A statistical index classification method and system based on tactile sensation experiments

Abstract

This invention discloses a statistical index classification method and system based on tactile sensation experiments. The method includes: collecting pressure data from users through multiple tactile sensation experiments; each tactile sensation experiment includes multiple samplings; extracting features from the pressure data; calculating statistical indices of the pressure data features; selecting significant difference indicators from the statistical indices; and inputting the significant difference indicators into a trained support vector machine classification model for classification. This invention selects objective and significantly different statistical indicators, which, when used for auxiliary assessment of children with ADHD, can obtain more accurate and objective classification results.

Description

Technical Field

[0001] This invention relates to the field of behavioral analysis technology, and in particular to a statistical index classification method and system based on tactile sensation experiments. Background Technology

[0002] Current methods for assessing children with Attention-Deficit Hyperactivity Disorder (ADHD) primarily rely on subjective descriptions from parents or teachers, using clinical interviews, questionnaires, and scales for diagnosis. These methods are subjective, lack appropriate objective indicators, and require highly qualified physicians with extensive clinical experience. Summary of the Invention

[0003] The purpose of this invention is to provide a statistical index classification method and system based on tactile sensation experiments, which is used to select objective and significantly different statistical indicators for auxiliary assessment of children with ADHD, so as to obtain more accurate and objective classification results.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A statistical index classification method based on tactile sensation experiments, including:

[0006] Users' pressure data were collected through multiple tactile sensation experiments; each tactile sensation experiment included multiple samplings.

[0007] Extract the features of the pressure data; the features include: the duration of each stage of each tactile sensation experiment, the success or failure of each tactile sensation experiment, and the pressure characteristics; the pressure characteristics include overshoot and force amplitude.

[0008] Calculate statistical indicators of the characteristics of the pressure data; the statistical indicators include: mean, variance, standard deviation, coefficient of variation, and the ratio of standard deviation to median;

[0009] Significant difference indicators are selected from the statistical indicators; the significant difference indicators are statistical indicators whose difference is less than a difference threshold; the difference is calculated using a paired t-test;

[0010] The significant difference index is input into the trained support vector machine classification model for classification.

[0011] Optionally, after collecting the user's pressure data through multiple tactile experiments, the method also includes:

[0012] The pressure data is preprocessed; specifically, the pressure data in each tactile sensation experiment where the previous sample was greater than the pressure threshold and the subsequent sample was less than the pressure threshold is replaced with the pressure data from the previous sample.

[0013] Optionally, features of the pressure data are extracted, specifically including:

[0014] Each tactile sensation experiment was divided into three stages, and the duration of each stage was recorded.

[0015] Calculate the maximum and minimum values ​​of the pressure data at different stages;

[0016] Subtract the target pressing force from the maximum value to obtain the overshoot;

[0017] The force amplitude is obtained by subtracting the minimum value from the maximum value.

[0018] The success or failure of each tactile sensation experiment was recorded.

[0019] Optionally, the training process of the support vector machine classification model is as follows:

[0020] Obtain the dataset; the dataset includes normal statistical indicators and abnormal statistical indicators;

[0021] The dataset is divided into a training set and a test set;

[0022] The support vector machine classification model is trained using the training set.

[0023] The trained support vector machine classification model is tested using the test set.

[0024] Optionally, the training set and test set are randomly divided in the dataset using the ten-fold cross-validation method.

[0025] This invention also provides a statistical index classification system based on tactile sensation experiments, including:

[0026] The pressure data acquisition module is used to collect the user's pressure data through multiple tactile experiments; each tactile experiment includes multiple samplings.

[0027] The feature extraction module is used to extract features from the pressure data; the features include: the duration of each stage of each tactile sensation experiment, the success or failure of each tactile sensation experiment, and pressure characteristics; the pressure characteristics include overshoot and force amplitude.

[0028] The statistical indicator calculation module is used to calculate statistical indicators of the characteristics of the pressure data; the statistical indicators include: mean, variance, standard deviation, coefficient of variation, and the ratio of standard deviation to median;

[0029] The indicator selection module is used to select significant difference indicators from the statistical indicators; the significant difference indicators are statistical indicators whose difference is less than a difference threshold; the difference is calculated using a paired t-test;

[0030] The indicator classification module is used to input the significant difference indicators into the trained support vector machine classification model for classification.

[0031] Optionally, it also includes:

[0032] The preprocessing module is used to preprocess the pressure data; specifically, it replaces the pressure data in each tactile sensation experiment where the previous sample was greater than the pressure threshold and the subsequent sample was less than the pressure threshold with the pressure data from the previous sample.

