An intelligent liquid nitrogen preservation effect evaluation system

Abstract

The present invention discloses an intelligent liquid nitrogen preservation effect evaluation system, comprising a data acquisition module, an alignment preprocessing module, a membership evaluation module, a reliability weight calculation module, a liquid nitrogen preservation evaluation model establishment module, and a liquid nitrogen preservation effect evaluation module. The present invention belongs to the field of liquid nitrogen preservation evaluation, and specifically refers to an intelligent liquid nitrogen preservation effect evaluation system. This scheme uses dual-center membership characterization to refine critical distinctions, and suppresses false anomalies introduced by occasional measurement errors in the neighborhood through local disharmony; introduces projection residual measurement to eliminate global projection distortion samples caused by instrument bias; automatically eliminates anomalies based on reliability weights; introduces reliability weights into the slack variable penalty for each sample to suppress interference from low-confidence detection results and prevent noise samples from distorting the decision surface; focuses on the most difficult-to-distinguish samples based on upper and lower bound separation losses and hyperbolic soft boundary bandwidth, and improves the sensitivity of discrimination for critical survival rate samples.

Description

Technical Field

[0001] The present invention relates to the field of liquid nitrogen preservation evaluation, and in particular to an intelligent liquid nitrogen preservation effect evaluation system. Background Art

[0002] The liquid nitrogen preservation effectiveness evaluation system is a comprehensive system used to assess the preservation effectiveness of samples stored in liquid nitrogen cryogenic environments. However, typical liquid nitrogen preservation effectiveness evaluation systems are unable to provide detailed judgments on sublethally damaged and critical samples, resulting in a high misjudgment rate for critical samples and poor accuracy in identifying false anomalies easily introduced by measurement noise, batch drift, or local instrument errors. Furthermore, these systems are unable to impose additional constraints on some of the most difficult liquid nitrogen-preserved samples, resulting in insufficient discrimination capabilities and a dominant decision-making surface dominated by noisy samples, leading to unstable evaluations. Summary of the Invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent liquid nitrogen preservation effect evaluation system. In view of the problem that the general liquid nitrogen preservation effect evaluation system is unable to make a delicate judgment on sublethal damage and critical state samples, resulting in a high misjudgment rate for critical samples and poor accuracy in identifying false anomalies easily introduced by measurement noise, batch drift or local instrument errors, this solution uses dual-center membership characterization to refine critical distinction and can distinguish between mildly damaged samples and severely inactivated samples; uses local disharmony to suppress false anomalies introduced by occasional measurement errors in the neighborhood; introduces projection residual measurement to eliminate false anomalies caused by batch drift or Global projection distortion samples caused by instrument deviation; automatic anomaly removal based on reliability weights; thereby improving the accuracy of liquid nitrogen preservation effect evaluation; in view of the fact that general liquid nitrogen preservation effect evaluation systems are unable to impose additional constraints on some of the most difficult to distinguish liquid nitrogen preservation samples, insufficient discrimination ability, and noise samples dominating the decision surface, resulting in unstable evaluation, this scheme introduces reliability weights into the slack variable penalty for each sample to suppress the interference of low-confidence detection results and prevent noise samples from distorting the decision surface; based on the upper and lower bound separation loss and hyperbolic soft boundary bandwidth, it focuses on the most difficult to distinguish samples and improves the discrimination sensitivity of critical survival rate samples, thereby improving the evaluation effect.

[0004] The technical solution adopted by the present invention is as follows: the present invention provides an intelligent liquid nitrogen preservation effect evaluation system, including a data acquisition module, an alignment preprocessing module, a membership evaluation module, a reliability weight calculation module, a liquid nitrogen preservation evaluation model establishment module and a liquid nitrogen preservation effect evaluation module;

[0005] The data acquisition module constructs a historical liquid nitrogen preservation sample set and a sample set to be evaluated;

[0006] The alignment preprocessing module maps the samples to a common subspace by solving the projection matrix;

[0007] The membership evaluation module calculates the distance between the sample to be evaluated and the center of the positive and negative classes in the projection subspace to characterize the sample attribution;

[0008] The reliability weight calculation module introduces projection residuals and generates sample reliability weights based on the aggregation index;

[0009] The liquid nitrogen preservation evaluation model establishment module adopts a weighted SVM with a hyperbolic soft boundary bandwidth to train a classifier under an objective function with a weighted slack variable penalty;

[0010] The liquid nitrogen preservation effect evaluation module evaluates the liquid nitrogen preservation effect of the sample set to be evaluated based on the liquid nitrogen preservation evaluation model.

