Analysis method for auxiliary diagnosis and treatment of autistic children

Abstract

The invention relates to the field of auxiliary diagnosis and treatment, in particular to an analysis method for auxiliary diagnosis and treatment of autistic children, which comprises the following steps: pre-detecting a target object and storing detection data so as to analyze the action characteristic representation quantity of the target object, determine the action characteristic feedback degree of the target object and perform auxiliary diagnosis and treatment analysis. Comprising the following steps: adjusting the image acquisition precision of an image acquisition device for a target object, calling a video record in a target object review case, and identifying an emotion fixation feature, a limb movement amplitude and a change ratio of a feature matching parameter relative to pre-detection based on the video record, comparing the data with expected contrast data determined based on a pre-stored intervention diagnosis and treatment scheme, and judging whether a feature matching standard is met or not; the recommended review times are generated based on the action feature representation quantity, the auxiliary diagnosis and treatment analysis result is output through the output end, the data analysis result is high in accuracy and reliability, auxiliary diagnosis and treatment data support can be provided for doctors, and the diagnosis and treatment efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of auxiliary diagnosis and treatment, and in particular to an analysis method for auxiliary diagnosis and treatment of autistic children. Background Art

[0002] With the development of computer vision and artificial intelligence technology, visual behavioral analysis is gaining more and more attention in assisting diagnosis of autism. By analyzing the patient's behavioral video data and extracting features for identification and diagnosis, it has the advantages of being non-interventional and concealed. At the same time, studies have shown that early intervention can significantly improve the social and communication skills of children with autism and improve their quality of life. Therefore, timely diagnosis and intervention of children with autism in the early stages of discovery have become an important part of clinical practice.

[0003] Chinese patent publication number: CN115578670A, discloses a method, device and diagnosis and treatment machine-assisted system for behavior recognition and motion prediction of autistic children, the method comprising: (1) acquiring original skeleton video data of the child from different angles; (2) inputting the acquired original skeleton video data into a view adaptive conversion unit to adaptively convert it into a consistent coordinate system to obtain an optimal representation of the human skeleton; (3) inputting the optimal representation of the human skeleton into a multi-scale feature extraction unit to extract high-performance spatiotemporal features of the human skeleton; (4) inputting the high-performance spatiotemporal features into a multi-task learning unit to perform behavior recognition and motion prediction on the human body, and obtaining behavior classification results and motion prediction results.

[0004] However, the existing technology still has the following problems: it only considers the motion characteristics, but not the gaze characteristics of the target object; and it does not adaptively adjust the data evaluation method according to the diagnosis and treatment plan to provide data support for doctors, resulting in poor auxiliary diagnosis and treatment effects. Summary of the invention

[0005] To this end, the present invention provides an analysis method for auxiliary diagnosis and treatment of autistic children, so as to overcome the problems in the prior art of only considering movement characteristics but not considering the gaze characteristics of the target object, and failing to adaptively adjust the data evaluation method according to the diagnosis and treatment plan to provide data support for doctors, resulting in poor auxiliary diagnosis and treatment effects.

[0006] To achieve the above object, the present invention provides an analysis method for auxiliary diagnosis and treatment of autistic children, comprising:

[0007] Step S1, pre-detecting the target object and storing detection data for the target object, wherein the pre-detection includes setting an image acquisition device to detect the action characteristics of the target object when it performs a communication action, wherein the action characteristics include the amplitude of the limb movement and the reaction time;

[0008] Step S2, analyzing the motion feature representation of the target object based on the motion feature to determine the motion feature feedback degree of the target object;

[0009] Step S3, based on the feedback degree of the target object's motion characteristics, performing auxiliary diagnosis and treatment analysis, including:

[0010] Adjusting the image acquisition accuracy of the image acquisition device for the target object, calling the video record in the target object's review case, identifying the emotional gaze features, the amplitude of body movements, and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standard is met;

[0011] Or, generating a recommended number of review times based on the action feature representation;

[0012] The feature matching parameter is determined by combining the ratio of the emotional gaze sub-features with the amplitude of the body movement, and the emotional gaze features include the gaze duration of the eye area and the gaze duration of the mouth area;

[0013] Step S4, outputting the auxiliary diagnosis and treatment analysis result through the output end.

