A joint disease diagnosis device based on gait characteristics

Abstract

The present invention provides a joint disease diagnosis device based on gait features, which belongs to the field of auxiliary diagnosis technology and specifically includes: an image recognition module, a feature screening module, a confidence assessment module, and a disease diagnosis module; the image recognition module is responsible for determining the probability of the user's different inferred disease types and suspected patients based on different gait images of the user; the feature screening module is responsible for extracting and screening gait features based on different gait images; the confidence assessment module is responsible for determining the comprehensive feature confidence of the suspected patient based on the confidence of different screened gait features and the comprehensive correlation coefficient; the disease diagnosis module is responsible for diagnosing the joint disease of the suspected patient based on the weight of the screened gait features and the screened gait features when the comprehensive feature confidence of the suspected patient meets the requirements, thereby further improving the efficiency and accuracy of disease diagnosis processing.

Description

Technical Field

[0001] The present invention belongs to the field of auxiliary diagnosis technology, and in particular relates to a joint disease diagnosis device based on gait characteristics. Background Art

[0002] Gait refers to the body's posture while walking. It is the overall external manifestation of the body's neuromuscular regulation system, limb function, and motor physiological and psychological indicators during walking. Normal gait exhibits numerous characteristics, including stability, periodicity, directionality, rhythmicity, coordination, and individual variability. However, gait characteristics change with aging joints or the presence of joint diseases. Therefore, identifying and diagnosing joint diseases based on gait characteristics has become a pressing technical challenge.

[0003] To address the above technical issues, the existing technical solution CN201811067995.8, "Knee Osteoarthritis Diagnostic System Based on Phase Space Reconstruction, Euclidean Distance, and Neural Network," utilizes the differences in gait system dynamics between healthy individuals and patients with knee osteoarthritis to perform classification, thereby assisting in the diagnosis and detection of knee osteoarthritis and achieving non-invasive diagnosis of knee osteoarthritis. However, this solution also presents the following technical issues:

[0004] The existing technical solutions ignore the similarity of gait features in different images. Since the gait features of people often have a certain periodicity and stability, the gait features of different time periods often have a certain similarity. If the similarity of gait features is not considered, it is impossible to accurately screen gait features with higher confidence.

[0005] In response to the above technical problems, the present invention provides a joint disease diagnosis device based on gait characteristics. Summary of the Invention

[0006] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0007] According to one aspect of the present invention, a method for dividing product modules based on a clustering algorithm is provided.

[0008] A device for diagnosing joint diseases based on gait characteristics, characterized by comprising:

[0009] Image recognition module, feature screening module, confidence assessment module, disease diagnosis module;

[0010] The image recognition module is responsible for determining the probability of the user suffering from different inferred disease types and the suspected patient according to different gait images of the user;

[0011] The feature screening module is responsible for determining the comprehensive correlation coefficients of different gait features and screening gait features based on the autocorrelation coefficients between different gait features and different presumed disease types and the probability of illness of different presumed disease types, and extracting and screening gait features based on different gait images;

[0012] The confidence assessment module is responsible for determining the confidence of different screened gait features based on the similarity of the screened gait features of different gait images and the number of gait images containing the screened gait features, and determining the confidence of the comprehensive features of the suspected patient based on the confidence of the different screened gait features and the comprehensive correlation coefficient;

[0013] The disease diagnosis module is responsible for determining the weights of different screening gait features according to the confidence and comprehensive correlation coefficient when the comprehensive feature confidence of the suspected patient meets the requirements, and diagnosing the joint disease of the suspected patient based on the weights of the screening gait features and the screening gait features.

[0014] The beneficial effects of the present invention are:

[0015] 1. By estimating the probability of a user's disease type and identifying suspected patients based on their different gait images, the system first assesses the probability of the disease type from the perspective of gait images, thereby screening suspected patients and improving the efficiency of disease diagnosis and treatment.

[0016] 2. By determining the confidence level of the comprehensive characteristics of suspected patients based on the confidence levels of different screening gait characteristics and the comprehensive correlation coefficient, the credibility of the screening gait characteristics and their correlation with the disease diagnosis results are fully considered, thus avoiding the problem of inaccurate disease diagnosis results due to low credibility of the screening gait characteristics, and further ensuring the accuracy of the disease diagnosis results.