[0011] Furthermore, the data acquisition module collects historical liquid nitrogen preservation sample sets and sample sets to be evaluated; excellent preservation and failed preservation are used as supervision labels; the liquid nitrogen preservation samples include biological indicators, physical parameters and environmental parameters.

[0012] Furthermore, the alignment preprocessing module first projects the source domain of the historical liquid nitrogen stored samples and the target domain of the samples to be evaluated, and the source domain data matrix is ​​X S , the target domain matrix is ​​X T , find the projection matrix P, expressed as: ; After the projection is obtained, it is expressed as: ; ;in, and are the mean vectors of source and target domain samples respectively; is a scalar hyperparameter used to balance the importance of the source domain and target domain divergence matrices; T is the transpose operation; and are the source and target domain sample positions after projection, respectively; tr(·) is the matrix trace.

[0013] Furthermore, the membership evaluation module first defines two category centers in the projection subspace: the positive category center Negative center ;in, It is a positive category center; It is a negative class center; and are respectively the sets of good preservation and failed preservation samples after projection; and are the total number of samples with good preservation and failed preservation respectively; x is the sample index; for any sample to be evaluated , calculate: distance to the center, the formula used is: ; ;in, and are the distances from the sample to be evaluated to the positive center and the negative center respectively; the degree of membership, the formula used is: ; ; Non-membership degree, the formula used is: ; ; ;in, is the positive class limit distance; is the positive class membership; is a smoothing term; is the minimum distance difference baseline; is the distance between centers; degree of non-affiliation; is the local disharmony, The set of k nearest neighbors N i , ; is the nearest neighbor set N i samples; and is the sample label.

[0014] Furthermore, the reliability weight calculation module integrates membership and non-membership to generate the reliability weight of each sample and introduces the projection residual metric. The formula used is: ;in, is the reliability weight; is the weight balancing factor; is the aggregation index.

[0015] Furthermore, the liquid nitrogen preservation evaluation model establishment module is based on SVM, which uses the reliability weight of each sample to weight the slack variable penalty and penalizes the upper and lower bound violations separately; it introduces the hyperbolic soft boundary bandwidth , the objective function is expressed as: ;

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] in, is the weight vector in kernel space; b is the bias term; is a slack variable; is the slack variable of the i-th sample; and is the loss coefficient; and They are the boundary violation losses for over-conservative and over-confident boundary samples respectively; is the upper bound scaling factor; is the true label; is the predicted label; C is the penalty coefficient; m is the total number of samples; is the kernel mapping function; is the initial bandwidth, is the boundary coefficient.

[0021] Furthermore, the liquid nitrogen preservation effect evaluation module jointly classifies the historical liquid nitrogen preservation sample set and the sample set to be evaluated based on the liquid nitrogen preservation evaluation model; divides the total sample data into a test set and a training set, initializes the SVM parameters, and performs iterative training based on the objective function gradient update parameters. The training is completed when the objective function converges or reaches the maximum number of iterations; sets a test threshold. If the prediction accuracy of the trained liquid nitrogen preservation evaluation model for the test set is higher than the test threshold, the liquid nitrogen preservation evaluation model is established, and the classification result of the sample to be evaluated at this time is used as the liquid nitrogen preservation evaluation result; otherwise, retraining is performed.

[0022] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0023] (1) In view of the problem that the general liquid nitrogen preservation effect evaluation system cannot make a detailed judgment on sublethal damage and critical state samples, resulting in a high misjudgment rate for critical samples and poor accuracy in identifying false anomalies easily introduced by measurement noise, batch drift or local instrument errors, this scheme uses dual-center membership characterization to refine critical distinction and distinguish between mildly damaged samples and severely inactivated samples; suppresses false anomalies introduced by occasional measurement errors in the neighborhood through local disharmony; introduces projection residual measurement to eliminate global projection distortion samples caused by batch drift or instrument deviation; and automatically removes anomalies based on reliability weights, thereby improving the accuracy of liquid nitrogen preservation effect evaluation.