[0014] Furthermore, in step S2, the action feature representation of the target object is parsed according to formula (1) based on the action feature, including:

[0015]

[0016] In formula (1), S represents the action feature representation, F represents the limb movement amplitude, F0 represents the limb movement amplitude threshold, T represents the reaction time, T0 represents the reaction time threshold, α represents the limb movement amplitude weight coefficient, and β represents the reaction time weight coefficient.

[0017] Furthermore, in step S2, determining the action feature feedback degree of the target object includes:

[0018] If the action feature representation amount is greater than or equal to the action feature representation amount threshold, the action feature of the target object is determined to be at a weak feedback level;

[0019] If the motion feature representation amount is less than the motion feature representation amount threshold, the motion feature of the target object is determined to be at a strong feedback level.

[0020] Furthermore, in step S3, auxiliary diagnosis and treatment analysis is performed based on the feedback degree of the target object's motion characteristics, including:

[0021] If the target object's motion feature is a weak feedback degree, the image acquisition accuracy of the image acquisition device for the target object is adjusted, the video record in the target object's review case is called, and the emotional gaze feature, the body movement amplitude, and the change ratio of the feature matching parameter relative to the pre-detection are identified based on the video record, and compared with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standard is met;

[0022] If the action feature of the target object is a strong feedback level, a recommended number of reviews is generated based on the action feature representation.

[0023] Furthermore, in step S3, adjusting the image acquisition accuracy of the image acquisition device for the target object includes:

[0024] Improve the resolution and maximum frame rate of the image acquisition device.

[0025] Furthermore, in step S3, the process of determining the feature matching parameter by calculating the ratio of the emotional gaze sub-features in combination with the body movement amplitude includes:

[0026] Obtaining video records captured by an image acquisition device to determine emotional gaze sub-features and body movement amplitudes;

[0027] The ratio of the fixation time of the eye area to the fixation time of the mouth area is used as the ratio of the emotion fixation sub-feature and as the first feature;

[0028] Calculate the ratio of the limb movement amplitude to the expected threshold of the limb movement amplitude as the second feature;

[0029] The sum of the first feature and the second feature is determined as a feature matching parameter.

[0030] Furthermore, in step S3, the process of identifying the emotional gaze features, the body movement amplitude, and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan includes:

[0031] Determine the change ratio of emotional gaze features, the change ratio of body movement amplitude, and the change ratio of feature matching parameters;

[0032] Determine the expected emotional gaze feature change ratio, expected limb movement amplitude change ratio, and expected feature matching parameter change ratio corresponding to the intervention diagnosis and treatment plan;

[0033] comparing the emotion gaze feature change ratio with an expected emotion gaze feature change ratio;

[0034] comparing the limb movement amplitude change ratio with the expected limb movement amplitude change ratio;

[0035] The feature matching parameter change ratio is compared with the expected feature matching parameter change ratio.

[0036] Furthermore, in step S3, the determination of whether the feature matching standard is met includes:

[0037] If the expected matching conditions are met, it is determined that the feature matching standard is met;

[0038] The expected matching conditions include that the feature matching parameter change ratio is less than the expected feature matching parameter change ratio, the emotional gaze feature change ratio is less than the expected emotional gaze feature change ratio, and the limb movement amplitude change ratio is less than the expected limb movement amplitude change ratio.

[0039] Furthermore, in step S3, generating a recommended number of review times based on the action feature representation includes:

[0040] The recommended review times are positively correlated with the action feature representation quantity.

[0041] Furthermore, the step S3 also includes pre-storing the expected control data of each intervention treatment plan, including:

[0042] Expected feature matching parameter change ratio, expected emotional gaze feature change ratio, and expected limb movement amplitude change ratio.

[0043] Compared with the prior art, the present invention performs pre-detection on the target object and stores the detection data to parse the target object's motion feature representation, determine the target object's motion feature feedback degree, and perform auxiliary diagnosis and treatment analysis, including: adjusting the image acquisition accuracy of the image acquisition device for the target object, calling the video record in the target object's review case, identifying the emotional gaze characteristics, limb movement amplitude and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standards are met; generating a recommended number of re-examinations based on the motion feature representation, and outputting the results of the auxiliary diagnosis and treatment analysis through the output end. The data analysis results are accurate and reliable, can provide doctors with auxiliary diagnosis and treatment data support, and improve diagnosis and treatment efficiency.

