Auxiliary decision-making method and device for ankle-foot orthosis adaptation

Abstract

The invention relates to an auxiliary decision-making method and device for adaptation of an ankle-foot orthosis. The method comprises the following steps: determining a first candidate orthosis matched with a first ankle-foot symptom set and a second ankle-foot symptom set associated with the first candidate orthosis from an ankle-foot case knowledge graph; determining the matching degree between the first ankle-foot symptom set and each second ankle-foot symptom set to obtain a first matching result; selecting an effective path and a first condition node set corresponding to the effective path from the expert rule knowledge graph; obtaining standard rule condition nodes of a second candidate orthosis from the expert rule knowledge graph to obtain a second condition node set; determining the matching degree of the first condition node set and the second condition node set to obtain a second matching result; and generating an auxiliary decision for carrying out ankle-foot orthosis adaptation on the object based on the first matching result and the second matching result. According to the scheme, the interpretability of the ankle foot orthosis in adapting to the intelligent decision is improved.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent rehabilitation equipment technology, and in particular to an auxiliary decision-making method and device for ankle-foot orthosis fitting. Background Technology

[0002] In related technologies, rehabilitation assistive devices are products used to improve, compensate for, and replace human functions, implement assistive treatments, and prevent disabilities. They are one of the important means of rehabilitation, education, employment, and participation in social life. Ankle-foot orthoses (AFOs) are core rehabilitation devices for correcting motor dysfunctions such as foot drop and ankle instability. Their fitting effect directly affects the reconstruction of patients' walking function and the improvement of their quality of life.

[0003] However, traditional ankle-foot orthosis fitting methods mainly rely on machine learning algorithms (such as random forests and deep learning) for assistive device recommendations. Although such methods have certain advantages in evaluation efficiency, their "black box" nature leads to a lack of interpretability in the decision-making process and makes it impossible to trace the generation logic of the recommendation results, which restricts the credibility and adoption rate of ankle-foot orthosis fitting decisions. Summary of the Invention

[0004] To overcome the problems existing in the related technologies, this disclosure provides an auxiliary decision-making method and device for ankle-foot orthosis fitting.

[0005] According to a first aspect of the present disclosure, an auxiliary decision-making method for ankle-foot orthosis fitting is provided, comprising: The first set of ankle and foot symptoms, the ankle and foot case knowledge graph, and the expert rule knowledge graph associated with the ankle and foot orthosis are obtained respectively; the ankle and foot case knowledge graph is generated based on historical ankle and foot orthosis fitting cases; the expert rule knowledge graph is generated based on expert knowledge. Identify at least one first candidate orthosis that matches the first ankle and foot symptom set from the ankle and foot case knowledge graph, and a second ankle and foot symptom set associated with each first candidate orthosis; The degree of matching between the first ankle and foot symptom set and each second ankle and foot symptom set is determined to obtain the first matching result; Select at least one valid path from the expert rule knowledge graph that matches the first ankle and foot symptom set, and a first condition node set corresponding to the at least one valid path; the first condition node set corresponds to at least one second candidate orthosis; The standard rule condition nodes for each second candidate orthotics are obtained from the expert rule knowledge graph to obtain the second condition node set; For each second candidate orthodontic device, the degree of matching between the first condition node set and the second condition node set of the second candidate orthodontic device is determined to obtain the second matching result; If there is no conflict between the first matching result and the second matching result, an auxiliary decision for ankle-foot orthosis fitting of the object is generated based on the first matching result and the second matching result.

[0006] According to a second aspect of the present disclosure, an assistive decision-making device for ankle-foot orthosis fitting is provided, comprising: The first acquisition unit is used to acquire the first ankle and foot symptom set, the ankle and foot case knowledge graph, and the expert rule knowledge graph associated with the ankle and foot orthosis, respectively; the ankle and foot case knowledge graph is generated based on historical ankle and foot orthosis fitting cases; the expert rule knowledge graph is generated based on expert knowledge. The determining unit is configured to determine, from the ankle and foot case knowledge graph, at least one first candidate orthosis that matches the first ankle and foot symptom set, and a second ankle and foot symptom set associated with each first candidate orthosis; The first matching unit is used to determine the degree of matching between the first ankle and foot symptom set and each second ankle and foot symptom set, and to obtain the first matching result; The selection unit is used to select at least one valid path from the expert rule knowledge graph that matches the first ankle and foot symptom set, and a first condition node set corresponding to the at least one valid path; the first condition node set corresponds to at least one second candidate orthosis. The second acquisition unit is used to acquire the standard rule condition node of each second candidate orthotics from the expert rule knowledge graph, and obtain the second condition node set; The second matching unit is used to determine the degree of matching between the first set of condition nodes of the second candidate orthodont and the second set of condition nodes for each second candidate orthodont, and to obtain the second matching result; The decision unit is configured to generate an auxiliary decision for fitting the object with an ankle-foot orthosis based on the first matching result and the second matching result, provided that there is no conflict between the first matching result and the second matching result.

[0007] According to a third aspect of the present disclosure, an electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in any one of the first aspects.

[0008] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of the first aspects.

[0009] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method as described in any one of the first aspects.