[0017] 3. By diagnosing joint diseases in suspected patients based on the weights and screening gait features, not only are the differences in the impact of different screening gait features on the accuracy of disease diagnosis results due to differences in confidence and correlation are fully considered, but also by mining screening gait features with higher correlation, the accuracy of disease diagnosis results is ensured on the basis of reducing the input feature dimension.

[0018] A further technical solution is that the gait image is extracted based on a walking video of the user within a specified time period.

[0019] A further technical solution is that the method for determining the confidence level of the comprehensive features of the suspected patient is:

[0020] Determining weight values of confidence levels of different screened gait features according to comprehensive correlation coefficients of different screened gait features, and determining modified confidence levels of different screened gait features in combination with the confidence levels of different screened gait features;

[0021] The number of screening gait features whose corrected confidence is greater than the preset confidence is determined by using different corrected confidences of the screening gait features, and the comprehensive feature confidence of the suspected patient is determined by combining the different corrected confidences of the screening gait features and the number of the screening gait features.

[0022] A further technical solution is that when the comprehensive feature confidence of the suspected patient is greater than the confidence threshold, it is determined that the comprehensive feature confidence of the suspected patient meets the requirements, and when the comprehensive feature confidence does not meet the requirements, the acquisition of the gait image of the suspected patient continues until the comprehensive feature confidence of the suspected patient meets the requirements.

[0023] Other features and advantages will be described in the following description, and in part will become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the accompanying drawings.

[0026] Figure 1 This is a framework diagram of a joint disease diagnosis device based on gait characteristics;

[0027] Figure 2 This is a flowchart of specific steps for determining the estimated probability of disease type. DETAILED DESCRIPTION

[0028] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. Like reference numerals in the figures represent like or similar structures, and thus their detailed description will be omitted.

[0029] The terms "a", "an", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "including" and "having" are used to express an open-ended inclusive meaning and mean that additional elements / components / etc. may be present in addition to the listed elements / components / etc.

[0030] Example 1

[0031] To solve the above problems, according to one aspect of the present invention, Figure 1 According to one aspect of the present invention, a device for diagnosing joint diseases based on gait characteristics is provided, which is characterized by specifically comprising:

[0032] Image recognition module, feature screening module, confidence assessment module, disease diagnosis module;

[0033] The image recognition module is responsible for determining the probability of the user suffering from different inferred disease types and the suspected patient according to different gait images of the user;

[0034] Specifically, the gait image is extracted based on a walking video of the user within a specified time period.

[0035] In one possible embodiment, Figure 2 As shown, the specific steps for determining the probability of the disease type are as follows:

[0036] Determining predicted disease probabilities of the inferred disease types for the different gait images based on recognition results of the different gait images of the user, and determining associated gait images of the inferred disease types based on the predicted disease probabilities of the inferred disease types for the different gait images;

[0037] Determining whether there is a strongly correlated image in the associated gait images based on the predicted disease probabilities of the associated gait images of the inferred disease type, and if so, proceeding to the next step; if not, determining the disease probability of the inferred disease type based on the average value of the predicted disease probabilities of the associated gait images;

[0038] Obtaining the number of strongly correlated images of the inferred disease type, and determining the comprehensive disease probability of the strongly correlated images of the inferred disease type in combination with the predicted disease probability of the strongly correlated images and the proportion in the gait image, and judging whether the comprehensive disease probability of the strongly correlated images of the inferred disease type meets the requirements; if so, proceeding to the next step; if not, using the comprehensive disease probability of the strongly correlated images as the disease probability of the inferred disease type;

[0039] Based on the number of associated gait images of the inferred disease type and the proportion of the gait images in the gait images, and the average value of the predicted disease probabilities of the associated gait images, the comprehensive disease probability of the strongly associated images is corrected to obtain the disease probability of the inferred disease type.

[0040] To give a specific example, the specific steps for determining the suspected patient are:

[0041] S11 determines whether there is a presumed disease type with a probability greater than a preset probability based on the presumed disease types of the user. If so, the user is considered a suspected patient. If not, proceed to the next step.