[0044] In particular, the present invention analyzes the motion feature representation of the target object through motion feature analysis. In actual situations, the basic motion development speed of the target object is relatively slow, and there will be problems of limb movement incoordination and movement disorders. By observing the limb movement amplitude of the target object when it generates communication movements, for example, the limb amplitude presented by the object who needs to be diagnosed and treated will be relatively small. This situation may be related to motor skill disorders, which are usually manifested as early motor lag, coordination disorders, physical decline and visual-motor integration disorders. At the same time, the target object needs to communicate or make corresponding responses. The reaction time required is longer. This situation may be related to its obstacles in social interaction and communication. They need more time to process and respond to social signals. Therefore, the present application analyzes the motion feature representation of the target object through the limb movement amplitude and reaction time when the target object generates communication movements, so as to characterize the degree of feedback that the target object can give when dealing with interaction and communication, and provide data support for the subsequent determination of the target object's motion feature feedback degree, so as to adaptively perform auxiliary diagnosis and treatment analysis, and then ensure the accuracy and reliability of the data analysis results, provide auxiliary diagnosis and treatment data support for doctors, and improve the efficiency of diagnosis and treatment.

[0045] In particular, the present invention selects different auxiliary diagnosis and treatment analysis methods according to the different levels of feedback of the action characteristics presented by the target object, including: when the action characteristics of the target object are at a weak feedback level, the present invention pre-adjusts the accuracy of the video recording of the target object's review diagnosis and treatment process, improves the resolution and maximum frame rate of image acquisition, and conducts a detailed observation of the relevant key features of the target object, and then extracts the emotional gaze sub-features and body movement amplitudes presented by the target object during interaction and communication in the high-precision video image to determine the feature matching parameters. Therefore, the present application characterizes the changes in the situation and the degree of improvement of the target object after a certain period of time after the pre-detection adopts the intervention diagnosis and treatment plan through feature matching parameters, and performs an overall analysis based on the feature matching parameters while considering the analysis of the various detailed features targeted by each sub-feature, so as to not only determine the use of intervention diagnosis and treatment as a whole, but also Whether the changes in the situation after a certain period of time under the treatment plan match the expected degree of change in diagnosis and treatment and the degree of matching, and by specifically analyzing the changes in each sub-feature and the matching of the sub-feature data corresponding to the expected changes in diagnosis and treatment, the degree of specific feature matching can be relatively accurately identified, so as to provide auxiliary support for the subsequent modification of targeted treatment plans, and consider the gaze characteristics of the target object combined with further analysis of the amplitude characteristics of limb movements to improve the accuracy and reliability of auxiliary diagnosis and treatment; when the target object's movement characteristics are at a strong feedback level, the abnormal tendency of the target object's movement characteristics is slight, and the data evaluation method is adjusted comprehensively and adaptively through the observation of the target object's movement characteristics combined with the data provided by the movement feature representation. The data analysis results are accurate and reliable, which can provide doctors with auxiliary diagnosis and treatment data support and improve the efficiency of diagnosis and treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of the steps of an analysis method for auxiliary diagnosis and treatment of autistic children according to an embodiment of the invention;

[0047] Figure 2 A logical decision diagram for determining the degree of feedback of the action characteristics of a target object for an embodiment of the invention;

[0048] Figure 3 A logical decision diagram for selecting an auxiliary diagnosis and treatment analysis method for an embodiment of the invention;

[0049] Figure 4 A logical decision diagram for determining whether a feature matching standard is met according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0051] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0052] See also Figures 1 to 4 As shown, Figure 1 Schematic diagram of the steps of the analysis method for auxiliary diagnosis and treatment of autistic children according to an embodiment of the present invention, Figure 2 A logical decision diagram for determining the degree of action feature feedback of a target object in an embodiment of the invention, Figure 3 A logical decision diagram for selecting an auxiliary diagnosis and treatment analysis method for an embodiment of the invention, Figure 4 The logical decision diagram for determining whether the feature matching standard is met in the embodiment of the invention is shown in the embodiment of the invention. The analysis method for auxiliary diagnosis and treatment of autistic children in the embodiment of the invention includes:

[0053] Step S1, pre-detecting the target object and storing detection data for the target object, wherein the pre-detection includes setting an image acquisition device to detect the action characteristics of the target object when it performs a communication action, wherein the action characteristics include the amplitude of the limb movement and the reaction time;