[0024] (2) In view of the fact that the general liquid nitrogen preservation effect evaluation system cannot impose additional constraints on some of the most difficult to distinguish liquid nitrogen preservation samples, the discrimination ability is insufficient, and the noise samples dominate the decision surface, resulting in unstable evaluation, this scheme introduces a reliability weight to the slack variable penalty of each sample to suppress the interference of low-confidence detection results and prevent noise samples from distorting the decision surface; based on the upper and lower bound separation loss and hyperbolic soft boundary bandwidth, it focuses on the most difficult to distinguish samples and improves the discrimination sensitivity of critical survival rate samples, thereby improving the evaluation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1This is a flow chart of an intelligent liquid nitrogen preservation effect evaluation system provided by the present invention.

[0026] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] In the description of the present invention, it should be understood that terms such as "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0029] Example 1, see Figure 1 The present invention provides an intelligent liquid nitrogen preservation effect evaluation system, which includes a data acquisition module, an alignment preprocessing module, a membership evaluation module, a reliability weight calculation module, a liquid nitrogen preservation evaluation model establishment module and a liquid nitrogen preservation effect evaluation module;

[0030] The data acquisition module constructs a historical liquid nitrogen preservation sample set and a sample set to be evaluated; and sends the data to the alignment preprocessing module;

[0031] The alignment preprocessing module maps the samples to the common subspace by solving the projection matrix; and sends the data to the membership evaluation module;

[0032] The membership evaluation module calculates the distance between the sample to be evaluated and the center of the positive and negative classes in the projection subspace, characterizing the sample attribution; and sends the data to the reliability weight calculation module;

[0033] The reliability weight calculation module introduces the projection residual and generates the sample reliability weight based on the aggregation index; and sends the data to the liquid nitrogen preservation assessment model establishment module;

[0034] The liquid nitrogen preservation evaluation model establishment module adopts a weighted SVM with a hyperbolic soft boundary bandwidth to train a classifier under an objective function with a weighted slack variable penalty; and sends the data to the liquid nitrogen preservation effect evaluation module;

[0035] The liquid nitrogen preservation effect evaluation module evaluates the liquid nitrogen preservation effect of the sample set to be evaluated based on the liquid nitrogen preservation evaluation model.

[0036] Example 2, see Figure 1 This embodiment is based on the above embodiment. The data acquisition module collects historical liquid nitrogen preservation sample sets and sample sets to be evaluated, including excellent preservation samples with high cell survival rate, membrane integrity, low ROS level, and small DNA fragmentation index after thawing, and failed preservation samples with significantly decreased survival rate / function or cell inactivation after thawing; excellent preservation and failed preservation are used as supervision labels; liquid nitrogen preservation samples include biological indicators, physical parameters and environmental parameters; the biological indicators include survival rate after thawing, membrane integrity, ROS level and DNA fragmentation index; the physical parameters include freezing rate, thawing rate, freezing program curve, CPA concentration; the environmental parameters include liquid nitrogen temperature, temperature change rate, humidity and storage container pressure.

[0037] Example 3, see Figure 1 This embodiment is based on the above embodiment. Due to the system drift of slight differences in cooling rate curves and CPA batch differences between different laboratories and different batches, the alignment preprocessing module first projects the source domain of the historical liquid nitrogen stored samples and the target domain of the samples to be evaluated. Let the source domain data matrix be X S , the target domain matrix is ​​X T , find the projection matrix P, expressed as: ; After the projection is obtained, it is expressed as: ; ;in, and are the mean vectors of source and target domain samples respectively; is a scalar hyperparameter used to balance the importance of the source domain and target domain divergence matrices; T is the transpose operation; and are the sample positions of the source and target domains after projection, respectively; tr(·) is the matrix trace; it eliminates the drift of different experimental conditions and instruments and improves the subsequent cross-batch generalization ability.

[0038] Example 4, see Figure 1 This embodiment is based on the above embodiment. In the projection subspace, the membership evaluation module uses the centers of the two categories of samples, good preservation and failed preservation, to measure the membership and non-membership of each sample to be evaluated. First, two category centers are defined: the positive category center Negative center ;in, It is a positive category center; It is a negative class center; and are respectively the sets of good preservation and failed preservation samples after projection; and are the total number of samples with good preservation and failed preservation respectively; x is the sample index; for any sample to be evaluated , calculate: distance to the center, the formula used is: ; ;in, and are the distances from the sample to be evaluated to the positive center and the negative center respectively; the degree of membership, the formula used is: ; ; Non-membership degree, the formula used is: ; ; ;in, is the positive class limit distance; is the positive class membership; is a smoothing term; is the minimum distance difference baseline; is the distance between centers; degree of non-affiliation; is the local disharmony, The set of k nearest neighbors N i , ; is the nearest neighbor set N i samples; and It is a sample label, and target domain samples are not considered; the dual center can simultaneously characterize the good and bad poles, greatly improving the ability to distinguish the critical state of sub-lethal damage samples; the non-membership term can quickly approach 1 when the sample has leaned towards the failure side, automatically eliminating those abnormal points that obviously failed to thaw; local disharmony suppresses false anomalies caused by occasional measurement noise; similarly, at the same distance to the good center, different samples can be reasonably distinguished based on the difference in distance from the failure center.

