Index When it is greater than the set abnormal threshold, it means that the target organoid is highly abnormal and has a large difference from the normal organoid, and a high abnormal data signal is output for subsequent evaluation;

[0057] When organoid abnormality is assessed When the value is less than or equal to the set abnormality threshold, it indicates that the target organoid has a low degree of abnormality and is not much different from normal organoids. The evaluation result is directly output without further steps.

[0058] In the abnormal impact assessment module, after receiving the high abnormal data signal, the abnormal impact assessment model is constructed. The specific process is to collect the cell level information, clinical impact information and cost information of the target organoid. The cell level information includes the cell level coefficient, which is calibrated as , clinical impact information includes clinical impact coefficient, calibrated as , the cost information includes the cost coefficient, denoted as ;

[0059] Cell level coefficient The number of cells surviving within one week of transplantation into the human body is measured as , obtain the number of cells surviving in the abnormal organoids of human transplantation target one week before, calibrated as , then the cell level coefficient ;

[0060] Clinical impact factor The time of normal life of the patient after the target organoid transplantation is collected and calibrated as , to obtain the normal life time of patients with the same disease who did not undergo organoid transplantation, calibrated as , w is the patient number who did not undergo organoid transplantation, w=1, 2, 3, 4, ..., v, v is a positive integer, then the clinical impact coefficient ;

[0061] Cost factor By collecting the target organoid culture and transplantation costs, the , the average annual income of the patient's family is calculated and calibrated as , then the cost coefficient ;

[0062] The abnormal impact assessment model is constructed by weighting the cell-level information, clinical impact information and cost information of the target organoid to generate the abnormal impact assessment index , and the corresponding coefficients are the cell level coefficients , Clinical Impact Factor , cost coefficient , the formula is ;

[0063] at the same time, All are greater than 0 and are set according to actual conditions. For example, the expert weighting method is adopted, that is, experts in related fields are invited to determine the weights of various indicators through professional opinion surveys and comprehensive evaluations to ensure that the weight coefficients can accurately reflect the importance of each indicator in the evaluation of organoid abnormalities. In addition, various methods such as hierarchical analysis method and fuzzy comprehensive evaluation method can be considered to determine the weight coefficients to ensure the objectivity and scientificity of the weight coefficients. I will not go into details here.

[0064] In the abnormal impact assessment module, the abnormal impact assessment index obtained from the abnormal impact assessment model is used to reflect the impact caused by the abnormality of the target organoid. The larger its value is, the lower the survival rate of the target organoid at the cellular level after transplantation, the less time the patient can live a normal life after transplantation, the greater the burden on the patient's family, and the greater the adverse impact.

[0065] In the abnormal impact assessment module, when the abnormal impact assessment index When it is greater than the set impact threshold, it indicates that the target organoid has a high degree of adverse impact on the patient, and a detection data signal is output. It is recommended to equip the organoid monitoring system after transplantation and conduct subsequent evaluation;

[0066] When abnormal impact assessment index When the value is less than or equal to the set impact threshold, it indicates that the impact of the target organoid is low, and the evaluation result is directly output without further steps.

[0067] In the algorithm processing module, the organoid abnormality assessment index, abnormal impact assessment index and organoid real-time data are used as construction factors of the convolutional neural network model. A convolutional neural network model is established to enable users to achieve real-time monitoring of organoid abnormalities after transplantation.

[0068] Furthermore, the method for constructing the convolutional neural network model of the present invention is as follows:

[0069] Step S4.1, collect real-time image data and related physiological parameters of the transplanted organoids through sensors, and preprocess the collected image data, including denoising, standardization, and data enhancement. The preprocessed data set is divided into a training set, a validation set, and a test set, with a ratio of 70% training set, 15% validation set, and 15% test set to ensure the generalization ability of the model and avoid overfitting;

[0070] Step S4.2, define the structure of the convolutional neural network model, including convolution layer, activation function, pooling layer, fully connected layer, input layer to determine the input image size, then add several convolution layers, extract the spatial features of the image, construct the convolution layer, mathematically expressed as , where x is the input image, w is the convolution kernel, b is the bias, and the activation function is defined as , then add a pooling layer to reduce the size of the feature map and retain important features, then add a fully connected layer to use the features extracted by the convolutional layer and the pooling layer for classification. The fully connected operation formula is , where W is the weight matrix, b is the bias, is the activation function, and finally add the output layer, setting the number of output nodes and activation function according to the specific task: ;