[0054] Step S2, analyzing the motion feature representation of the target object based on the motion feature to determine the motion feature feedback degree of the target object;

[0055] Step S3, based on the feedback degree of the target object's motion characteristics, performing auxiliary diagnosis and treatment analysis, including:

[0056] Adjusting the image acquisition accuracy of the image acquisition device for the target object, calling the video record in the target object's review case, identifying the emotional gaze features, the amplitude of body movements, and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standard is met;

[0057] Or, generating a recommended number of review times based on the action feature representation;

[0058] The feature matching parameter is determined by combining the ratio of the emotional gaze sub-features with the amplitude of the body movement, and the emotional gaze features include the gaze duration of the eye area and the gaze duration of the mouth area;

[0059] Step S4, outputting the auxiliary diagnosis and treatment analysis result through the output end.

[0060] Specifically, there is no limitation on the image acquisition device, which only needs to have the function of acquiring the image of the target object, such as a photographic device. Of course, authorization is required when acquiring the image of the target object, which will not be elaborated here.

[0061] Specifically, there is no specific limitation on the method of storing the detection data for the target object. A database specifically for storing detection data and treatment plans can be established at the output end, and the detection data and treatment plans can be stored in the database. At the same time, the type of database is not limited and will not be elaborated here.

[0062] Specifically, a preliminary test is performed when the target object is interviewed. There is no limitation on the method of determining the amplitude of body movements. The amplitude of movement can be determined through video analysis methods. For example, the amplitude of movement can be represented by obtaining the average moving distance of each joint of the target object within a predetermined time during the interview. Of course, other methods can also be used, which will not be elaborated here.

[0063] It is understandable that the reaction time is the time duration from the moment the communication object sends a communication to the target object to the moment the target object responds, which will not be elaborated here.

[0064] It is understandable that the time of communication and the time of response can be determined by analyzing the mouth features, for example, by pre-training an algorithm or model that can realize the corresponding function, and importing a logic component to realize the corresponding function, which will not be elaborated here.

[0065] It can be understood that the eye region fixation duration refers to the duration of the target object's fixation on the eye region, and the mouth region fixation duration refers to the duration of the target object's fixation on the mouth region;

[0066] In clinical manifestations, typically developing autistic subjects have significantly shortened gaze durations on key emotional feature areas such as the eyes and mouth, and prefer to gaze at the mouth area and avoid gaze at the eye area.

[0067] Specifically, in step S2, the action feature representation of the target object is parsed according to formula (1) based on the action feature, including:

[0068]

[0069] In formula (1), S represents the action feature representation, F represents the limb movement amplitude, F0 represents the limb movement amplitude threshold, T represents the reaction time, T0 represents the reaction time threshold, α represents the limb movement amplitude weight coefficient, and β represents the reaction time weight coefficient.

[0070] In this embodiment, the limb movement amplitude threshold F0 and the reaction time threshold T0 are obtained by pre-measurement. By collecting historical data of relevant video data when a number of target objects diagnosed with autism generate communication actions, limb movement amplitude data and reaction time data are extracted, and the average limb movement amplitude △F and the average reaction time △T presented by the number of target objects are solved. It is set that F0 = r1×△F, T0 = r2×△T, where r1 is the first precision coefficient and r2 is the second precision coefficient. 1.1 < r1 < 1.15, 1.05 < r1 < 1.1, α is taken as 0.55, and β is taken as 0.45.

[0071] The present invention analyzes the action feature characterization quantity of the target object through action features. In actual situations, the basic action development speed of the target object is relatively slow, and there will be problems such as uncoordinated limb movements and movement disorders. By observing the limb movement amplitude when the target object generates communication actions, for example, the limb amplitude presented by the object to be treated will be relatively small, and this situation may be related to motor skill disorders. These disorders usually manifest as early motor delay, coordination disorders, physical fitness decline, and visuomotor integration disorders, etc. At the same time, when the target object is communicating or needs to make a corresponding reaction, the required reaction time is longer. This situation may be related to their disorders in social interaction and communication. They need more time to process and respond to social signals. Therefore, this application analyzes the action feature characterization quantity for the target object through the limb movement amplitude and reaction time when the target object generates communication actions, so as to characterize the feedback degree that the target object can give when treating interaction and communication, provide data support for determining the action feature feedback degree of the target object subsequently, and perform auxiliary diagnosis and treatment analysis adaptively. Furthermore, to ensure the accuracy and reliability of the data analysis results, it can provide auxiliary diagnosis and treatment data support for doctors and improve the diagnosis and treatment efficiency.

