Assessment system for bullfrog nerve experiment

Abstract

The invention discloses an examination system for a bullfrog nerve experiment. The examination system comprises a data acquisition module, a data processing module and a data output module. The data acquisition module is used for acquiring experimental data of a bullfrog nerve experiment; the data processing module is used for processing and analyzing the experimental data to generate assessment data; and the data output module is used for outputting the assessment data. The bullfrog reflex arc analysis experiment monitoring system can realize automatic monitoring of bullfrog reflex arc analysis experiments operated by students, and improves accuracy, normalization and effectiveness of experiment operations of the students.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of animal experiment definition, and particularly relates to a system for evaluating bullfrog nerve experiment. BACKGROUND

[0002] The bullfrog reflex arc analysis experiment is an important experiment project in clinic. Medical colleges score and evaluate medical students by observing their comprehensive performance in the bullfrog reflex arc analysis experiment. At present, the scoring and evaluation mainly rely on naked eyes, which has great subjectivity and error probability. Therefore, how to develop a new experiment evaluation system to overcome the above problems in the prior art is a direction for further research of those skilled in the art. SUMMARY

[0003] The present application aims to provide a system for evaluating bullfrog nerve experiment, which can realize automatic monitoring of students' operation of the bullfrog reflex arc analysis experiment, and improve the accuracy, standardization and effectiveness of students' experiment operation.

[0004] The present application discloses a system for evaluating bullfrog nerve experiment, which comprises:

[0005] A data acquisition module, which is used for acquiring experiment data of the bullfrog nerve experiment;

[0006] A data processing module, which is used for processing and analyzing the experiment data and generating evaluation data;

[0007] A data output module, which is used for outputting the evaluation data.

[0008] Preferably, in the system for evaluating bullfrog nerve experiment,

[0009] The data acquisition module comprises:

[0010] A camera, which is used for acquiring action video in the bullfrog nerve experiment; the action video comprises action video of the bullfrog and action video of the operator

[0011] The data processing module is used for extracting key frames from the action video, extracting abnormal behaviors based on the key frames, and generating evaluation data based on the abnormal behaviors.

[0012] Preferably, in the system for evaluating bullfrog nerve experiment,

[0013] The key frames are extracted by using a clustering algorithm to obtain the key frames in the action video.

[0014] ​The extraction of abnormal behavior based on the keyframes includes: extracting abnormal behavior based on histogram entropy value as the total discriminative feature of abnormal behavior.

[0015] Preferably, in the above-mentioned assessment system for the bullfrog neural experiment,

[0016] The use Clustering algorithms are used to obtain keyframes from bullfrog behavior videos, including:

[0017] Step 110: Define the video image frame sample as... , Define the number of cluster categories. and the initial membership matrix ;

[0018] Step 120: Calculate the cluster centers using the following formula based on the results of the first iteration:

[0019]

[0020] in, For the first The first iteration of the class contains elements like... For the first In the results of the next iteration Belonging to the Membership values ​​of the classes; the similarity between the cluster centers and each sample is obtained using Euclidean distance:

[0021]

[0022] like Then its membership degree It equals 1, and The membership degree of belonging to other categories is 0; if hour, The Secondary membership matrix Calculate using the following formula:

[0023]

[0024] Step 130: Definition For weighted index, if , ;like , ;Compare and ,like , As the first After the next iteration, the membership matrix satisfies the above two conditions; otherwise, the iteration stops. It is a very small positive value;

[0025] Step 140: Calculate each cluster domain The total average distance of all samples to the corresponding cluster center and the average distance of each sample to the cluster center

[0026]

[0027]

[0028] When the number of cluster centers is less than or equal to half of the defined value, that is, The existing cluster center is split;

[0029] Calculate the standard deviation vector of the sample cluster in each clustering Where the number of clusters The dimension is

[0030] Calculate the maximum component value of , if and one of the two conditions is met: Or , the corresponding cluster center is split into two new cluster centers and ;

[0031] When the distance between the new cluster centers and is less than , the two cluster centers are merged;

[0032] Step 150: If , the clustering ends; otherwise, go to step 120.

