A Multi-Timescale Rehabilitation Evaluation Method and System Based on Fuzzy Hierarchical Analysis

Abstract

This invention relates to the field of rehabilitation assessment, providing a multi-timescale rehabilitation assessment method and system based on fuzzy hierarchical analysis (AHP), comprising: S1: constructing a hierarchical structure for rehabilitation assessment, including an information layer, a factor layer, a criterion layer, and a target layer; S2: constructing the association between the factor layer and the criterion layer to obtain the criterion layer evaluation vector; S3: constructing the association between the criterion layer and the target layer through the criterion layer evaluation vector to obtain the target layer evaluation vector; S4: obtaining the final rehabilitation assessment result through the target layer evaluation vector. This invention provides a feasible consistency check and adjustment method for the fuzzy judgment matrix of fuzzy hierarchical analysis, making the analysis results of fuzzy hierarchical analysis more accurate; and conducting rehabilitation assessments at different time scales allows the rehabilitation assessment to meet real-time requirements and improves the accuracy of the rehabilitation assessment.

Description

Technical Field

[0001] This invention relates to the field of rehabilitation assessment, and in particular to a multi-timescale rehabilitation assessment method and system based on fuzzy hierarchical analysis. Background Technology

[0002] Due to the high incidence of accidents and the accelerating aging process, the demand for rehabilitation therapy is constantly increasing. With the development of bio-mechatronics technology, rehabilitation robots have become an important technical means for rehabilitation therapy. Rehabilitation assessment is a crucial part of rehabilitation therapy and plays a guiding role in rehabilitation prescription. Robot-assisted rehabilitation therapy often involves real-time motion assessment as well as assessments at different time scales, such as assessments of motor function recovery at different treatment cycles.

[0003] However, existing rehabilitation assessment methods, due to their fixed time scale, cannot conduct rehabilitation assessments at different time scales, resulting in the inability of existing rehabilitation assessments to meet the requirements of real-time performance, and the accuracy of rehabilitation assessments is also affected.

[0004] Furthermore, fuzzy hierarchical analysis is a multi-objective decision-making method based on fuzzy rules, which uses fuzzy rules to solve the consistency problem in traditional hierarchical analysis.

[0005] However, existing rehabilitation assessments using fuzzy analytic hierarchy process (AHP) only provide the testing methods without proposing specific adjustment algorithms, resulting in inaccurate results from AHP and consequently, inaccurate rehabilitation assessments.

[0006] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis, comprising:

[0008] S1: Construct a hierarchical structure for rehabilitation assessment, which includes: information layer, factor layer, criterion layer, and goal layer;

[0009] S2: Construct the relationship between the factor layer and the criterion layer to obtain the criterion layer evaluation vector;

[0010] S3: Construct the relationship between the criterion layer and the target layer through the criterion layer evaluation vector to obtain the target layer evaluation vector;

[0011] S4: Obtain the final rehabilitation evaluation result through the target layer evaluation vector.

[0012] Preferably, step S1 specifically includes:

[0013] Acquire multi-source individual information in lower limb rehabilitation scenarios, and construct an information layer based on this multi-source individual information;

[0014] Set evaluation indicators and construct a factor layer based on these indicators;

[0015] Set criteria indicators and construct a criteria layer based on these indicators;

[0016] The hierarchical structure of rehabilitation evaluation, from bottom to top, is as follows: information layer - factor layer - criterion layer - goal layer.

[0017] Preferably, step S2 specifically includes:

[0018] S21: Standardize the evaluation indicators to obtain standardized evaluation indicators. The standardized evaluation indicators constitute a standardized factor layer. The standardized evaluation indicators are scored by individual multi-source information in the information layer.

[0019] S22: Construct a fuzzy judgment matrix for the criteria indicators of the criteria layer, and calculate the weight matrix of the criteria evaluation indicators through the fuzzy judgment matrix of the criteria layer.

[0020] S23: Construct the criterion layer weight vector matrix, and calculate the criterion layer evaluation vector through the criterion layer evaluation index weight matrix and the criterion layer weight vector matrix.

[0021] Preferably, step S22 specifically includes:

[0022] S221: The fuzzy judgment matrix of the criterion layer is represented as follows:

[0023] Where B represents the criterion layer, i is the criterion indicator number, a and b are both standardized evaluation indicator numbers, and r ab To determine the relative importance of standardized evaluation indicator a to standardized evaluation indicator b, n i Let a and b be the total number of standardized evaluation indicators for criterion indicator i, where a and b are both in the range of n. i Within the range;

[0024] S222: Will Convert to matrix in For matrix K i Perform a consistency check. If the consistency check passes, proceed to step S225; otherwise, proceed to step S223.