[0033] Optionally, the feature extraction module specifically includes:

[0034] The division and statistical unit is used to divide each tactile sensory experiment into three stages and to count the time length of each stage;

[0035] The calculation unit is used to calculate the maximum and minimum values ​​of the pressure data at different stages;

[0036] An overshoot calculation unit is used to subtract the target pressing force from the maximum value to obtain the overshoot.

[0037] A force amplitude calculation unit is used to subtract the minimum value from the maximum value to obtain the force amplitude.

[0038] The statistical unit is used to determine the success or failure of each tactile sensory experiment.

[0039] Optionally, it further includes: a training module; the training module includes:

[0040] A dataset acquisition unit is used to acquire a dataset; the dataset includes normal statistical indicators and abnormal statistical indicators.

[0041] A partitioning unit is used to divide the dataset into a training set and a test set;

[0042] A training unit is used to train the support vector machine classification model using the training set;

[0043] The testing unit is used to test the trained support vector machine classification model using the test set.

[0044] Optionally, the partitioning unit uses a ten-fold cross-validation method to randomly partition the dataset into training and test sets.

[0045] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0046] This invention leverages the high temporal and spatial resolution of the tactile channel in human attention mechanisms to collect users' pressure data, extract features, and select significant indicators that can characterize the differences between two groups (healthy children and children with ADHD). This allows the data to be used for auxiliary assessment of children with ADHD, resulting in more accurate and objective classification results. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 The flowchart shows the statistical index classification method based on tactile sensation experiments provided by this invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] The purpose of this invention is to provide a statistical index classification method and system based on tactile sensation experiments, which is used to select objective and significantly different statistical indicators for auxiliary assessment of children with ADHD, so as to obtain more accurate and objective classification results.

[0051] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Example 1

[0053] This invention provides a statistical index classification method based on tactile sensation experiments, such as... Figure 1 As shown, the method includes the following steps:

[0054] Step 101: Collect the user's pressure data through multiple tactile sensation experiments; each tactile sensation experiment includes multiple samplings.

[0055] In practical applications, the pressure applied is displayed in real-time via a virtual reality headset and a computer screen. The pressure level is represented by the height of a moving disc, while the target pressure value and allowable error range are represented by a cylinder of a certain height. Only one finger is pressed at a time, and only one target pressure cylinder is displayed. The virtual reality headset provides a better sense of immersion for the user and reduces distractions from surrounding elements. Data on pressure changes over time is collected at a frequency of 50Hz. The experiment consists of three tasks, which the user must perform sequentially. All three tasks have a fixed duration set for each trial, and the trial duration must not exceed the set duration. The first two tasks also have a set duration for which the user must maintain the pressure within the allowable error range for that duration; otherwise, the trial is considered successful. The two tasks have different set durations. The third task does not have a pre-set duration; the user must maintain the pressure as long as possible until the trial ends.

[0056] The collected pressure data is preprocessed by setting a threshold force value for starting to press. Then, pressure data that were greater than the threshold in the previous sampling and suddenly less than the threshold in the next sampling are replaced with the data from the previous sampling to remove data with abnormally decreased pressure.

[0057] Step 102: Extract the features of the pressure data; the features include: the duration of each stage of each tactile sensation experiment, the success or failure of each tactile sensation experiment, and the pressure features; the pressure features include overshoot and force amplitude.

[0058] Specifically, this includes: dividing each tactile sensation experiment into three stages and calculating the time length of each stage; calculating the maximum and minimum values ​​of the pressure data in different stages; subtracting the target pressure from the maximum value to obtain the overshoot; subtracting the minimum value from the maximum value to obtain the force amplitude; and calculating the success or failure of each tactile sensation experiment.

[0059] In practical applications, the collected pressure data for each press is first divided into three stages, based on the following criteria: The first stage is the initial pressing process, starting when the target cylinder appears and ending when the pressure reaches the pre-set effective pressure threshold. The second stage is the pressure increase stage, starting at the end of the first stage and ending when the pressure stabilizes within the allowable error range near the target value and is maintained. The third stage is the stabilization stage, starting at the end of the second stage. The end time is discussed under two experimental scenarios. For the first two tasks, the experiment is considered successful when the pressure reaches the required time, and the single trial ends; the successful moment is recorded as the end time of the third stage. For the third task, the end time of the trial is recorded as the end time of the third stage. Then, force amplitude features are extracted within and between stages. First, the maximum and minimum pressure values ​​are calculated in different time stages. The overshoot is obtained by subtracting the target pressure from the maximum value, and the force amplitude is obtained by subtracting the minimum value from the maximum value.

[0060] Step 103: Calculate the statistical indicators of the pressure data characteristics; the statistical indicators include: mean, variance, standard deviation, coefficient of variation, and the ratio of standard deviation to median.

[0061] Step 104: Select significant difference indicators from the statistical indicators; the significant difference indicators are statistical indicators whose difference is less than the difference threshold; the difference is calculated using a paired t-test.