[0010] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: determining at least one first candidate orthosis matching a first ankle and foot symptom set from an ankle and foot case knowledge graph, and a second ankle and foot symptom set associated with each first candidate orthosis; determining the degree of matching between the first ankle and foot symptom set and each second ankle and foot symptom set respectively to obtain a first matching result; selecting at least one valid path matching the first ankle and foot symptom set from an expert rule knowledge graph, and a first condition node set corresponding to at least one valid path; obtaining standard rule condition nodes for each second candidate orthosis from the expert rule knowledge graph to obtain a second condition node set; determining the degree of matching between the first condition node set and the second condition node set of the second candidate orthosis to obtain a second matching result; and generating an auxiliary decision for ankle and foot orthosis fitting of an object based on the first matching result and the second matching result, provided that there is no conflict between the first matching result and the second matching result. By employing a dual-reasoning collaborative approach based on ankle and foot case knowledge graphs and expert rule knowledge graphs, the advantages of data-driven and knowledge-driven approaches are complemented. Furthermore, the dual-reasoning collaborative approach enables auxiliary decision-making results to have dual interpretability, thereby allowing users to clearly understand the decision-making logic of ankle and foot orthosis fitting and improving the credibility of ankle and foot orthosis fitting decisions.

[0011] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0012] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0013] Figure 1 This is a flowchart illustrating an auxiliary decision-making method for ankle-foot orthosis fitting according to an exemplary embodiment.

[0014] Figure 2 This is a flowchart of the ankle-foot orthosis matching process proposed in the embodiments of this disclosure.

[0015] Figure 3This is a rule-based ankle-foot orthosis decision flowchart proposed in the embodiments of this disclosure.

[0016] Figure 4 This is a flowchart of the ankle-foot orthosis rule matching process proposed in the embodiments of this disclosure.

[0017] Figure 5 This is a schematic diagram of the system hierarchy of the auxiliary decision-making method for ankle-foot orthosis fitting proposed in the embodiments of this disclosure.

[0018] Figure 6 This is a flowchart illustrating the reasoning process for ankle-foot orthosis fitting assistance decision-making as proposed in this embodiment.

[0019] Figure 7 This is a flowchart of the ankle-foot orthosis fitting auxiliary decision-making process proposed in the embodiments of this disclosure.

[0020] Figure 8 This is a block diagram illustrating an auxiliary decision-making device for ankle-foot orthosis fitting according to an exemplary embodiment.

[0021] Figure 9 This is a block diagram illustrating an apparatus for an auxiliary decision-making method for ankle-foot orthosis fitting, according to an exemplary embodiment. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0023] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. The singular forms “a” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0024] It should be understood that although the terms first, second, third, etc., may be used to describe various information in embodiments of this disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of embodiments of this disclosure, and similarly, second information may also be referred to as first information. Depending on the context, the words “if” and “suppose” as used herein may be interpreted as “when”, “when”, or “in response to a determination”.

[0025] Furthermore, various forms of processes shown in the embodiments of this disclosure can be used to reorder, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0026] It should be noted that the collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0027] Figure 1 This is a flowchart illustrating an auxiliary decision-making method for ankle-foot orthosis fitting according to an exemplary embodiment, such as... Figure 1 As shown, it should be noted that the auxiliary decision-making method for ankle-foot orthosis fitting in this embodiment of the present disclosure is applied to an auxiliary decision-making device for ankle-foot orthosis fitting. For example... Figure 1 As shown, the method may include the following steps: Step 101: Obtain the first ankle and foot symptom set, ankle and foot case knowledge graph, and expert rule knowledge graph associated with the ankle and foot orthosis for the object.

[0028] Among them, the ankle-foot case knowledge graph is generated based on historical ankle-foot orthotic fitting cases; the expert rule knowledge graph is generated based on expert knowledge.

[0029] In one embodiment, the object may be a patient.

[0030] In some embodiments, patient case data input by the user can be obtained through a terminal device. The patient case data is then standardized and hierarchically mapped according to the same rules used when constructing the ankle-foot case knowledge graph and the expert rule knowledge graph to ensure consistent format. Then, information such as the patient's disease and symptoms is extracted from the patient case data to form a first ankle-foot symptom set S1, for example, S1={'cerebral palsy', 'foot drop', 'ankle inversion', 'muscle strength grade 2'}.

[0031] In one embodiment, an ankle-foot orthosis fitting assessment form and an ankle-foot orthosis configuration form can be determined by combining existing lower limb orthosis configuration assessment forms. Based on this, ankle-foot orthosis configuration cases are collected to obtain semi-structured case data.

[0032] The development of the ankle-foot orthosis fitting assessment and prescription forms can be achieved by collecting and organizing existing assessment forms and prescriptions from multiple institutions, as well as current standards, and combining this with the data required for ankle-foot orthosis assessment, using the Delphi method. Ankle-foot orthosis fitting cases are obtained from hospitals and are completed using the developed assessment and prescription forms to ensure the standardization and consistency of the case data. For semi-structured case data, data anonymization and cleaning are performed to remove patient privacy information and null values. Knowledge extraction is then performed based on rules to construct an ankle-foot case knowledge graph, which is stored in the Neo4j graph database.

[0033] In some embodiments of this disclosure, the ankle-foot case knowledge graph is generated using the following steps: Historical ankle-foot orthosis fitting cases are obtained, and knowledge extraction is performed on these cases to obtain target entities and the relationships between them. Target entities include symptoms and orthosis types. An ankle and foot case knowledge graph is generated based on the target entities and the relationships between them.