[0025] S223: Calculate and obtain matrix K i eigenvectors Constructing a matrix Where α∈(0,1);

[0026] S224: Calculate the fuzzy judgment matrix of the criterion layer that satisfies consistency. The calculation formula is:

[0027] in use replace Return to step S222;

[0028] S225: Weight Matrix of Evaluation Indicators at the Criterion Level W i B The calculation formula is:

[0029]

[0030]

[0031]

[0032] Where i is from 1 to n i Positive integers.

[0033] Preferably, step S23 specifically includes:

[0034] S231: The weight vector matrix of the criterion layer is represented as follows:

[0035] Where B represents the criterion layer, i is the criterion indicator number, a and b are both standardized evaluation indicator numbers, and e ab The proportion of standardized evaluation index b obtained by standardized evaluation index a under the selected time scale is the standard index i.

[0036] S232: Criterion Layer Evaluation Vector S i B The calculation formula is:

[0037]

[0038] in, s1 to Let be the membership degree of criterion index i.

[0039] Preferably, step S3 specifically includes:

[0040] S31: The target layer is scored by the criteria indicators in the criteria layer, and the final rehabilitation evaluation result is standardized by the scoring to obtain the standardized final rehabilitation evaluation result. The standardized target layer is composed of the standardized final rehabilitation evaluation result.

[0041] S32: Construct a standardized target layer fuzzy judgment matrix for the target layer, and calculate the target layer evaluation index weight matrix through the target layer fuzzy judgment matrix;

[0042] S33: Obtain the target layer weight vector matrix by calculating the criterion layer evaluation vector, and obtain the target layer evaluation vector by calculating the target layer evaluation index weight matrix and the target layer weight vector matrix.

[0043] Preferably, step S32 specifically includes:

[0044] S321: The target layer fuzzy judgment matrix is ​​represented as: R A =(r′) df ) c×c ;

[0045] Where A represents the evaluation layer, d and f are the index numbers of the criteria indicators, and r' ab Let c represent the relative importance of criterion indicator a to criterion indicator b, c represent the total number of criterion indicators, and d and f both fall within the range of c.

[0046] S322: R A Convert to matrix K = (k′) df ) c×c ,in Perform a consistency check on matrix K. If the consistency check passes, proceed to step S325; otherwise, proceed to step S323.

[0047] S323: Calculate the eigenvectors P′=[p′1,p′2,…,p′] of matrix K. c ] T Constructing a matrix Where β∈(0,1);

[0048] S324: Calculate the fuzzy judgment matrix of the criterion layer that satisfies consistency. The calculation formula is:

[0049] in use Replace R A Return to step S322;

[0050] S325: Weight Matrix of Evaluation Indicators at the Criterion Level W A The calculation formula is:

[0051] W A =[w′1,w′2,…w′ a …, w′ c ]

[0052]

[0053]

[0054] Preferably, step S33 specifically includes:

[0055] S331: Target layer weight vector matrix E A The calculation formula is:

[0056]

[0057] Where U is the criterion-level evaluation vector S i B The maximum value of the index number i in the standard;

[0058] S332: Target layer evaluation vector S A The calculation formula is:

[0059] S A =W A E A

[0060] Among them, W A This is the weight matrix of the evaluation indicators for the target layer.

[0061] Preferably, step S4 specifically includes:

[0062] A quantitative score vector G is obtained based on rehabilitation assessment needs;

[0063] The formula for calculating the final rehabilitation evaluation result Y is:

[0064] Y = S A G T

[0065] Among them, S A This is the evaluation vector for the target layer.

[0066] A multi-timescale rehabilitation assessment system based on fuzzy hierarchical analysis includes:

[0067] The structural construction module is used to build a hierarchical structure for rehabilitation evaluation, which includes: information layer, factor layer, criterion layer and goal layer.

[0068] The first correlation module is used to construct the correlation between the factor layer and the criterion layer to obtain the criterion layer evaluation vector;

[0069] The second association module is used to construct the association between the criterion layer and the target layer through the criterion layer evaluation vector, and obtain the target layer evaluation vector.

[0070] The evaluation module is used to obtain the final rehabilitation evaluation result through the target layer evaluation vector.