[0071] Step S4.3, train the model using the training set and calculate the model output through forward propagation: ,in, is the predicted output, x is the input image, is the model parameter, and then the loss is calculated based on the predicted results and the true label: , and then calculate the gradient of the loss with respect to the model parameters by backpropagation: ,in, is the learning rate, is the gradient of the loss function with respect to the parameters, and the optimizer is used to update the model parameters based on the obtained results;

[0072] Step S4.4, use the validation set to verify the model, adjust the model parameters and hyperparameters, use the test set to evaluate the performance of the model, calculate the accuracy, precision, recall, and F1 score indicators, deploy the trained model in the real-time monitoring system, continuously monitor the organoids, and issue an alarm in time when an abnormality is detected. Example

[0073] like Figure 2 As shown, a method for monitoring organoid abnormalities based on big data comprises the following steps:

[0074] Step S1, monitoring the target organoid, collecting morphological abnormality information, functional abnormality information, gene expression information, cell level information, clinical impact information, cost information, and real-time data after transplantation;

[0075] Step S2, constructing an organoid abnormality assessment model to assess the degree of abnormality of the target organoid;

[0076] Step S3, for regions with high degree of organoid abnormality, an abnormality impact assessment model is constructed by combining cell-level information, clinical impact information, and cost information to obtain an abnormality impact index of the target organoid on the patient;

[0077] Step S4, establishing a convolutional neural network model using the organoid abnormality assessment index, the abnormal impact assessment index and real-time data after transplantation as factors, performing real-time monitoring of organoid abnormalities on the convolutional neural network model, and deciding whether to issue an alarm based on the monitoring results.

[0078] The above formulas are all dimensionless and calculated numerically. Specific dimension removal can be achieved by various means such as standardization, which will not be elaborated here. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0079] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, an ATA hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state ATA hard disk.