[0062] In practical applications, pressure data characteristics were collected from two groups (healthy children and children with ADHD). For each characteristic, the characteristic values ​​were calculated for an equal number of normal children and children with ADHD, forming two sets of data. The differences between the two sets of data were calculated. The p-value was used as a representation of the differences, and statistical indicators with p-values ​​less than a set threshold were selected as indicators of significant differences for classification.

[0063] Step 105: Input the significant difference index into the trained support vector machine classification model for classification.

[0064] The process of training a Support Vector Machine (SVM) classification model is as follows:

[0065] Statistical indicators were calculated for the stress data characteristics of two groups (healthy children and children with ADHD), and labels were assigned, including two categories: normal indicators and abnormal indicators. The dataset was randomly divided into training and test sets using a 10-fold cross-validation method. A support vector machine classification model was trained using the training set. Classification results matching the labels were considered correct, and those matching were considered incorrect. The average of these results was used as the accuracy of the classification model. Training and testing were repeated multiple times, and the average classification accuracy was calculated.

[0066] This invention leverages the high temporal and spatial resolution of the tactile channel in human attention mechanisms to collect users' pressure data, extract features, and select significant indicators that can characterize the differences between two groups (healthy children and children with ADHD). This allows the data to be used for auxiliary assessment of children with ADHD, resulting in more accurate and objective classification results.

[0067] Example 2

[0068] In order to implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a statistical index classification system based on tactile sensation experiment is provided below.

[0069] The system includes:

[0070] The pressure data acquisition module is used to collect the user's pressure data through multiple tactile experiments; each tactile experiment includes multiple samplings.

[0071] The feature extraction module is used to extract features from the pressure data. These features include: the duration of each stage of each tactile sensation experiment, the success or failure of each experiment, and pressure characteristics. The pressure characteristics include overshoot and force amplitude. Specifically, it includes: a division and statistics unit for dividing each tactile sensation experiment into three stages and calculating the duration of each stage; a calculation unit for calculating the maximum and minimum values ​​of the pressure data within different stages; an overshoot calculation unit for subtracting the target pressure from the maximum value to obtain the overshoot; a force amplitude calculation unit for subtracting the minimum value from the maximum value to obtain the force amplitude; and a statistics unit for statistically analyzing the success or failure of each tactile sensation experiment.

[0072] The statistical indicator calculation module is used to calculate statistical indicators of the characteristics of the pressure data; the statistical indicators include: mean, variance, standard deviation, coefficient of variation, and the ratio of standard deviation to median.

[0073] The indicator selection module is used to select significant difference indicators from the statistical indicators; the significant difference indicators are statistical indicators whose difference is less than the difference threshold; the difference is calculated using a paired t-test.

[0074] The indicator classification module is used to input the significant difference indicators into the trained support vector machine classification model for classification.

[0075] The system also includes:

[0076] The preprocessing module is used to preprocess the pressure data; specifically, it replaces the pressure data in each tactile sensation experiment where the previous sample was greater than the pressure threshold and the subsequent sample was less than the pressure threshold with the pressure data from the previous sample.

[0077] The training module includes: a dataset acquisition unit for acquiring a dataset, the dataset including normal statistical indicators and abnormal statistical indicators; a partitioning unit for partitioning the dataset into a training set and a test set; a training unit for training the support vector machine classification model using the training set; and a testing unit for testing the trained support vector machine classification model using the test set.

[0078] The partitioning unit uses a ten-fold cross-validation method to randomly divide the dataset into training and test sets.

[0079] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0080] This article uses specific examples to illustrate the principles and implementation methods of the invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. The described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

Claims

1. A statistical index classification method based on tactile sensation experiments, characterized in that, include: Data on the pressure applied by users was collected through multiple tactile experiments. Each tactile sensory experiment involves multiple samplings. Specifically, it includes: real-time display of pressing pressure via a virtual reality headset and computer screen; the magnitude of the pressing pressure is represented by the height of a moving disc; the target pressing pressure value and allowable error range are represented by a cylinder of a certain height; only one finger is pressed at a time, and only one target pressing pressure cylinder is displayed; data on the change of pressing pressure over time is collected at a frequency of 50Hz; the tactile sensory experiment consists of three tasks, which the user must perform in sequence; all three tasks have a fixed duration set for each trial, and the trial duration must not exceed the set duration; the first two tasks also have a set duration for which the user needs to maintain the pressing pressure within the allowable error range for the specified duration, which is considered a successful trial; otherwise, failure to reach the specified duration within the fixed trial duration is considered a failure; the two tasks have different set durations, and the third task does not have a pre-set duration. Extracting features from the pressure data; these features include: the duration of each stage of each tactile sensation experiment, the success or failure of each tactile sensation experiment, and pressure characteristics; the pressure characteristics include overshoot and force amplitude; specifically, this includes: dividing each tactile sensation experiment into three stages and calculating the duration of each stage; the first stage is the initial pressing process, the second stage is the pressure increase stage, and the third stage is the stabilization stage; calculating the maximum and minimum values ​​of the pressure data within different stages; subtracting the target pressure from the maximum value to obtain the overshoot; subtracting the minimum value from the maximum value to obtain the force amplitude; and calculating the success or failure of each tactile sensation experiment. Calculate statistical indicators of the characteristics of the pressure data; the statistical indicators include: mean, variance, standard deviation, coefficient of variation, and the ratio of standard deviation to median; Significant difference indicators are selected from the statistical indicators; the significant difference indicators are statistical indicators whose difference is less than a difference threshold; the difference is calculated using a paired t-test; The significant difference index is input into the trained support vector machine classification model for classification. The training process of the support vector machine classification model is as follows: obtain the dataset; the dataset includes normal statistical indicators and abnormal statistical indicators; divide the dataset into a training set and a test set; train the support vector machine classification model using the training set; and test the trained support vector machine classification model using the test set.