[0071] The present invention has the following beneficial effects:

[0072] 1. This invention provides a feasible consistency check and adjustment method for the fuzzy judgment matrix of fuzzy hierarchical analysis, making the analysis results of fuzzy hierarchical analysis more accurate;

[0073] 2. Conducting rehabilitation assessments at different time scales allows the assessments to meet real-time requirements and improves their accuracy. Attached Figure Description

[0074] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0075] Figure 2 A structural diagram of the hierarchical structure of rehabilitation assessment;

[0076] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0077] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0078] Reference Figure 1 This invention provides a multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis, comprising:

[0079] S1: Construct a hierarchical structure for rehabilitation assessment, which includes: information layer, factor layer, criterion layer, and goal layer;

[0080] S2: Construct the relationship between the factor layer and the criterion layer to obtain the criterion layer evaluation vector;

[0081] S3: Construct the relationship between the criterion layer and the target layer through the criterion layer evaluation vector to obtain the target layer evaluation vector;

[0082] S4: Obtain the final rehabilitation evaluation result through the target layer evaluation vector.

[0083] In this embodiment, in the robot-assisted lower limb rehabilitation training scenario, the information that needs to be acquired mainly includes: limb movement information, contact force with the rehabilitation robot, and bioelectrical information during the movement process; wherein, the limb movement information includes the movement position, velocity, and acceleration of the hip, knee, and ankle joints of the lower limbs, which are collected by images and seven inertial sensors located on the waist, left and right thighs, left and right calves, and left and right insteps respectively; the contact force with the rehabilitation robot is the contact force between the end pedal of the rehabilitation robot and the contact surface of the foot, which is collected by pressure sensors placed on the pedal; the bioelectrical information includes electrocardiogram (ECG) and lower limb electromyography (EMG) signals, the ECG signals are collected by a wrist ECG meter worn by the individual, and the lower limb EMG signals are collected by eight surface-mounted EMG sensors distributed on the tibialis anterior, gastrocnemius, rectus femoris, and biceps femoris muscles of both lower limbs;

[0084] Step S1 is as follows:

[0085] Acquire multi-source individual information in lower limb rehabilitation scenarios, and construct an information layer based on this multi-source individual information;

[0086] Set evaluation indicators and construct a factor layer based on these indicators;

[0087] Set criteria indicators and construct a criteria layer based on these indicators;

[0088] The hierarchical structure of rehabilitation evaluation, from bottom to top, is as follows: information layer - factor layer - criterion layer - goal layer.

[0089] Specifically, such as Figure 2 The diagram shown is a structural diagram of the hierarchical structure of rehabilitation assessment.

[0090] The individual multi-source information in information layer D includes: joint position D1, movement velocity D2, movement acceleration D3, contact force D4, electromyography D5, and electrocardiogram D6;

[0091] Evaluation indicators for factor layer C include muscle strength (C1), muscle tone (C2), joint range of motion (C3), muscle coordination (C4), movement smoothness (C5), and fatigue level (C6).

[0092] The criteria indicators for criterion layer B include: sarcomere evaluation B1 and overall motion evaluation B2;

[0093] The target layer outputs the final rehabilitation evaluation results;

[0094] Based on the needs of lower limb rehabilitation assessment, a four-layer structure is constructed from bottom to top: information layer, factor layer, criterion layer, and target layer.

[0095] In this embodiment, step S2 specifically includes:

[0096] S21: Standardize the evaluation indicators to obtain standardized evaluation indicators. The standardized evaluation indicators constitute a standardized factor layer. The standardized evaluation indicators are scored by individual multi-source information in the information layer.

[0097] Specifically, in order to obtain consistent quantitative scores, it is necessary to standardize each indicator. First, a five-level evaluation set V = {v1, v2, v3, v4, v5} is set, where v1 to v5 represent five evaluation levels: very poor, poor, average, good, and very good, respectively. Second, the evaluation indicators of the factor layer are standardized, as follows.

[0098] Standardization of muscle strength C1: The actual measured torque value (F) is compared with the standard value (Fnor) of the healthy limb, and scored according to the criteria in Table 1. The scoring is on a five-point scale, and the scoring criteria are as follows:

[0099] Table 1 Muscle Strength Scoring

[0100] Muscle strength recovery assessment level 5 (Very good) <![CDATA[F-F nor ≤3]]> 4 (Good) <![CDATA[3<F-F nor ≤5]]> 3 (General) <![CDATA[5<F-F nor ≤10]]> 2 (Poor) <![CDATA[10<F-F nor ≤15]]> 1 (Very poor) <![CDATA[F-F nor >15]]>

[0101] Standardization of muscle tone C2: Similar to muscle tone, a five-level score is determined based on the muscle tone value.