[0034] In some embodiments, patient privacy information, such as patient name, ID number, and home address, can be anonymized first to ensure data privacy. Then, the data is cleaned to remove null values. The characteristics of semi-structured case data are analyzed, and a top-down approach is used to construct a knowledge graph. First, entities and relationships are defined. Referring to the Chinese medical text annotation system and UMLS semantic types, and with guidance from hospital experts, a schema layer is determined, containing seven types of entities: patient, disease, symptoms or signs, examination, orthopedic effect, orthopedic device configuration, and body part, as well as 25 types of relationships. A rule-based knowledge extraction method is used. Based on the schema layer, knowledge extraction rules are defined to extract entities and relationships and represent them as triples. The nodes and edges of the ankle and foot case knowledge graph are created using the py2neo toolkit in Python, and then the data is stored in the Neo4j graph database for easy searching and querying. Finally, the knowledge graph is checked for entity and relationship simplicity and relationship consistency, and duplicate nodes and erroneous relationships are removed.

[0035] Step 102: Identify at least one first candidate orthosis that matches the first ankle-foot symptom set from the ankle-foot case knowledge graph, and a second ankle-foot symptom set associated with each first candidate orthosis.

[0036] In one embodiment, the same entity extraction rules as those used in the ankle and foot case knowledge graph can be employed to extract entities from the data, forming a first ankle and foot symptom set S1. The one-hop relation nodes (i.e., directly associated node entities) of each symptom node in S1 are queried from the ankle and foot case knowledge graph, and all orthotic entities (i.e., first candidate orthotics) are selected from the queried node set, forming a set of ankle and foot orthotic configurations to be matched, O1. For each orthotic in O1, all related one-hop relation nodes in the case knowledge graph are queried using Cypher language, and all symptom and disease nodes are selected to form a second ankle and foot symptom set S2 corresponding to the orthotic.

[0037] Step 103: Determine the degree of matching between the first ankle-foot symptom set and each second ankle-foot symptom set to obtain the first matching result.

[0038] It is understandable that the symptoms in the first ankle-foot symptom set are the actual symptoms of the subject, while the second ankle-foot symptom set is the symptoms corresponding to a certain type of ankle-foot orthosis in historical ankle-foot fitting cases. The higher the degree of matching, the higher the fit between the ankle-foot orthosis corresponding to the second ankle-foot symptom set and the patient.

[0039] In some embodiments of this disclosure, step 103 may specifically include the following steps: For each second ankle-foot symptom set, select the target symptom that exists in the second ankle-foot symptom set from all the symptoms in the first ankle-foot symptom set; Obtain the first preset weight value for each target symptom; The first preset weight values ​​corresponding to all target symptoms in the second ankle and foot symptom set are added together to obtain the first matching score of the second ankle and foot symptom set. All first candidate orthotics are sorted according to the first matching degree score to obtain the first matching result.

[0040] In one embodiment, the patient's first ankle and foot symptom set S1 can be traversed. For each symptom in S1, it is checked whether it also appears in the second ankle and foot symptom set S2. If it does, the TF-IDF weight value (i.e., the first preset weight value) of this symptom is found. The first preset weight values ​​corresponding to all target symptoms in the second ankle and foot symptom set are added together to obtain the first matching score of the second ankle and foot symptom set.

[0041] In some embodiments of this disclosure, the first preset weight value is calculated using the following steps: For the first candidate orthosis corresponding to the first preset weight value, determine the first frequency of the target symptom corresponding to the first preset weight value in the cases associated with the first candidate orthosis; Identify the second most frequent occurrence of the target symptom across all orthodontic-associated cases; Calculate the product of the first frequency and the second frequency to obtain the first preset weight value of the target symptom.

[0042] In some embodiments, all cases in the ankle-foot case knowledge graph can be grouped according to the type of ankle-foot orthosis used. For example, all cases using a "Dynamic Ankle-Foot Orthosis (DAFO)" could be grouped together, and all cases using a "Hinged Ankle-Foot Orthosis" could be grouped together.

[0043] In one embodiment, term frequency (TF) represents the frequency of a symptom occurring in cases of a particular orthosis. For example, in the "DAFO" group, the symptom "foot drop" occurs very frequently, so the TF value for "foot drop" in "DAFO" is high. This indicates that the symptom is a typical indicator of this orthosis. Inverse document frequency (DF) calculates the frequency of a symptom occurring in cases across all orthosis categories. If a symptom (such as "lower limb weakness") occurs in almost all orthosis types, its IDF value is low because it has poor discriminative power and cannot help distinguish one orthosis from another. TF-IDF = TF × IDF. The final first preset weight value combines the above two points. A high TF-IDF weight for a symptom with respect to a certain orthosis means that it is not only a typical symptom of this orthosis but also effectively distinguishes this orthosis from other orthoses.

[0044] In some embodiments of this disclosure, a reasoning model can also be constructed based on an ankle-foot case knowledge graph using traditional machine learning methods, and the model performance can be evaluated through cross-validation. The performance of the model can then be compared with that of the algorithm in step 3.