[0042] S12 uses the inferred disease type with the disease probability within the set probability interval as the screening disease type, and determines whether the number of the user's screening disease types is less than the preset number of disease types. If so, proceed to the next step; if not, proceed to step S14;

[0043] S13: determining an average value of the probability of the user's screening disease type suffering from the screening disease type and a maximum value of the probability of the user's screening disease type suffering from the screening disease type based on the user's screening disease type, and determining the disease credibility of the user's screening disease type in combination with the number of the user's screening disease types, and determining whether the user is not a suspected patient based on the disease credibility; if so, determining that the user is not a suspected patient; if not, proceeding to the next step;

[0044] S14 obtains the number of the user's presumed disease types and the probability of illness of different presumed disease types, and determines the user's comprehensive illness credibility based on the illness credibility of the user's screened disease types, and determines whether the user is a suspected patient based on the comprehensive credibility.

[0045] It should be noted that the preset probability is determined according to the number of the gait images and the recognition accuracy of the gait images. The more gait images of the user and the higher the recognition accuracy of the gait images, the greater the preset probability.

[0046] In another possible embodiment, the specific steps of determining the suspected patient are:

[0047] The inferred disease type with a probability of illness within a set probability interval is used as a screening disease type, and if the user does not have the screening disease type, it is determined that the user is not a suspected patient;

[0048] When the user has a screening disease type, determine whether the number of presumed disease types whose probability of the user suffering from the disease is greater than a preset probability meets the requirement. If so, determine that the user is a suspected patient. If not, proceed to the next step.

[0049] Determining the sum of the number of disease probabilities of the user's screened disease types based on the number of the user's screened disease types and the disease probabilities of different screened disease types, and judging whether the sum of the number of disease probabilities of the user's screened disease types meets the requirements; if so, proceeding to the next step; if not, determining that the patient is not a suspected patient;

[0050] Determining an average value of the disease probabilities of the user's screened disease types and a maximum value of the disease probabilities of the user's screened disease types based on the user's screened disease types, and determining the disease credibility of the user's screened disease types in combination with the number of the user's screened disease types, and determining whether the user is not a suspected patient based on the disease credibility; if so, determining that the user is not a suspected patient; if not, proceeding to the next step;

[0051] The number of the user's presumed disease types and the probability of illness of different presumed disease types are obtained, and the comprehensive credibility of the user's illness is determined in combination with the credibility of the user's screened disease types, and whether the user is a suspected patient is determined based on the comprehensive credibility.

[0052] The feature screening module is responsible for determining the comprehensive correlation coefficients of different gait features and screening gait features based on the autocorrelation coefficients between different gait features and different presumed disease types and the probability of illness of different presumed disease types, and extracting and screening gait features based on different gait images;

[0053] In a possible embodiment, the method for determining the screening gait characteristics is:

[0054] S21: determining the autocorrelation coefficients between the gait characteristics and different presumed disease types by principal component analysis, and determining the corrected correlation coefficients between the gait characteristics and different presumed disease types in combination with the morbidity probabilities of the different presumed disease types, and determining whether the corrected correlation coefficients between the gait characteristics and different presumed disease types are all less than a minimum correlation coefficient limit; if so, determining that the gait characteristics do not belong to the screening gait characteristics; if not, proceeding to the next step;

[0055] S22: Determine whether the gait feature has a presumed disease type whose modified correlation coefficient is greater than a preset correlation threshold. If so, proceed to step S24; if not, proceed to the next step.

[0056] S23: determining a sum of the corrected correlation coefficients of the gait feature based on the corrected correlation coefficients of the gait feature and different presumed disease types, and judging whether the sum of the corrected correlation coefficients of the gait feature meets the requirement; if so, proceeding to the next step; if not, determining that the gait feature does not belong to the screening gait feature;

[0057] S24: the inferred disease type with the corrected correlation coefficient greater than the preset correlation threshold is regarded as a strongly correlated type, and the characteristic correlation coefficient of the strongly correlated type of the gait feature is determined according to the number of the strongly correlated types of the gait feature and the corrected correlation coefficients of different strongly correlated types, and whether the characteristic correlation coefficient of the strongly correlated type of the gait feature meets the requirements is judged; if so, the process proceeds to the next step; if not, the process determines that the gait feature does not belong to the screening gait feature;

[0058] S25 takes the inferred disease type whose corrected correlation coefficient is not less than the minimum correlation coefficient limit as the associated disease type, and determines the comprehensive correlation coefficient of the gait feature based on the number of associated disease types of the gait feature, the average value of the corrected correlation coefficients of different associated disease types, and the characteristic correlation coefficient of the strongly correlated type of the gait feature, and determines the screened gait feature based on the comprehensive correlation coefficient.