[0072] Specifically, in the step S2, determining the action feature feedback degree of the target object includes,

[0073] If the action feature characterization quantity is greater than or equal to the action feature characterization quantity threshold, it is determined that the action feature of the target object is a weak feedback degree;

[0074] If the action feature characterization quantity is less than the action feature characterization quantity threshold, it is determined that the action feature of the target object is a strong feedback degree.

[0075] The action feature characterization quantity threshold S0 is selected within the interval [1.12, 1.23].

[0076] Specifically, in the step S3, based on the action feature feedback degree of the target object, auxiliary diagnosis and treatment analysis are performed, including,

[0077] If the target object's motion feature is a weak feedback degree, the image acquisition accuracy of the image acquisition device for the target object is adjusted, the video record in the target object's review case is called, and the emotional gaze feature, the body movement amplitude, and the change ratio of the feature matching parameter relative to the pre-detection are identified based on the video record, and compared with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standard is met;

[0078] If the action feature of the target object is a strong feedback level, a recommended number of reviews is generated based on the action feature representation.

[0079] Specifically, in step S3, adjusting the image acquisition accuracy of the image acquisition device for the target object includes:

[0080] Improve the resolution and maximum frame rate of the image acquisition device.

[0081] It is understandable that during the pre-detection process, conventional high-definition resolution and conventional maximum frame rate can be used for image acquisition, where the conventional high-definition resolution is 720p (1280×720 pixels) and the conventional maximum frame rate is 30Fps. During the review of the target object with weak feedback characteristics, the resolution and maximum frame rate of the image acquisition device can be increased to 1440p (2560×1440 pixels) and the maximum frame rate can be increased to 60~120Fps to observe the reaction and action of the target object more clearly and meticulously.

[0082] Specifically, in step S3, the process of determining the feature matching parameter by calculating the ratio of the emotional gaze sub-feature and combining the body movement amplitude includes:

[0083] Obtaining video records captured by an image acquisition device to determine emotional gaze sub-features and body movement amplitudes;

[0084] The ratio of the fixation time of the eye area to the fixation time of the mouth area is used as the ratio of the emotion fixation sub-feature and as the first feature;

[0085] Calculate the ratio of the limb movement amplitude to the expected threshold of the limb movement amplitude as the second feature;

[0086] The sum of the first feature and the second feature is determined as a feature matching parameter.

[0087] In this embodiment, the feature matching parameters are determined according to the following formula:

[0088]

[0089] Where P represents the feature matching parameter, E represents the fixation duration of the eye area, Y represents the fixation duration of the mouth area, F represents the amplitude of limb movement, and Fe represents the expected threshold of the amplitude of limb movement.

[0090] Among them, the eye area fixation time E and the mouth area fixation time Y are obtained according to the video information collected by the image acquisition device during the review of the target object, and Fe is pre-set, and Fe=F0×1.15 is set.

[0091] Specifically, in step S3, the process of identifying the emotional gaze features, the body movement amplitude, and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan includes:

[0092] Determine the change ratio of emotional gaze features, the change ratio of body movement amplitude, and the change ratio of feature matching parameters;

[0093] Determine the expected emotional gaze feature change ratio, expected limb movement amplitude change ratio, and expected feature matching parameter change ratio corresponding to the intervention diagnosis and treatment plan;

[0094] comparing the emotion gaze feature change ratio with an expected emotion gaze feature change ratio;

[0095] comparing the limb movement amplitude change ratio with the expected limb movement amplitude change ratio;

[0096] comparing the feature matching parameter change ratio with an expected feature matching parameter change ratio;

[0097] It can be understood that the change ratio is the ratio of the change amount to the initial amount;

[0098] The emotional gaze feature change ratio is the mean of the eye gaze feature change ratio and the mouth gaze feature change ratio.

[0099] Specifically, in step S3, the determination of whether the feature matching standard is met includes:

[0100] If the expected matching conditions are met, it is determined that the feature matching standard is met;

[0101] The expected matching conditions include that the feature matching parameter change ratio is less than the expected feature matching parameter change ratio, the emotional gaze feature change ratio is less than the expected emotional gaze feature change ratio, and the limb movement amplitude change ratio is less than the expected limb movement amplitude change ratio.