[0045] The process begins by extracting disease and symptom entities from all cases. Then, Cypher is used to query all disease, symptom, and therapeutic operation nodes to form a feature set. Based on this feature set, a two-dimensional data table suitable for machine learning algorithms is obtained by traversing each case dataset. Subsequently, random forest, support vector machine, and Naive Bayes models are used for training and testing, respectively. Top-1 and Top-3 accuracy are used as evaluation metrics, and cross-validation and confusion matrix analysis are employed to analyze model performance. The machine learning algorithm is compared with a text-weighted algorithm to verify the text-weighted method's effectiveness in enhancing interpretability and improving doctor credibility compared to the machine learning method.

[0046] As an example, such as Figure 2As shown, a process for matching ankle and foot orthoses is provided. First, the set of input case symptoms, the set of ankle and foot orthotic configurations to be matched, and the list of orthotic symptoms that meet the criteria are obtained. Then, it is determined whether the input case symptoms are in the list. If so, the symptom weight is obtained and the matching degree is calculated. Then, it is determined whether all orthoses have been calculated. If so, the matching degrees are sorted in descending order and the results are returned and the process ends. If the input symptoms are not in the list or all orthoses have not been calculated, the process continues to loop until the process ends.

[0047] Step 104: Select at least one valid path that matches the first ankle and foot symptom set from the expert rule knowledge graph, and the first condition node set corresponding to the at least one valid path.

[0048] The first set of condition nodes corresponds to at least one second candidate orthotics.

[0049] In some embodiments, the rule content can be determined in advance by combining the ankle-foot orthosis fitting assessment form to ensure consistency with the patient's case format. Based on the rules, knowledge extraction is completed, an expert rule knowledge graph is constructed, and stored in the Neo4j graph database.

[0050] The rule definition method and content can be determined by combining the ankle-foot orthosis fitting assessment form, which can include four parts: disease, functional abnormality manifestations, muscle strength and tone requirements, and orthotic configuration, to ensure consistency with the patient's case format. Then, the characteristics of the expert rule data structure are analyzed, and based on the storage format of the case data, the rules are extracted and stored as a dictionary file in the same format as the case. Next, to ensure that the input case symptom entities are also searchable in the expert rule knowledge graph, based on the case knowledge graph schema layer and knowledge extraction rules, the knowledge graph schema layer is defined to include six types of entities and 25 types of relationships: disease, functional assessment, muscle strength and tone assessment, examination, body parts, and orthotic devices. Knowledge extraction is then performed based on the rules, and entity alignment and fusion ensure the consistency of entities and relationships between the two knowledge graphs, completing the extraction of entities and relationships. Simultaneously, during the construction process, duplicate entity relationships are removed to ensure the uniqueness of entity relationships throughout the knowledge graph, which is then stored in the Neo4j graph database.

[0051] In some embodiments of this disclosure, step 104, which involves selecting at least one effective path from the expert rule knowledge graph that matches the first ankle-foot symptom set, may specifically include the following steps: Each symptom in the first ankle and foot symptom set is used as the starting point in the expert rule knowledge graph, and the reachable path of each starting point is determined according to the preset search depth; Filter candidate paths from all reachable paths that terminate at the orthotics node; A candidate path is determined to be a valid path if each condition node in the candidate path matches a symptom in the first ankle-foot symptom set.

[0052] In one embodiment, a path reasoning algorithm can be used to search all reachable paths in the expert rule knowledge graph with a maximum depth limit, and extract effective paths from all reachable paths, including ankle-foot orthosis nodes, path condition nodes, and patient symptoms.

[0053] In one embodiment, a normalized patient symptom entity (e.g., [cerebral palsy, ankle inversion, muscle strength 1.5]) can be used as a starting point. In an expert rule knowledge graph, a depth-first search algorithm is used, with a maximum search depth (e.g., 3 layers), to find all possible paths leading from the starting point. From all the searched paths, those whose endpoints are "orthopedic nodes" are selected. Each "condition node" on these paths is checked to see if it is satisfied by the patient's symptoms. Only when all condition nodes on a path are satisfied is the path considered a "valid path".

[0054] Step 105: Obtain the standard rule condition node for each second candidate orthodontic device from the expert rule knowledge graph to obtain the second condition node set.

[0055] It should be noted that the standard rule condition node is the preset standard condition for the second candidate orthosis, which represents the ideal fit standard of the orthosis as determined by experts.

[0056] Step 106: For each second candidate orthodontic device, determine the degree of matching between the first condition node set and the second condition node set of the second candidate orthodontic device to obtain the second matching result.

[0057] In some embodiments of this disclosure, step 106 may specifically include the following steps: Select the common condition nodes from the first condition node set and the second condition node set; Obtain the second preset weight value for each common condition node; The second preset weight values ​​of all common condition nodes are added together to obtain the second matching score of the second candidate orthotics; All second candidate orthotics are sorted according to the second matching degree score to obtain the second matching result.

[0058] In one embodiment, by querying the condition nodes of each second candidate orthosis in the expert rule knowledge graph and matching them with the first condition node set of the second candidate orthosis, a second preset weight value is assigned to the common nodes of the two. The second preset weight value can be obtained based on the doctor's advice or by calculating and comparing the accuracy under different weights through relevant algorithms. Then, the second matching degree score is obtained by weighting the weights. The orthosis configuration, the content of the matching rule, and the search path are returned to the front end to complete the recommendation of ankle and foot orthosis.