[0059] Specifically, determining the screening gait features based on the comprehensive correlation coefficient includes:

[0060] When the comprehensive correlation coefficient of the gait feature is greater than a preset correlation coefficient threshold, the gait feature is determined to be a screening gait feature.

[0061] In another possible embodiment, the method for determining the screening gait characteristics is:

[0062] Determining the autocorrelation coefficients between the gait characteristics and different presumed disease types by principal component analysis, and determining the corrected correlation coefficients between the gait characteristics and different presumed disease types in combination with the probabilities of the different presumed disease types;

[0063] When the corrected correlation coefficients between the gait feature and different presumed disease types are all greater than a preset correlation threshold, determining the gait feature as a screening gait feature, and determining a comprehensive correlation coefficient of the gait feature using the preset correlation coefficient;

[0064] When the corrected correlation coefficient between the gait feature and different presumed disease types is not greater than a preset correlation threshold for the presumed disease type:

[0065] When the corrected correlation coefficients between the gait feature and different presumed disease types are all less than a minimum correlation coefficient limit, determining that the gait feature does not belong to the screening gait feature;

[0066] When the modified correlation coefficients between the gait characteristics and different presumed disease types are all less than the minimum correlation coefficient limit:

[0067] The inferred disease type whose corrected correlation coefficient is greater than a preset correlation threshold is regarded as a strongly correlated type, and the characteristic correlation coefficient of the strongly correlated type of the gait feature is determined according to the number of the strongly correlated types of the gait feature and the corrected correlation coefficients of different strongly correlated types, and whether the characteristic correlation coefficient of the strongly correlated type of the gait feature meets the requirements is judged, if so, proceeding to the next step, if not, determining that the gait feature does not belong to the screening gait feature;

[0068] The inferred disease type whose corrected correlation coefficient is not less than the minimum correlation coefficient limit is taken as the associated disease type, and the comprehensive correlation coefficient of the gait feature is determined based on the number of associated disease types of the gait feature, the average value of the corrected correlation coefficients of different associated disease types, and the characteristic correlation coefficient of the strongly correlated type of the gait feature, and the screening gait feature is determined based on the comprehensive correlation coefficient.

[0069] The confidence assessment module is responsible for determining the confidence of different screened gait features based on the similarity of the screened gait features of different gait images and the number of gait images containing the screened gait features, and determining the confidence of the comprehensive features of the suspected patient based on the confidence of the different screened gait features and the comprehensive correlation coefficient;

[0070] Specifically, the method for determining the confidence level of the gait feature screening is as follows:

[0071] Extracting the screened gait features based on different gait images to obtain screened gait features of different gait images, determining similarities of the screened gait features of different gait images based on the screened gait features of different gait images, determining valid gait features among the screened gait features of different gait images based on the similarities, and using the gait image corresponding to the valid gait feature as the valid gait image;

[0072] Determining whether the number of the valid gait images is less than a preset number of images; if so, determining the confidence of the screened gait features based on the number of the valid gait images; if not, proceeding to the next step;

[0073] Based on the effective gait features of different effective gait images, the similarities between different effective gait features are determined, and based on the similarities between different effective gait features, the average value of the similarities between different effective gait features and the number of effective gait features with similarities greater than a preset similarity are determined, and the confidence of the screened gait features is determined in combination with the number of the effective gait images and the proportion of the number of the gait images.

[0074] It is understandable that the method for determining the confidence level of the comprehensive features of the suspected patient is:

[0075] Determining weight values of confidence levels of different screened gait features according to comprehensive correlation coefficients of different screened gait features, and determining modified confidence levels of different screened gait features in combination with the confidence levels of different screened gait features;

[0076] The number of screening gait features whose corrected confidence is greater than the preset confidence is determined by using different corrected confidences of the screening gait features, and the comprehensive feature confidence of the suspected patient is determined by combining the different corrected confidences of the screening gait features and the number of the screening gait features.

[0077] It should be noted that when the comprehensive feature confidence of the suspected patient is greater than the confidence threshold, it is determined that the comprehensive feature confidence of the suspected patient meets the requirements, and when the comprehensive feature confidence does not meet the requirements, the acquisition of the gait image of the suspected patient continues until the comprehensive feature confidence of the suspected patient meets the requirements.