[0102] Specifically, in the step S3, generating a recommended number of review times based on the action feature representation includes:

[0103] The recommended number of reexaminations is positively correlated with the action feature characterization quantity.

[0104] In this embodiment, optionally,

[0105] Compare the action feature characterization quantity S with the first action feature characterization quantity comparison threshold S1 and the second action feature characterization quantity comparison threshold S2.

[0106] If S > S2, then determine it as the first recommended number of reexaminations c1, and set c1 = 4c0;

[0107] If S1 ≤ S ≤ S2, then determine it as the second recommended number of reexaminations c2, and set c2 = 2c0;

[0108] If S < S1, then determine the precision improvement value as the third precision improvement value c3, and set c3 = [1.5c0];

[0109] Among them, c0 represents the benchmark number of reexaminations, S1 = 1.2S0, and S2 = 1.4S0.

[0110] It can be understood that the benchmark number of reexaminations is determined based on historical prior data, which will not be elaborated here.

[0111] The present invention selects different auxiliary diagnosis and treatment analysis methods according to the different degrees of feedback of the action characteristics presented by the target object, including: when the action characteristics of the target object are at a weak feedback degree, the present invention pre-adjusts the accuracy of the video recording of the target object's review diagnosis and treatment process, improves the resolution and maximum frame rate of image acquisition, and conducts a detailed observation of the relevant key features of the target object, and then extracts the emotional gaze sub-features and body movement amplitudes presented by the target object during interaction and communication in the high-precision video image to determine the feature matching parameters. Therefore, the present application characterizes the changes in the situation and the degree of improvement of the target object after a certain period of time after the pre-detection adopts the intervention diagnosis and treatment plan through the feature matching parameters, and performs an overall analysis based on the feature matching parameters while considering the analysis of the various detailed features targeted by each sub-feature, which not only determines the use of the intervention diagnosis and treatment plan as a whole Whether the changes in the case after a certain period of time match the expected degree of change in diagnosis and treatment, and the degree of matching, and by specifically analyzing the changes in each sub-feature and the matching of the sub-feature data corresponding to the expected changes in diagnosis and treatment, the degree of specific feature matching can be relatively accurately identified, so as to provide auxiliary support for the subsequent modification of targeted diagnosis and treatment plans, and consider the gaze characteristics of the target object combined with further analysis of the amplitude characteristics of limb movements to improve the accuracy and reliability of auxiliary diagnosis and treatment; when the target object's movement characteristics are at a strong feedback level, the abnormal tendency of the target object's movement characteristics is slight, and the data evaluation method is adjusted comprehensively and adaptively through the observation of the target object's movement characteristics combined with the data provided by the movement feature representation. The data analysis results are accurate and reliable, which can provide doctors with auxiliary diagnosis and treatment data support and improve the efficiency of diagnosis and treatment.

[0112] Specifically, the step S3 also includes pre-storing expected control data for each intervention treatment plan, including.

[0113] Expected feature matching parameter change ratio, expected emotional gaze feature change ratio, and expected limb movement amplitude change ratio.

[0114] It will be appreciated that the expected control data for each intervention treatment regimen are determined based on historical a priori data, including.

[0115] Record the test data of several target subjects after treatment with different types of intervention treatment plans, and classify the test data corresponding to the same intervention treatment plan.

[0116] Calculate the mean change ratio of the feature matching parameters from the implementation of the intervention treatment plan to the reexamination in the test data as the expected feature matching parameter change ratio;

[0117] Calculate the mean change ratio of emotional gaze features from the implementation of the intervention treatment plan to the reexamination in the test data as the expected change ratio of emotional gaze features;

[0118] Calculate the mean change ratio of limb movement amplitude from the implementation of the intervention treatment plan to the reexamination in the test data as the expected change ratio of limb movement amplitude;

[0119] The expected feature matching parameter change ratio, the expected emotion gaze feature change ratio, and the expected limb movement amplitude change ratio are determined as the expected control data corresponding to the intervention treatment plan.