[0059] As an example, patient case data acquired from the front end can be standardized and mapped according to expert rules, such as mapping 'Level 1' to '1', 'Level 1+' to '1.5', and 'Level 2' to '2', to ensure consistent format. The standardized case symptom entities are then input into the expert rule knowledge graph, and a depth-first search method is used, with a maximum search depth of 3, to search all possible paths. Paths containing orthotics nodes are extracted, and then condition nodes within these paths are extracted. It is determined whether the condition node satisfies the patient's symptoms; if so, the path is valid; otherwise, it is invalid. Valid paths are categorized by orthotics, and condition nodes are extracted from each category to obtain the first set of condition nodes A1, which contains the orthotics and their path condition nodes.

[0060] Furthermore, in the expert rule knowledge graph, the rule number corresponding to each type of orthotic configuration is obtained using a relational query method. Then, the second condition node set A2 that each orthotic configuration must satisfy is obtained using the relational query method. For the effective path condition node set A1 of each orthotic, A2 obtained by querying the knowledge graph is traversed, common nodes are filtered, and they are assigned a second preset weight value set according to the doctor's advice. The weight settings of the three conditions, namely disease, functional abnormality, and muscle strength and tone requirements, are all defined in the range of 0-1. Combined with the algorithm, the importance of the three parts is determined as follows: weight of functional abnormality > weight of disease > weight of muscle strength and tone level. The weight of disease is set to 0.4, the weight of functional abnormality is set to 0.6, and the weight of muscle strength and tone level requirements is set to 0.2. Then, the first matching degree score of each rule is calculated by traversing and weighting. The top three orthotic configurations with the highest scores are selected. The orthotic configurations corresponding to the descending rule matching degree, the matching rule content, and the path obtained by the algorithm search are returned to the front end to complete the ankle and foot orthotic recommendation for the case.

[0061] In one embodiment, the results obtained from path reasoning (the first set of condition nodes) are combined with the results from the rule engine (the second set of condition nodes). Based on the rule descriptions and search paths provided by both, the doctor determines whether the recommended results are suitable for the patient and modifies and improves the rules accordingly.

[0062] As an example, such as Figure 3 As shown, a rule-based decision-making process for ankle-foot orthoses is provided. First, the set of symptoms of the input cases is obtained, and path search and extraction of valid paths are performed sequentially. Then, it is determined whether the path condition nodes match the patient's symptoms. If they match, the valid path is stored, path AFO classification is performed, and each rule condition node is queried. Next, it is determined whether the path condition node exists in the rule condition node. If it exists, the node weight is obtained, the rule score is calculated, the scores are sorted in descending order, and the result is returned, and the process ends. If the intermediate judgment is not satisfied, the next node is judged until the process is completed.

[0063] As another example, such as Figure 4 As shown, a process for matching rules for ankle and foot orthoses is provided. First, the patient's symptoms are selected, and then the rules are matched. If the rules are matched, the orthoses and rule descriptions are inferred, and the process ends after the patient information is stored. If the rules are not matched, a message is displayed indicating that no orthoses were matched, and the process ends after the patient information is stored.

[0064] Step 107: If there is no conflict between the first matching result and the second matching result, generate an auxiliary decision for ankle-foot orthosis fitting of the object based on the first matching result and the second matching result.

[0065] In one embodiment, to facilitate rule matching, the expert rules are first extracted into a dictionary format using a rule language, taking into account the structure and language characteristics of the expert rules. For example, the key value corresponding to the disease cerebral palsy is 'Dynamic Ankle-Foot Orthosis (DAFO), Static Ankle-Foot Orthosis, Hinged Ankle-Foot Orthosis, Ground Reaction Orthosis'. Then, a script is written to build a rule engine, and a GUI human-computer interaction interface is designed. This interface includes all orthotics purposes, symptoms, diseases, and muscle strength / muscle tone conditions in the expert rules. By selecting patient symptoms, rule matching is performed after statistical processing. If the symptoms exist in the rules, an orthosis recommendation and rule explanation will be provided. Otherwise, the system will indicate that no matching orthosis was found and display the rule conflict.

[0066] In some embodiments of this disclosure, prior to step 107, the method may further include the following steps: If the first candidate orthosis with the highest first matching score is the same as the second candidate orthosis with the highest second matching score, it is determined that there is no conflict between the first matching result and the second matching result.

[0067] As an example, through the web front-end, patient information is filled in according to the assessment form in the patient assessment module. In the prescription recommendation module, the dual-engine reasoning algorithm based on cases and rules can obtain ankle and foot orthotic recommendation results based on cases and expert rules respectively. It also includes corresponding supporting explanations, such as orthotic matching degree, rule matching degree, rule matching explanation, and graph search path display, etc. When the two recommendation results are consistent, the system directly gives the optimal matching result. If the two conflict, a doctor arbitration interface will pop up, and the doctor will make the final recommendation according to the recommendation explanation.

[0068] In some embodiments of this disclosure, such as Figure 5 As shown, a hierarchical architecture for an auxiliary decision-making method for ankle-foot orthosis fitting proposed in this disclosure is provided, consisting of a data layer, a dual-engine inference layer, and a decision output layer from bottom to top. The data layer collects data through an evaluation table, and after desensitization, cleaning, and knowledge extraction, stores the case knowledge graph and expert rule knowledge graph in the Neo4j database. The dual-engine inference layer operates in parallel; the case engine generates recommendation results using weighted matching based on the TF-IDF algorithm, while the expert rule engine completes the recommendation through path reasoning algorithms and the rule engine. The decision output layer receives the dual-engine inference results; if the results are consistent, a prescription is directly output; if there is a conflict, the final prescription is determined through a doctor arbitration interface.