[0033] Preferably, in the above examination system of bullfrog neural experiment,

[0034] The abnormal behavior based on histogram entropy value as the total feature extraction of abnormal behavior includes:

[0035] Define the histogram model as:

[0036] Where, Indicates the frame, The number of histogram intervals,

[0037]

[0038] In the formula, Indicates the​​​ the first behavior feature in the frame video image the first behavior feature, a weight value of the first behavior feature, a feature value interval corresponding to the first behavior feature, a histogram interval number, a Kronecker function; define the entropy of the image as:

[0039] define the entropy value calculation formula as:

[0040]

[0041] the greater the obtained value is, the greater the motion behavior change of the bullfrog is;

[0042]

[0043]

[0044] when abnormal behavior is determined to occur, otherwise, normal behavior is determined to occur.

[0045] Preferably, in the above-mentioned bullfrog nerve experiment examination system,

[0046] the generating examination data based on the abnormal behavior comprises: assigning a weight value to the abnormal behavior, and performing weighted summation on the abnormal behavior occurring in one bullfrog nerve experiment based on the weight value to obtain an examination deduction coefficient.

[0047] Preferably, in the above-mentioned bullfrog nerve experiment examination system,

[0048] the data acquisition module further comprises:

[0049] a heart rate sensor, which is used for monitoring the heart rate of the experimental bullfrog;

[0050] a temperature and humidity sensor, which is used for collecting the environmental temperature and humidity of the bullfrog nerve experiment.

[0051] Preferably, in the above-mentioned bullfrog nerve experiment examination system, the data output module comprises a printer / display.

[0052] Compared with the prior art, the present application has the following advantages:

[0053] Firstly, the present application can realize automatic monitoring of the student's operation of the bullfrog reflex arc analysis experiment, and improve the accuracy, standardization and effectiveness of the student's experimental operation. At the same time, the present application has simple structure, convenient operation process, and is easy to prepare and use. BRIEF DESCRIPTION OF DRAWINGS​​​

[0054] Figure 1 A module diagram for the embodiment 1.

[0055] In the figure, the component names corresponding to each reference numeral are as follows:

[0056] 100, data acquisition module; 200, data processing module; 300, data output module. DETAILED DESCRIPTION

[0057] The present application is described below by way of specific embodiments, and those skilled in the art can easily understand other advantages and effects of the present application from the disclosure. The present application can also be implemented or applied by different specific embodiments, and various modifications or changes can be made to the details in the specification without departing from the spirit of the present application.

[0058] Embodiment 1, please refer to Figure 1 :

[0059] An evaluation system for bullfrog nerve experiment, comprising: a data acquisition module 100, a data processing module 200 and a data output module 300.

[0060] The data acquisition module 100 is configured to acquire experimental data of the bullfrog nerve experiment; the data processing module 200 is configured to process and analyze the experimental data and generate evaluation data; and the data output module is configured to output the evaluation data. Further, the data acquisition module 100 comprises a camera device, a physiological sensing device and an environmental sensing device. Among them,

[0061] The camera device can be installed in the laboratory and used to acquire action videos in the bullfrog nerve experiment. The number of camera devices can be configured to be multiple, and multiple camera devices are used to acquire action videos of bullfrogs and action videos of operators, respectively. The physiological sensing device is used to obtain the physiological coefficients of the experimental bullfrogs. For example, the physiological sensing device can include a heart rate sensor, which monitors the heart rate of the experimental bullfrog through a detection electrode patch installed on the surface of the experimental bullfrog. The physiological sensing device can also include existing devices for detecting other physical signs of the experimental bullfrog. The environmental sensing device can include a temperature and humidity sensor, which is used to acquire the environmental temperature and humidity during the bullfrog nerve experiment.

[0062] The data processing module 200 is signal connected with the data acquisition module 100, configured to extract key frames from the action videos, extract abnormal behaviors based on the key frames, and generate evaluation data based on the abnormal behaviors. In this example, the step of extracting key frames specifically includes: adopting Clustering algorithm to obtain keyframes in the action video: Extracting abnormal behavior based on the keyframes includes: extracting abnormal behavior based on histogram entropy value as the total discriminative feature of abnormal behavior.

[0063] in, Clustering algorithms belong to unsupervised classification algorithms. An improvement on the mean-based algorithm, this method does not require prior knowledge; instead, it determines cluster centers through iterative computation. Fuzzy iterative self-organizing clustering is used to cluster video image frames and select keyframes because it pre-sets iteration parameters, cluster centers, and the number of clusters. Clusters are automatically merged and split based on similarity criteria, effectively avoiding the randomness of cluster center selection in ordinary clustering methods.