2. The statistical index classification method based on tactile sensation experiment according to claim 1, characterized in that, After collecting users' pressure data through multiple tactile experiments, it also includes: The pressure data is preprocessed; specifically, the pressure data in each tactile sensation experiment where the previous sample was greater than the pressure threshold and the subsequent sample was less than the pressure threshold is replaced with the pressure data from the previous sample.

3. The statistical index classification method based on tactile sensation experiment according to claim 1, characterized in that, The training and test sets in the dataset are randomly divided using the 10-fold cross-validation method.

4. A statistical index classification system based on tactile sensation experiments, characterized in that, include: The pressure data acquisition module is used to collect the user's pressure data through multiple tactile experiments. Each tactile sensory experiment involves multiple samplings. Specifically, it includes: real-time display of pressing pressure via a virtual reality headset and computer screen; the magnitude of the pressing pressure is represented by the height of a moving disc; the target pressing pressure value and allowable error range are represented by a cylinder of a certain height; only one finger is pressed at a time, and only one target pressing pressure cylinder is displayed; data on the change of pressing pressure over time is collected at a frequency of 50Hz; the tactile sensory experiment consists of three tasks, which the user must perform in sequence; all three tasks have a fixed duration set for each trial, and the trial duration must not exceed the set duration; the first two tasks also have a set duration for which the user needs to maintain the pressing pressure within the allowable error range for the specified duration, which is considered a successful trial; otherwise, failure to reach the specified duration within the fixed trial duration is considered a failure; the two tasks have different set durations, and the third task does not have a pre-set duration. A feature extraction module is used to extract features from the pressure data. These features include: the duration of each stage of each tactile sensation experiment, the success or failure of each experiment, and pressure characteristics. The pressure characteristics include overshoot and force amplitude. The module includes: a division and statistics unit for dividing each tactile sensation experiment into three stages and calculating the duration of each stage; the first stage is the initial pressing process, the second stage is the pressure increase stage, and the third stage is the stabilization stage; a calculation unit for calculating the maximum and minimum values ​​of the pressure data within different stages; an overshoot calculation unit for subtracting the target pressure from the maximum value to obtain the overshoot; a force amplitude calculation unit for subtracting the minimum value from the maximum value to obtain the force amplitude; and a statistics unit for calculating the success or failure of each tactile sensation experiment. The statistical indicator calculation module is used to calculate statistical indicators of the characteristics of the pressure data; the statistical indicators include: mean, variance, standard deviation, coefficient of variation, and the ratio of standard deviation to median; The indicator selection module is used to select significant difference indicators from the statistical indicators; the significant difference indicators are statistical indicators whose difference is less than a difference threshold; the difference is calculated using a paired t-test; The indicator classification module is used to input the significant difference indicators into the trained support vector machine classification model for classification. It also includes: a training module; the training module includes: A dataset acquisition unit is used to acquire a dataset; the dataset includes normal statistical indicators and abnormal statistical indicators. A partitioning unit is used to divide the dataset into a training set and a test set; A training unit is used to train the support vector machine classification model using the training set; The testing unit is used to test the trained support vector machine classification model using the test set.

5. The statistical index classification system based on tactile sensation experiment according to claim 4, characterized in that, Also includes: The preprocessing module is used to preprocess the pressure data; Specifically: Replace the pressure data in each tactile sensation experiment where the previous sample was greater than the pressure threshold and the subsequent sample was less than the pressure threshold with the pressure data from the previous sample.

6. The statistical index classification system based on tactile sensation experiment according to claim 4, characterized in that, The partitioning unit uses a ten-fold cross-validation method to randomly divide the dataset into training and test sets.

Images