[0069] In some embodiments of this disclosure, such as Figure 6 As shown, a reasoning process for assisting decision-making in ankle-foot orthosis fitting is provided. First, the patient data (e.g., patient case) input from the front end is preprocessed. Then, engine reasoning based on case knowledge graph is carried out in parallel (generating recommendation results through TF-IDF-based weight calculation, matching degree calculation, and descending matching degree sorting). Rule engine recommendation is performed using expert rules (through rule extraction, rule matching, and orthosis output results). Engine reasoning based on expert rule knowledge graph is performed (generating results through path search, effective path extraction, rule condition query, and rule score calculation). Finally, result conflict detection is performed. If there is no conflict, the result is output; if there is a conflict, the decision is made by doctor arbitration.

[0070] In other embodiments of this disclosure, such as Figure 7 As shown, a complete process for assisting decision-making in ankle and foot orthosis fitting is provided. First, the front end fills out an assessment form and performs data standardization processing. Then, it enters a dual-engine parallel reasoning stage. The ankle and foot case knowledge graph engine generates a recommended solution after data anonymization and cleaning, knowledge graph query, and TF-IDF weighted matching. The expert rule knowledge graph engine generates a recommended solution through path search, effective path extraction, rule condition query, and rule score selection. After that, it is determined whether the results of the two engines are consistent. If they are consistent, a prescription is output and the process ends. If they are inconsistent, the process ends after doctor arbitration and prescription confirmation.

[0071] According to the auxiliary decision-making method for ankle-foot orthosis fitting proposed in this disclosure, the method involves: determining at least one first candidate orthosis matching a first ankle-foot symptom set from an ankle-foot case knowledge graph, and a second ankle-foot symptom set associated with each first candidate orthosis; determining the degree of matching between the first ankle-foot symptom set and each second ankle-foot symptom set to obtain a first matching result; selecting at least one valid path matching the first ankle-foot symptom set from an expert rule knowledge graph, and a first condition node set corresponding to the at least one valid path; obtaining standard rule condition nodes for each second candidate orthosis from the expert rule knowledge graph to obtain a second condition node set; determining the degree of matching between the first condition node set and the second condition node set for the second candidate orthosis to obtain a second matching result; and generating an auxiliary decision for ankle-foot orthosis fitting based on the first and second matching results, provided there is no conflict between the first and second matching results. By employing a dual-reasoning collaborative approach based on ankle and foot case knowledge graphs and expert rule knowledge graphs, the advantages of data-driven and knowledge-driven approaches are complemented. Furthermore, the dual-reasoning collaborative approach enables auxiliary decision-making results to have dual interpretability, thereby allowing users to clearly understand the decision-making logic of ankle and foot orthosis fitting and improving the credibility of ankle and foot orthosis fitting decisions.

[0072] Figure 8 This is a block diagram illustrating an assistive decision-making device for ankle-foot orthosis fitting according to an exemplary embodiment. (Refer to...) Figure 8 The device includes a first acquisition unit 801, a determination unit 802, a first matching unit 803, a selection unit 804, a second acquisition unit 805, a second matching unit 806, and a decision unit 807.

[0073] The first acquisition unit 801 is used to acquire the first ankle and foot symptom set, the ankle and foot case knowledge graph, and the expert rule knowledge graph associated with the ankle and foot orthosis, respectively; the ankle and foot case knowledge graph is generated based on historical ankle and foot orthosis fitting cases; the expert rule knowledge graph is generated based on expert knowledge. The determining unit 802 is used to determine at least one first candidate orthosis that matches a first ankle and foot symptom set from an ankle and foot case knowledge graph, and a second ankle and foot symptom set associated with each first candidate orthosis; The first matching unit 803 is used to determine the degree of matching between the first ankle and foot symptom set and each second ankle and foot symptom set respectively, and to obtain the first matching result; The selection unit 804 is used to select at least one valid path that matches the first ankle and foot symptom set from the expert rule knowledge graph, and a first condition node set corresponding to the at least one valid path; the first condition node set corresponds to at least one second candidate orthosis. The second acquisition unit 805 is used to acquire the standard rule condition node of each second candidate orthotics from the expert rule knowledge graph, and obtain the second condition node set. The second matching unit 806 is used to determine the degree of matching between the first condition node set and the second condition node set of each second candidate orthodont, and to obtain the second matching result. Decision unit 807 is used to generate an auxiliary decision for ankle-foot orthosis fitting of an object based on the first matching result and the second matching result, provided that there is no conflict between the first matching result and the second matching result.

[0074] In some embodiments of this disclosure, the first matching unit 803 may specifically be used for: In some embodiments of this disclosure, the apparatus further includes a computing unit, which may specifically be used for: For the first candidate orthosis corresponding to the first preset weight value, determine the first frequency of the target symptom corresponding to the first preset weight value in the cases associated with the first candidate orthosis; Identify the second most frequent occurrence of the target symptom across all orthodontic-associated cases; Calculate the product of the first frequency and the second frequency to obtain the first preset weight value of the target symptom.

[0075] In some embodiments of this disclosure, the selection unit 804 may specifically be used for: Each symptom in the first ankle and foot symptom set is used as the starting point in the expert rule knowledge graph, and the reachable path of each starting point is determined according to the preset search depth; Filter candidate paths from all reachable paths that terminate at the orthotics node; A candidate path is determined to be a valid path if each condition node in the candidate path matches a symptom in the first ankle-foot symptom set.