[0078] The disease diagnosis module is responsible for determining the weights of different screening gait features according to the confidence and comprehensive correlation coefficient when the comprehensive feature confidence of the suspected patient meets the requirements, and diagnosing the joint disease of the suspected patient based on the weights of the screening gait features and the screening gait features.

[0079] Through the above embodiments, the present application achieves the following technical effects:

[0080] 1. By estimating the probability of a user's disease type and identifying suspected patients based on their different gait images, the system first assesses the probability of the disease type from the perspective of gait images, thereby screening suspected patients and improving the efficiency of disease diagnosis and treatment.

[0081] 2. By determining the confidence level of the comprehensive characteristics of suspected patients based on the confidence levels of different screening gait characteristics and the comprehensive correlation coefficient, the credibility of the screening gait characteristics and their correlation with the disease diagnosis results are fully considered, thus avoiding the problem of inaccurate disease diagnosis results due to low credibility of the screening gait characteristics, and further ensuring the accuracy of the disease diagnosis results.

[0082] 3. By diagnosing joint diseases in suspected patients based on the weights and screening gait features, not only are the differences in the impact of different screening gait features on the accuracy of disease diagnosis results due to differences in confidence and correlation are fully considered, but also by mining screening gait features with higher correlation, the accuracy of disease diagnosis results is ensured on the basis of reducing the input feature dimension.

[0083] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, apparatus, and non-volatile computer storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant details, refer to the descriptions of the method embodiments.

[0084] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The foregoing description is merely one or more embodiments of this specification and is not intended to limit this specification. It will be apparent to those skilled in the art that various modifications and variations may be made to one or more embodiments of this specification. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of one or more embodiments of this specification are intended to be within the scope of the claims of this specification.

Claims

1. A device for diagnosing joint diseases based on gait characteristics, characterized in that: Specifically include: Image recognition module, feature screening module, confidence assessment module, disease diagnosis module; The image recognition module is responsible for determining the probability of the user suffering from different inferred disease types and the suspected patient according to different gait images of the user; The feature screening module is responsible for determining the comprehensive correlation coefficients of different gait features and screening gait features based on the correlation coefficients between different gait features and different presumed disease types and the probability of morbidity of different presumed disease types, and extracting and screening gait features based on different gait images; The confidence assessment module is responsible for determining the confidence of different screened gait features based on the similarity of the screened gait features of different gait images and the number of gait images containing the screened gait features, and determining the confidence of the comprehensive features of the suspected patient based on the confidence of the different screened gait features and the comprehensive correlation coefficient; The disease diagnosis module is responsible for determining the weights of different screening gait features according to the confidence and comprehensive correlation coefficient when the comprehensive feature confidence of the suspected patient meets the requirements, and diagnosing the joint disease of the suspected patient based on the weights of the screening gait features and the screening gait features.

2. The joint disease diagnosis device based on gait characteristics according to claim 1, characterized in that: The gait image is extracted based on a walking video of the user within a specified time period.

3. The joint disease diagnosis device based on gait characteristics according to claim 1, characterized in that: The specific steps of determining the probability of the disease type are as follows: Determining predicted disease probabilities of the inferred disease types for the different gait images based on recognition results of the different gait images of the user, and determining associated gait images of the inferred disease types based on the predicted disease probabilities of the inferred disease types for the different gait images; Determining whether there is a strongly correlated image in the associated gait images based on the predicted disease probabilities of the associated gait images of the inferred disease type, and if so, proceeding to the next step; if not, determining the disease probability of the inferred disease type based on the average value of the predicted disease probabilities of the associated gait images; Obtaining the number of strongly correlated images of the inferred disease type, and determining the comprehensive disease probability of the strongly correlated images of the inferred disease type in combination with the predicted disease probability of the strongly correlated images and the proportion in the gait image, and judging whether the comprehensive disease probability of the strongly correlated images of the inferred disease type meets the requirements; if so, proceeding to the next step; if not, using the comprehensive disease probability of the strongly correlated images as the disease probability of the inferred disease type; Based on the number of associated gait images of the inferred disease type and the proportion of the gait images in the gait images, and the average value of the predicted disease probabilities of the associated gait images, the comprehensive disease probability of the strongly associated images is corrected to obtain the disease probability of the inferred disease type.