[0039] Example 5, see Figure 1 This embodiment is based on the above embodiment. The reliability weight calculation module integrates membership and non-membership to generate the reliability weight of each sample, and introduces the projection residual metric to balance local and global credibility and better eliminate anomalies. The formula used is: ;in, is the reliability weight; is the weight balancing factor; is the aggregation index; it gives greater weight to samples with high membership + low non-membership, weakening the suspicious points introduced by measurement noise or batch drift.

[0040] By performing the above operations, the general liquid nitrogen preservation effect evaluation system is unable to make detailed judgments on sublethal damage and critical state samples, resulting in a high misjudgment rate for critical samples and poor accuracy in identifying false anomalies easily introduced by measurement noise, batch drift or local instrument errors. This scheme uses dual-center membership characterization to refine critical distinctions and distinguish between mildly damaged samples and severely inactivated samples; uses local disharmony to suppress false anomalies introduced by occasional measurement errors in the neighborhood; introduces projection residual measurement to eliminate global projection distortion samples caused by batch drift or instrument deviation; and automatically removes anomalies based on reliability weights, thereby improving the accuracy of liquid nitrogen preservation effect evaluation.

[0041] Example 6, see Figure 1 This embodiment is based on the above embodiment. The liquid nitrogen preservation evaluation model establishment module is based on SVM, and the slack variable penalty is weighted by the reliability weight of each sample, and the upper and lower bound violations are penalized separately; the hyperbolic soft boundary bandwidth is introduced , automatically amplify the distinction penalty for samples near the boundary, and the objective function is expressed as: ;

[0042] ;

[0043] ;

[0044] ;

[0045] ;

[0046] in, is the weight vector in kernel space; b is the bias term; is a slack variable; is the slack variable of the i-th sample; and is the loss coefficient; and They are the boundary violation losses for over-conservative and over-confident boundary samples respectively; is the upper bound scaling factor; is the true label; is the predicted label; C is the penalty coefficient; m is the total number of samples; is the kernel mapping function; is the initial bandwidth, is the boundary coefficient.

[0047] By performing the above operations, we address the problems of general liquid nitrogen preservation effect evaluation systems, which are unable to impose additional constraints on some of the most difficult-to-distinguish liquid nitrogen preservation samples, resulting in insufficient discrimination ability and noisy samples dominating the decision surface, leading to unstable evaluation. This scheme introduces a reliability weight into the slack variable penalty for each sample to suppress the interference of low-confidence detection results and prevent noisy samples from distorting the decision surface. Based on the upper and lower bound separation loss and hyperbolic soft boundary bandwidth, this scheme focuses on the most difficult-to-distinguish samples and improves the discrimination sensitivity of samples with critical survival rates, thereby improving the evaluation effect.

[0048] Example 7, see Figure 1 This embodiment is based on the above embodiment. The liquid nitrogen preservation effect evaluation module classifies the historical liquid nitrogen preservation sample set and the sample set to be evaluated based on the liquid nitrogen preservation evaluation model; the total sample data is divided into a test set and a training set, the SVM parameters are initialized, and the parameters are updated based on the objective function gradient for iterative training. The training is completed when the objective function converges or reaches the maximum number of iterations; a test threshold is set. If the prediction accuracy of the trained liquid nitrogen preservation evaluation model for the test set is higher than the test threshold, the liquid nitrogen preservation evaluation model is established, and the classification result of the sample to be evaluated at this time is used as the liquid nitrogen preservation evaluation result; otherwise, retraining is performed.