[0076] In some embodiments of this disclosure, the second matching unit 806 may specifically be used for: Select the common condition nodes from the first condition node set and the second condition node set; Obtain the second preset weight value for each common condition node; The second preset weight values ​​of all common condition nodes are added together to obtain the second matching score of the second candidate orthotics; All second candidate orthotics are sorted according to the second matching degree score to obtain the second matching result.

[0077] In some embodiments of this disclosure, the apparatus further includes a conflict determination unit, which can be specifically used to: determine that there is no conflict between the first matching result and the second matching result when the first candidate orthodontic with the highest first matching degree score is the same as the second candidate orthodontic with the highest second matching degree score.

[0078] In some embodiments of this disclosure, the apparatus further includes a generation unit, which may specifically be used for: Historical ankle-foot orthosis fitting cases are obtained, and knowledge extraction is performed on these cases to obtain target entities and the relationships between them. Target entities include symptoms and orthosis types. An ankle and foot case knowledge graph is generated based on the target entities and the relationships between them.

[0079] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0080] According to the auxiliary decision-making device for ankle-foot orthosis fitting proposed in the embodiments of this disclosure, the device determines at least one first candidate orthosis matching a first ankle-foot symptom set from an ankle-foot case knowledge graph, and a second ankle-foot symptom set associated with each first candidate orthosis; determines the degree of matching between the first ankle-foot symptom set and each second ankle-foot symptom set to obtain a first matching result; selects at least one valid path matching the first ankle-foot symptom set from an expert rule knowledge graph, and a first condition node set corresponding to the at least one valid path; obtains the standard rule condition node for each second candidate orthosis from the expert rule knowledge graph to obtain a second condition node set; determines the degree of matching between the first condition node set and the second condition node set of the second candidate orthosis to obtain a second matching result; and, if there is no conflict between the first matching result and the second matching result, generates an auxiliary decision for ankle-foot orthosis fitting of the object based on the first matching result and the second matching result. By employing a dual-reasoning collaborative approach based on ankle and foot case knowledge graphs and expert rule knowledge graphs, the advantages of data-driven and knowledge-driven approaches are complemented. Furthermore, the dual-reasoning collaborative approach enables auxiliary decision-making results to have dual interpretability, thereby allowing users to clearly understand the decision-making logic of ankle and foot orthosis fitting and improving the credibility of ankle and foot orthosis fitting decisions.

[0081] Figure 9This is a block diagram illustrating an apparatus for an assistive decision-making method for ankle-foot orthosis fitting, according to an exemplary embodiment. For example, apparatus 900 may be an electronic device, such as a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0082] Reference Figure 9 The device 900 may include one or more of the following components: a processing component 902, a memory 904, a power component 906, a multimedia component 908, an audio component 910, an input / output (I / O) interface 912, a sensor component 914, and a communication component 916.

[0083] Processing component 902 typically controls the overall operation of device 900, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 902 may include one or more processors 920 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 902 may include one or more modules to facilitate interaction between processing component 902 and other components. For example, processing component 902 may include a multimedia module to facilitate interaction between multimedia component 908 and processing component 902.

[0084] Memory 904 is configured to store various types of data to support the operation of device 900. Examples of this data include instructions for any application or method operating on device 900, contact data, phonebook data, messages, pictures, videos, etc. Memory 904 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0085] The power supply component 906 provides power to the various components of the device 900. The power supply component 906 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device 900.

[0086] Multimedia component 908 includes a screen that provides an output interface between the device 900 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 908 includes a front-facing camera and / or a rear-facing camera. When the device 900 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0087] Audio component 910 is configured to output and / or input audio signals. For example, audio component 910 includes a microphone (MIC) configured to receive external audio signals when device 900 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 904 or transmitted via communication component 916. In some embodiments, audio component 910 also includes a speaker for outputting audio signals.

[0088] I / O interface 912 provides an interface between processing component 902 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0089] Sensor assembly 914 includes one or more sensors for providing status assessments of various aspects of device 900. For example, sensor assembly 914 may detect the on / off state of device 900, the relative positioning of components such as the display and keypad of device 900, changes in position of device 900 or a component of device 900, the presence or absence of user contact with device 900, orientation or acceleration / deceleration of device 900, and temperature changes of device 900. Sensor assembly 914 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 914 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 914 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0090] Communication component 916 is configured to facilitate wired or wireless communication between device 900 and other devices. Device 900 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 916 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 916 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0091] In an exemplary embodiment, the apparatus 900 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0092] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 904 including instructions, which can be executed by a processor 920 of the device 900 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0093] In an exemplary embodiment, a computer program product is also provided, including a computer program that implements the above-described method when executed by a processor 920 of the device 900.