[0120] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An analysis method for auxiliary diagnosis and treatment of autistic children, characterized in that: include: Step S1, pre-detecting the target object and storing detection data for the target object, wherein the pre-detection includes setting an image acquisition device to detect the action characteristics of the target object when it performs a communication action, wherein the action characteristics include the amplitude of the limb movement and the reaction time; Step S2, analyzing the motion feature representation of the target object based on the motion feature to determine the motion feature feedback degree of the target object; Step S3, based on the feedback degree of the target object's motion characteristics, performing auxiliary diagnosis and treatment analysis, including: Adjusting the image acquisition accuracy of the image acquisition device for the target object, calling the video record in the target object's review case, identifying the emotional gaze features, the amplitude of body movements, and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standard is met; Or, generating a recommended number of review times based on the action feature representation; The feature matching parameter is determined by combining the ratio of the emotional gaze sub-features with the amplitude of the body movement, and the emotional gaze features include the gaze duration of the eye area and the gaze duration of the mouth area; Step S4, outputting the auxiliary diagnosis and treatment analysis result through the output end.

2. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: In step S2, the action feature representation of the target object is analyzed based on the action feature according to formula (1): include, In formula (1), S represents the action feature representation, F represents the limb movement amplitude, F0 represents the limb movement amplitude threshold, T represents the reaction time, T0 represents the reaction time threshold, α represents the limb movement amplitude weight coefficient, and β represents the reaction time weight coefficient.

3. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: In step S2, determining the action feature feedback degree of the target object includes: If the action feature representation amount is greater than or equal to the action feature representation amount threshold, the action feature of the target object is determined to be at a weak feedback level; If the motion feature representation amount is less than the motion feature representation amount threshold, the motion feature of the target object is determined to be at a strong feedback level.

4. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: In step S3, auxiliary diagnosis and treatment analysis is performed based on the feedback degree of the target object's motion characteristics. include, If the target object's motion feature is a weak feedback degree, the image acquisition accuracy of the image acquisition device for the target object is adjusted, the video record in the target object's review case is called, and the emotional gaze feature, the body movement amplitude, and the change ratio of the feature matching parameter relative to the pre-detection are identified based on the video record, and compared with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan to determine whether the feature matching standard is met; If the action feature of the target object is a strong feedback level, a recommended number of reviews is generated based on the action feature representation.

5. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: In step S3, the image acquisition accuracy of the image acquisition device for the target object is adjusted. include, Improve the resolution and maximum frame rate of the image acquisition device.

6. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: In step S3, the feature matching parameters are determined by calculating the ratio of the emotional gaze sub-features and combining them with the body movement amplitude. include, Obtaining video records captured by an image acquisition device to determine emotional gaze sub-features and body movement amplitudes; The ratio of the fixation time of the eye area to the fixation time of the mouth area is used as the ratio of the emotion fixation sub-feature and as the first feature; Calculate the ratio of the limb movement amplitude to the expected threshold of the limb movement amplitude as the second feature; The sum of the first feature and the second feature is determined as a feature matching parameter.

7. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 6, characterized in that: In step S3, the process of identifying the emotional gaze features, the body movement amplitude, and the change ratio of the feature matching parameters relative to the pre-detection based on the video record, and comparing them with the expected control data determined based on the pre-stored intervention diagnosis and treatment plan includes: Determine the change ratio of emotional gaze features, the change ratio of body movement amplitude, and the change ratio of feature matching parameters; Determine the expected emotional gaze feature change ratio, expected limb movement amplitude change ratio, and expected feature matching parameter change ratio corresponding to the intervention diagnosis and treatment plan; comparing the emotion gaze feature change ratio with an expected emotion gaze feature change ratio; comparing the limb movement amplitude change ratio with the expected limb movement amplitude change ratio; The feature matching parameter change ratio is compared with the expected feature matching parameter change ratio.

8. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 7, characterized in that: In step S3, the determination of whether the feature matching standard is met includes: If the expected matching conditions are met, it is determined that the feature matching standard is met; The expected matching conditions include that the feature matching parameter change ratio is less than the expected feature matching parameter change ratio, the emotional gaze feature change ratio is less than the expected emotional gaze feature change ratio, and the limb movement amplitude change ratio is less than the expected limb movement amplitude change ratio.

9. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: In the step S3, generating the recommended number of review times based on the action feature representation includes: The recommended review times are positively correlated with the action feature representation quantity.

10. The analysis method for auxiliary diagnosis and treatment of autistic children according to claim 1, characterized in that: The step S3 also includes pre-storing the expected control data of each intervention treatment plan, including: Expected feature matching parameter change ratio, expected emotional gaze feature change ratio, and expected limb movement amplitude change ratio.

Images