[0049] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

[0050] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. An intelligent liquid nitrogen preservation effect evaluation system, characterized by: The system includes a data acquisition module, an alignment preprocessing module, a membership evaluation module, a reliability weight calculation module, a liquid nitrogen preservation evaluation model establishment module, and a liquid nitrogen preservation effect evaluation module; The data acquisition module constructs a historical liquid nitrogen preservation sample set and a sample set to be evaluated; The alignment preprocessing module maps the samples to a common subspace by solving the projection matrix; The membership evaluation module calculates the distance between the sample to be evaluated and the center of the positive and negative classes in the projection subspace to characterize the sample attribution; The reliability weight calculation module introduces projection residuals and generates sample reliability weights based on the aggregation index; The liquid nitrogen preservation evaluation model establishment module adopts a weighted SVM with a hyperbolic soft boundary bandwidth to train a classifier under an objective function with a weighted slack variable penalty; The liquid nitrogen preservation effect evaluation module evaluates the liquid nitrogen preservation effect of the sample set to be evaluated based on the liquid nitrogen preservation evaluation model; The data acquisition module collects historical liquid nitrogen preservation sample sets and sample sets to be evaluated; excellent preservation and failed preservation are used as supervision labels; liquid nitrogen preservation samples include biological indicators, physical parameters and environmental parameters; The liquid nitrogen preservation evaluation model establishment module is based on SVM, which uses the reliability weight of each sample to weight the slack variable penalty and penalizes the upper and lower bound violations separately; it introduces hyperbolic soft boundary bandwidth , the objective function is expressed as: ; ； ； ； ； in, is the weight vector in kernel space; b is the bias term; is a slack variable; is the slack variable of the i-th sample; and is the loss coefficient; and They are the boundary violation losses for over-conservative and over-confident boundary samples respectively; is the upper bound scaling factor; is the true label; is the predicted label; C is the penalty coefficient; m is the total number of samples; is the kernel mapping function; is the initial bandwidth, is the boundary coefficient; is the reliability weight.

2. The intelligent liquid nitrogen preservation effect evaluation system according to claim 1, characterized in that: The alignment preprocessing module first projects the source domain of the historical liquid nitrogen storage sample and the target domain of the sample to be evaluated, and the source domain data matrix is ​​X S , the target domain matrix is ​​X T , find the projection matrix P, expressed as: ; After the projection is obtained, it is expressed as: ; ;in, and are the mean vectors of source and target domain samples respectively; is a scalar hyperparameter used to balance the importance of the source domain and target domain divergence matrices; T is the transpose operation; and are the source and target domain sample positions after projection, respectively; tr(·) is the matrix trace.

3. The intelligent liquid nitrogen preservation effect evaluation system according to claim 2, characterized in that: The membership evaluation module first defines two category centers in the projection subspace: the positive category center Negative center ;in, It is a positive category center; It is a negative class center; and are respectively the sets of good preservation and failed preservation samples after projection; and are the total number of samples with good preservation and failed preservation respectively; x is the sample index; for any sample to be evaluated , calculate: distance to the center, the formula used is: ; ;in, and are the distances from the sample to be evaluated to the positive center and the negative center respectively; the degree of membership, the formula used is: ; ; Non-membership degree, the formula used is: ; ; ;in, is the positive class limit distance; is the positive class membership; is a smoothing term; is the minimum distance difference baseline; is the distance between centers; degree of non-affiliation; is the local disharmony, The set of k nearest neighbors N i , ; is the nearest neighbor set N i samples; and is the sample label.

4. The intelligent liquid nitrogen preservation effect evaluation system according to claim 3, characterized in that: The reliability weight calculation module integrates membership and non-membership to generate the reliability weight of each sample and introduces the projection residual metric. The formula used is: ;in, is the weight balancing factor; is the aggregation index.

5. The intelligent liquid nitrogen preservation effect evaluation system according to claim 4, characterized in that: The liquid nitrogen preservation effect evaluation module is based on the liquid nitrogen preservation evaluation model to jointly classify the historical liquid nitrogen preservation sample set and the sample set to be evaluated; the total sample data is divided into a test set and a training set, the SVM parameters are initialized, and the parameters are updated based on the objective function gradient to perform iterative training. When the objective function converges or reaches the maximum number of iterations, the training is completed; A test threshold is set. If the prediction accuracy of the trained liquid nitrogen preservation evaluation model for the test set is higher than the test threshold, the liquid nitrogen preservation evaluation model is established, and the classification result of the sample to be evaluated at this time is used as the liquid nitrogen preservation evaluation result; otherwise, retraining is performed.

Images