[0080] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0081] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0082] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0083] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0085] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile ATA hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0086] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A big data-based organoid abnormality monitoring system, characterized in that: Includes the following modules: The data collection module monitors the target organoids and collects information on morphological abnormalities, functional abnormalities, gene expression, cell-level information, clinical impact information, cost information, and real-time data after transplantation; The organoid abnormality assessment module builds an organoid abnormality assessment model to assess the degree of abnormality of the target organoid; Abnormal impact assessment module: For areas with high abnormality in organoids, an abnormal impact assessment model is constructed by combining cell-level information, clinical impact information, and cost information to obtain the abnormal impact index of the target organoid on the patient; The algorithm processing module uses the organoid abnormality assessment index, abnormal impact assessment index, and real-time data after transplantation as factors to establish a convolutional neural network model, monitor the organoid abnormality in real time on the convolutional neural network model, and decide whether to issue an alarm based on the monitoring results; In the organoid abnormality assessment module, an organoid abnormality assessment model is constructed. The specific process is to collect the morphological abnormality information, functional abnormality information and gene expression information of the target organoid. The morphological abnormality information includes the morphological abnormality coefficient, which is calibrated as X yc , the dysfunction information includes the dysfunction coefficient, calibrated as G yc , gene expression information includes gene expression coefficient, denoted as J bd ; Morphological abnormality coefficient X yc By collecting the perimeter and area of ​​the target organoid, they are marked as Z c and M j , then the circularity of the target organoid is Get the average roundness of the same type of organoids in the database and calibrate it as The coefficient of morphological abnormality Functional abnormality coefficient G yc By collecting the secretions corresponding to the target organoid, including hormones, enzymes or metabolites, the secretion concentration in the organoid is calibrated as F n , obtain the concentration of the same type of secretion of the same type of organoid in the database and mark it as F x , x is the organoid number in the database, x = 1, 2, 3, 4, ..., u, u is a positive integer, then the dysfunction coefficient Gene expression coefficient J bd The expression profile of genes in organoids was analyzed by qPCR, RNA sequencing or single-cell RNA sequencing technology. The gene overlap was obtained by comparing the gene expression of the same type of organoids in the database and marked as C j The expression level and modification status of proteins in organoids were detected by Western Blot, ELISA or mass spectrometry. The protein expression coincidence was obtained by comparing the expression level and modification status of proteins of the same type in the database, which was calibrated as C d , then the gene expression coefficient J bd =C j +C d ; The organoid abnormality assessment model is constructed by weighting the morphological abnormality information, functional abnormality information and gene expression information of the target organoid to generate the organoid abnormality assessment index YC pg , the corresponding coefficients are morphological abnormality coefficient X yc , dysfunction coefficient G yc , gene expression coefficient J bd , the formula is YC pg =a1×X yc +a2×G yc -a3×J bd , a1, a2, and a3 are the weight coefficients of the corresponding indicators, all greater than 0; In the Organoid Abnormality Assessment Module, when the Organoid Abnormality Assessment Index YC pg When it is greater than the set abnormal threshold, a high abnormal data signal is output for subsequent evaluation; When the organoid abnormality assessment index YC pg When the value is less than or equal to the set abnormal threshold, the evaluation result is directly output without further steps. In the abnormal impact assessment module, after receiving the high abnormal data signal, the abnormal impact assessment model is constructed. The specific process is to collect the cell level information, clinical impact information and cost information of the target organoid. The cell level information includes the cell level coefficient, which is calibrated as S xb , clinical impact information includes clinical impact coefficient, denoted as Y lc , cost information includes cost coefficient, denoted as C bx ; Cell level coefficient S xb The number of cells that survived within one week of transplantation into the human body was collected and labeled as H. y , obtain the number of cells surviving in the abnormal organoids of human transplantation target one week before, denoted as H q , then the cell level coefficient S xb =H q -H y ; Clinical impact coefficient Y lc The time of normal life of the patient after the target organoid transplantation is collected and marked as T h , obtain the normal life time of patients with the same disease who have not undergone organoid transplantation, calibrated as T w , w is the patient number who did not undergo organoid transplantation, w = 1, 2, 3, 4, ..., v, v is a positive integer, then the clinical impact coefficient Cost coefficient C bx By collecting the target organoid culture and transplantation costs, labeled as C y , calculate the average annual income of the patient's family, denoted as N s , then the cost coefficient C bx =C y / N s ; The abnormal impact assessment model is constructed by weighting the cell-level information, clinical impact information and cost information of the target organoid to generate the abnormal impact assessment index YX pg , the corresponding coefficients are the cell level coefficient S xb , Clinical impact coefficient Y lc , cost coefficient C bx , the formula is YX pg =β1×S xb -β2×Y lc +β3×C bx , β1, β2, and β3 are the weight coefficients of the corresponding indicators, all greater than 0; In the abnormal impact assessment module, when the abnormal impact assessment index YX pg When it is greater than the set impact threshold, the detection data signal is output, and it is recommended to equip the organoid monitoring system after transplantation and conduct subsequent evaluation; When the abnormal impact assessment index YX pg When the impact value is less than or equal to the set impact threshold, it indicates that the impact of the target organoid is low, and the evaluation result is directly output without further steps. In the algorithm processing module, the organoid abnormality assessment index, abnormal impact assessment index and organoid real-time data are used as the construction factors of the convolutional neural network model, and a convolutional neural network model is established to enable users to achieve real-time monitoring of organoid abnormalities after transplantation; Furthermore, the convolutional neural network model is constructed as follows: Step S4.1, collecting real-time image data and related physiological parameters of the transplanted organoids through sensors, and preprocessing the collected image data to divide the preprocessed data set into a training set, a validation set, and a test set; Step S4.2, define the structure of the convolutional neural network model, including convolution layer, activation function, pooling layer, and fully connected layer. The input layer determines the input image size, and then adds several convolution layers to extract the spatial features of the image and construct the convolution layer. The mathematical expression is Conv(x)=x*w+b, where x is the input image, w is the convolution kernel, and b is the bias. The activation function is defined as ReLU(x)=max(0,x). Then add a pooling layer to reduce the size of the feature map and retain important features. Then add a fully connected layer to use the features extracted by the convolution layer and the pooling layer for classification. The fully connected operation formula is y=σ(Wx+b), where W is the weight matrix, b is the bias, and σ is the activation function. Finally, add the output layer and set the number of output nodes and activation function according to the specific task: Step S4.3, train the model using the training set and calculate the model output through forward propagation: in, is the predicted output, x is the input image, θ is the model parameter, and the loss is calculated based on the predicted result and the true label: Then backpropagate to calculate the gradient of the loss with respect to the model parameters: Where η is the learning rate, is the gradient of the loss function with respect to the parameters, and the optimizer is used to update the model parameters based on the obtained results; Step S4.4, use the validation set to verify the model, adjust the model parameters and hyperparameters, use the test set to evaluate the performance of the model, calculate the accuracy, precision, recall, and F1 score indicators, deploy the trained model in the real-time monitoring system, continuously monitor the organoids, and issue an alarm in time when an abnormality is detected.

Citation Information

Patent Citations

  • High-throughput organoid phenotype screening system and method

    CN110283724A

  • Method for providing information for diagnosis and drug screening of brain-related diseases by using three-dimensional brain cell aggregate

    CN116598007A

  • Prediction method and system for drug-induced acute injury based on liver organ model

    CN117305402A

  • AI prediction method for drug-induced acute kidney injury based on kidney organ model

    CN117393158A

  • Quantifying and analyzing organoids from digital images

    EP4120119A1