[0064] Using histogram entropy as the overall discriminative feature for different bullfrog behaviors, the fundamental characteristic that histogram twisting or translating does not change the behavioral characteristics, can effectively reflect changes in bullfrog movement. Feature extraction of bullfrog behavior is performed, and the extracted features are divided into equal-sized intervals using the oriented gradient histogram method. The extracted feature values ​​are then projected into these intervals to obtain the histogram distribution of each bullfrog movement behavior. Research shows that during normal bullfrog movement, the direction and speed of the bullfrog's movement do not change significantly, remaining within a certain range. However, targets exhibiting abnormal behavior show significant changes in both direction and speed. Therefore, calculating the histogram entropy value based on the histogram feature distribution can detect whether abnormal behavior has occurred.

[0065] The use Clustering algorithms are used to obtain keyframes from bullfrog behavior videos, including:

[0066] Step 110: Define the video image frame sample as , Define the number of cluster categories. and the initial membership matrix ;

[0067] Step 120: Calculate the cluster centers using the following formula based on the results of the first iteration:

[0068]

[0069] in, For the first The first iteration of the class contains elements like... For the first In the results of the next iteration Belonging to the Membership values ​​of the classes; the similarity between the cluster centers and each sample is obtained using Euclidean distance:

[0070]

[0071] If , its membership equals 1, while the membership of other categories is 0; if , the first membership matrix is calculated as follows:

[0072]

[0073] Step 130: define as the weight index, if , ; if , ; compare with , if , as the membership matrix after the first iteration, when the above two conditions are met, the iteration stops, otherwise jump to step 120, wherein is a minimum positive value;

[0074] Step 140: calculate the total average distance of all samples in each cluster domain to the corresponding cluster center and the average distance of each sample to the cluster center :

[0075]

[0076]

[0077] When the number of cluster centers is less than or equal to half of the defined value, i.e. , split the existing cluster center;

[0078] Calculate the standard deviation vector of sample clustering in each cluster , wherein the number of clusters is ;

[0079] Calculate the maximum component value of , if and one of the two conditions is met: or , its corresponding cluster center is split into two new cluster centers and ;

[0080] ​When the new cluster center and The distance is less than In this case, the two cluster centers will be merged;

[0081] Step 150: If Clustering ends; otherwise, proceed to step 120.

[0082] Clustering involves seven initial variables with numerous and difficult-to-determine parameter values. These parameters interact with each other, and their values ​​are also related to the sample set for clustering. After multiple experiments and comparisons, a suitable initial setting was determined. (Sample) In Membership matrix Expected number of cluster centers Number of iterations Minimal positive value Minimum cluster center distance Standard deviation Number of cluster centers Therefore, based on the above steps, Clustering methods are used to obtain keyframes from a video.

[0083] The extraction of anomalous behaviors based on histogram entropy as the overall feature for discriminating anomalous behavior includes:

[0084] Define the histogram model as follows:

[0085] in, Indicates the first frame, The number of intervals in the histogram.

[0086]

[0087] In the formula Indicates the first The first frame of the video image A behavioral characteristic, Indicates the first The weight of each behavioral feature Indicates the first The feature value range corresponding to each behavioral feature Indicates the interval number of the histogram. Represents the Kronecker function;

[0088] The entropy of an image is defined as:

[0089]

[0090] Obtained The greater the value, the greater the change in the motion behavior of the bullfrog; since abnormal behavior occurs in a time period rather than at a moment, the feature information in a time period should be considered to avoid the influence of the mutation of the entropy value of a frame on the detection result. Therefore, the histogram fusion value of the key image frame obtained through clustering is used as the final discrimination feature.

[0091] The entropy value calculation formula is defined as:

[0092]

[0093] When abnormal behavior is determined, otherwise, normal behavior is determined.

[0094] The generating of the examination data based on the abnormal behavior includes: based on the pre-written operation program, the same or different weight values are given to the abnormal behavior, the abnormal behavior occurring in the first bullfrog neural experiment is weighted and summed based on the weight values, and the examination deduction coefficient is obtained. Teachers can determine whether the students pass the test score line in experimental teaching based on whether the examination deduction coefficient reaches the preset threshold.