[0094] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0095] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An auxiliary decision-making method for ankle-foot orthosis fitting, characterized in that, include: Obtain the object's first ankle and foot symptom set, ankle and foot case knowledge graph, and expert rule knowledge graph associated with the ankle and foot orthosis, respectively; The ankle-foot case knowledge graph was generated based on historical ankle-foot orthotic fitting cases; the expert rule knowledge graph was generated based on expert knowledge. Identify at least one first candidate orthosis that matches the first ankle and foot symptom set from the ankle and foot case knowledge graph, and a second ankle and foot symptom set associated with each first candidate orthosis; The degree of matching between the first ankle and foot symptom set and each second ankle and foot symptom set is determined to obtain the first matching result; Select at least one valid path from the expert rule knowledge graph that matches the first ankle and foot symptom set, and a first condition node set corresponding to the at least one valid path; The first set of conditional nodes corresponds to at least one second candidate orthodontic device; The standard rule condition nodes for each second candidate orthotics are obtained from the expert rule knowledge graph to obtain the second condition node set; For each second candidate orthodontic device, the degree of matching between the first condition node set and the second condition node set of the second candidate orthodontic device is determined to obtain the second matching result; If there is no conflict between the first matching result and the second matching result, an auxiliary decision for ankle-foot orthosis fitting of the object is generated based on the first matching result and the second matching result.

2. The auxiliary decision-making method for ankle-foot orthosis fitting according to claim 1, characterized in that, The step of determining the degree of matching between the first ankle-foot symptom set and each second ankle-foot symptom set to obtain a first matching result includes: For each second ankle and foot symptom set, select the target symptom that exists in the second ankle and foot symptom set from all the symptoms in the first ankle and foot symptom set; Obtain the first preset weight value for each target symptom; The first preset weight values ​​corresponding to all target symptoms in the second ankle and foot symptom set are added together to obtain the first matching score of the second ankle and foot symptom set. All first candidate orthotics are sorted according to the first matching degree score to obtain the first matching result.

3. The auxiliary decision-making method for ankle-foot orthosis fitting according to claim 2, characterized in that, The first preset weight value is calculated using the following steps: For the first candidate orthosis corresponding to the first preset weight value, determine the first frequency of occurrence of the target symptom corresponding to the first preset weight value in the cases associated with the first candidate orthosis; Determine the second frequency of the target symptom in all orthodontic-associated cases; The product of the first frequency and the second frequency is calculated to obtain the first preset weight value of the target symptom.

4. The auxiliary decision-making method for ankle-foot orthosis fitting according to claim 1, characterized in that, The step of selecting at least one effective path from the expert rule knowledge graph that matches the first ankle-foot symptom set includes: Each symptom in the first ankle and foot symptom set is used as the starting point in the expert rule knowledge graph, and the reachable path of each starting point is determined according to the preset search depth; Filter candidate paths from all reachable paths that terminate at the orthotics node; The candidate path is determined to be a valid path if each condition node in the candidate path matches a symptom in the first ankle and foot symptom set.

5. The auxiliary decision-making method for ankle-foot orthosis fitting according to claim 1, characterized in that, Determine the degree of matching between the first set of condition nodes and the second set of condition nodes for the second candidate orthotics to obtain a second matching result, including: Select the common condition nodes from the first set of condition nodes and the second set of condition nodes; Obtain the second preset weight value for each common condition node; The second preset weight values ​​of all common condition nodes are added together to obtain the second matching degree score of the second candidate orthotics; All second candidate orthotics are sorted according to the second matching degree score to obtain the second matching result.

6. The auxiliary decision-making method for ankle-foot orthosis fitting according to claim 1, characterized in that, Before generating an auxiliary decision for ankle-foot orthosis fitting of the subject based on the first matching result and the second matching result, the method further includes: If the first candidate orthosis with the highest first matching score is the same as the second candidate orthosis with the highest second matching score, it is determined that there is no conflict between the first matching result and the second matching result.

7. The auxiliary decision-making method for ankle-foot orthosis fitting according to claim 1, characterized in that, The ankle-foot case knowledge graph was generated using the following steps: Historical ankle-foot orthosis fitting cases are obtained, and knowledge extraction is performed on these cases to obtain target entities and the relationships between them; the target entities include symptoms and orthosis types. The ankle and foot case knowledge graph is generated based on the target entities and the relationships between them.

8. An auxiliary decision-making device for ankle-foot orthosis fitting, characterized in that, include: The first acquisition unit is used to acquire the first ankle and foot symptom set, ankle and foot case knowledge graph, and expert rule knowledge graph associated with the ankle and foot orthosis of the object, respectively. The ankle-foot case knowledge graph was generated based on historical ankle-foot orthotic fitting cases; the expert rule knowledge graph was generated based on expert knowledge. The determining unit is configured to determine, from the ankle and foot case knowledge graph, at least one first candidate orthosis that matches the first ankle and foot symptom set, and a second ankle and foot symptom set associated with each first candidate orthosis; The first matching unit is used to determine the degree of matching between the first ankle and foot symptom set and each second ankle and foot symptom set, and to obtain the first matching result; The selection unit is used to select at least one valid path from the expert rule knowledge graph that matches the first ankle and foot symptom set, and a first condition node set corresponding to the at least one valid path; The first set of conditional nodes corresponds to at least one second candidate orthodontic device; The second acquisition unit is used to acquire the standard rule condition node of each second candidate orthotics from the expert rule knowledge graph, and obtain the second condition node set; The second matching unit is used to determine the degree of matching between the first set of condition nodes of the second candidate orthodont and the second set of condition nodes for each second candidate orthodont, and to obtain the second matching result; The decision unit is configured to generate an auxiliary decision for fitting the object with an ankle-foot orthosis based on the first matching result and the second matching result, provided that there is no conflict between the first matching result and the second matching result.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images