4. The joint disease diagnosis device based on gait characteristics according to claim 1, characterized in that: The specific steps for determining the suspected patient are: S11 determines whether there is a presumed disease type with a probability greater than a preset probability based on the presumed disease types of the user. If so, the user is considered a suspected patient. If not, proceed to the next step. S12 uses the inferred disease type with the disease probability within the set probability interval as the screening disease type, and determines whether the number of the user's screening disease types is less than the preset number of disease types. If so, proceed to the next step; if not, proceed to step S14; S13: determining an average value of the probability of the user's screening disease type suffering from the screening disease type and a maximum value of the probability of the user's screening disease type suffering from the screening disease type based on the user's screening disease type, and determining the disease credibility of the user's screening disease type in combination with the number of the user's screening disease types, and determining whether the user is not a suspected patient based on the disease credibility; if so, determining that the user is not a suspected patient; if not, proceeding to the next step; S14 obtains the number of the user's presumed disease types and the probability of illness of different presumed disease types, and determines the user's comprehensive illness credibility based on the illness credibility of the user's screened disease types, and determines whether the user is a suspected patient based on the comprehensive credibility.

5. The joint disease diagnosis device based on gait characteristics according to claim 4, characterized in that: The preset probability is determined according to the number of the gait images and the recognition accuracy of the gait images. The more gait images of the user, the higher the recognition accuracy of the gait images, and the greater the preset probability.

6. The joint disease diagnosis device based on gait characteristics according to claim 1, characterized in that: The method for determining the screening gait characteristics is: Determining the correlation coefficients between the gait characteristics and different presumed disease types by principal component analysis, and determining the corrected correlation coefficients between the gait characteristics and different presumed disease types in combination with the probabilities of the different presumed disease types; When the corrected correlation coefficients between the gait feature and different presumed disease types are all greater than a preset correlation threshold, determining the gait feature as a screening gait feature, and determining a comprehensive correlation coefficient of the gait feature using the preset correlation coefficient; When the corrected correlation coefficient between the gait feature and different presumed disease types is not greater than a preset correlation threshold for the presumed disease type: When the corrected correlation coefficients between the gait feature and different presumed disease types are all less than a minimum correlation coefficient limit, determining that the gait feature does not belong to the screening gait feature; When the modified correlation coefficients between the gait characteristics and different presumed disease types are all less than the minimum correlation coefficient limit: The inferred disease type whose corrected correlation coefficient is greater than a preset correlation threshold is regarded as a strongly correlated type, and the characteristic correlation coefficient of the strongly correlated type of the gait feature is determined according to the number of the strongly correlated types of the gait feature and the corrected correlation coefficients of different strongly correlated types, and whether the characteristic correlation coefficient of the strongly correlated type of the gait feature meets the requirements is judged, if so, proceeding to the next step, if not, determining that the gait feature does not belong to the screening gait feature; The inferred disease type whose corrected correlation coefficient is not less than the minimum correlation coefficient limit is taken as the associated disease type, and the comprehensive correlation coefficient of the gait feature is determined based on the number of associated disease types of the gait feature, the average value of the corrected correlation coefficients of different associated disease types, and the characteristic correlation coefficient of the strongly correlated type of the gait feature, and the screening gait feature is determined based on the comprehensive correlation coefficient.

7. The joint disease diagnosis device based on gait characteristics according to claim 6, characterized in that: Determining the screening gait features based on the comprehensive correlation coefficient specifically includes: When the comprehensive correlation coefficient of the gait feature is greater than a preset correlation coefficient threshold, the gait feature is determined to be a screening gait feature.

8. The joint disease diagnosis device based on gait characteristics according to claim 1, characterized in that: The method for determining the confidence level of the comprehensive features of the suspected patient is: Determining weight values of confidence levels of different screened gait features according to comprehensive correlation coefficients of different screened gait features, and determining modified confidence levels of different screened gait features in combination with the confidence levels of different screened gait features; The number of screening gait features whose corrected confidence is greater than the preset confidence is determined by using different corrected confidences of the screening gait features, and the comprehensive feature confidence of the suspected patient is determined by combining the different corrected confidences of the screening gait features and the number of the screening gait features.

9. The joint disease diagnosis device based on gait characteristics according to claim 1, characterized in that: When the comprehensive feature confidence of the suspected patient is greater than the confidence threshold, it is determined that the comprehensive feature confidence of the suspected patient meets the requirements, and when the comprehensive feature confidence does not meet the requirements, the acquisition of the suspected patient's gait image continues until the comprehensive feature confidence of the suspected patient meets the requirements.

Images