[0095] The data output module 300 includes a printer / display. The display can be used to directly display the examination deduction coefficient, and the printer is used to print the examination deduction coefficient as a paper document.

[0096] The embodiments of the present application are described in detail above in combination with the drawings, but the present application is not limited to the above-described embodiments. Even if various changes are made to the present application, as long as the changes fall within the scope of the present application and equivalent technologies, they still fall within the protection scope of the present application.

Claims

1. A testing system for bullfrog neural experiments, characterized in that, The system comprises: a data acquisition module for acquiring experimental data of bullfrog neural experiments; a data processing module for processing and analyzing the experimental data and generating assessment data; a data output module for outputting the assessment data.

2. The assessment system of the bullfrog neural experiment according to claim 1, wherein the data acquisition module comprises: a camera for acquiring action videos in the bullfrog neural experiment; the action videos include action videos of the bullfrog and action videos of the operator the data processing module is configured to extract key frames from the action videos, extract abnormal behaviors based on the key frames, and generate assessment data based on the abnormal behaviors.

3. The assessment system of the bullfrog neural experiment according to claim 2, wherein the extraction of the abnormal behaviors based on the key frames comprises: extracting the abnormal behaviors based on histogram entropy values as the total features for distinguishing the abnormal behaviors. The extracting key frames comprises: The clustering algorithm obtains the key frames in the action video:

4. The assessment system of the bullfrog neural experiment according to claim 3, wherein step 120: the first iteration result is calculated according to the following formula to obtain the clustering center:

5. The assessment system of the bullfrog neural experiment according to claim 4, wherein the extraction of the abnormal behaviors based on the histogram entropy values as the total features for distinguishing the abnormal behaviors comprises: The adoption The clustering algorithm obtains the key frame of the bullfrog behavior video, which comprises: Step 110: defining the video image frame samples as , ; defining the number of cluster classes and the initial membership matrix ; defining the entropy of an image as: wherein, is the first class initial iteration contains image elements, is the first iteration calculation result belongs to the first class membership value; the degree of similarity between the cluster center and each sample is obtained by the Euclidean distance: If its membership is equal to 1, while its membership to other classes is 0; if , the first membership matrix is calculated as follows: ； Step 130: Definition For weighted index, if , ;like , ;Compare and ,like , As the first After the next iteration, the membership matrix satisfies the above two conditions; otherwise, the iteration stops. It is a very small positive value; Step 140: Calculate each cluster field the total average distance of all samples to the corresponding cluster center and the average distance of each sample to the cluster center : When the number of cluster centers is less than or equal to half of the defined value, i.e. the existing cluster centers are split. calculating a standard deviation vector of the sample clusters in each cluster where the number of clusters dimension is ; The maximum component value of the cluster center is calculated The maximum component value of the cluster center is calculated If the following conditions are satisfied And one of the following two conditions is satisfied: Or The corresponding cluster center Is split into two new cluster centers And ; When the distance of the new cluster center and is less than then the two cluster centers are merged; Step 150: If , clustering is finished; otherwise go to step 120. defining the entropy value calculation formula as:

6. The assessment system of the bullfrog neural experiment according to claim 3, wherein the generation of the assessment data based on the abnormal behaviors comprises: The histogram model is defined as: wherein, represents the frame, is the number of histogram bins, In the formula represents the first frame video image behavior feature, represents the weight of the first behavior feature, represents the feature value interval corresponding to the first behavior feature, represents the histogram interval number, represents the Kronecker function; assigning a weight value to the abnormal behaviors, performing weighted summation on the abnormal behaviors in a bullfrog neural experiment based on the weight values, and obtaining an assessment deduction coefficient. obtained The greater the value, the greater the change in locomotor behavior of the bullfrog; 7. The assessment system of the bullfrog neural experiment according to claim 1, wherein the data acquisition module further comprises: When then the abnormal behavior is determined to have occurred, and otherwise, normal behavior is determined. a heart rate sensor for monitoring the heart rate of the experimental bullfrog; a temperature and humidity sensor for acquiring the environmental temperature and humidity of the bullfrog neural experiment.

8. The assessment system of the bullfrog neural experiment according to claim 1, wherein the data output module comprises a printer / display. ​ ​ ​ ​ ​

Images