[0102] Standardized processing of joint range of motion (ROM) C3: The ratio of the patient's ROM to that of a normal individual was used as an indicator for evaluating joint rehabilitation. Lower limb ROM was measured including the hip and knee joints. Hip ROM was measured as follows: flexion 125 degrees, extension 15 degrees, adduction 35 degrees, abduction 45 degrees, and internal / external rotation 45 degrees each. Knee ROM was measured as follows: knee flexion and extension 0–130 degrees. Patients' ROM values ​​were recorded, and a five-point scoring system was used, as shown in Table 2.

[0103] Table 2 Joint Range of Motion Scoring

[0104] Joint recovery evaluation level 5 (Very good) ROM / ROM≥90% 4 (Good) 80% ≤ ROM / ROM < 90% 3 (General) 60% ≤ ROM / ROM < 80% 2 (Poor) 40% ≤ ROM / ROM < 60% 1 (Very poor) ROM / ROM < 40%

[0105] Standardization of muscle coordination C4: Biomechanical modeling was performed using motion information from various joints of the lower limbs. The theoretical activation of four selected key muscles was analyzed, and the actual activation was obtained through electromyography. The two were compared and scored on a five-point scale, with 5 points indicating that the theoretical situation is completely consistent with the actual situation and 1 point indicating that the actual situation is completely inconsistent.

[0106] Standardization of motion smoothness C5: Using motion information of each joint in the lower limb, calculate the acceleration, spectral arc length and number of discontinuous pauses during the motion process, and score it on a five-point scale, with 5 being the best and 1 being the worst.

[0107] Standardization of fatigue level C6: Principal component analysis was used to determine characteristic parameters related to muscle fatigue. ECG pulse rate, total distance traveled, and electromyographic signal characteristics were selected. Fisher discriminant analysis was used to determine the discriminant function for muscle fatigue. A five-point scoring system was applied, assigning 5 points to normal muscle fatigue, 3 points to fatigue, 1 point to severe fatigue, and 2 and 4 points to intermediate values.

[0108] The information directly collected from the factor layer is processed to obtain real-time quantities related to the above indicators. By comparing with the scoring criteria, the scores of each standardized evaluation indicator can be obtained.

[0109] S22: Construct a fuzzy judgment matrix for the criteria indicators of the criteria layer, and calculate the weight matrix of the criteria evaluation indicators through the fuzzy judgment matrix of the criteria layer.

[0110] S23: Construct the criterion layer weight vector matrix, and calculate the criterion layer evaluation vector through the criterion layer evaluation index weight matrix and the criterion layer weight vector matrix.

[0111] In this embodiment, step S22 specifically includes:

[0112] S221: The fuzzy judgment matrix of the criterion layer is represented as follows:

[0113] Where B represents the criterion layer, i is the criterion indicator number, a and b are both standardized evaluation indicator numbers, and r ab To determine the relative importance of standardized evaluation indicator a to standardized evaluation indicator b, n i Let a and b be the total number of standardized evaluation indicators for criterion indicator i, where a and b are both in the range of n. i Within the range;

[0114] Specifically, for example, if i = 1, then It can be expanded to the following form:

[0115]

[0116] The importance of the standardized evaluation indicators is calibrated using a scale of 0.1-0.9, and the quantitative standards are shown in Table 3 below:

[0117] Table 3: Quantitative Standards for Importance Assessment

[0118]

[0119]

[0120] S222: Will Convert to matrix in For matrix K i Perform a consistency check. If the consistency check passes, proceed to step S225; otherwise, proceed to step S223.

[0121] Specifically, consistency testing refers to the use of consistency indices CI and CR for verification, which are:

[0122]

[0123] Where, λ max For matrix The largest eigenvalue, RI, can be obtained by referring to Table 4 below;

[0124] Table 4 RI Values

[0125]

[0126] When CR < 0.1, the consistency test is passed; when CR ≥ 0.1, the first-time test is failed, and the matrix needs to be adjusted.

[0127] S223: Calculate and obtain matrix K i eigenvectors Constructing a matrix Where α∈(0,1);

[0128] S224: Calculate the fuzzy judgment matrix of the criterion layer that satisfies consistency. The calculation formula is:

[0129] in use replace Return to step S222;

[0130] S225: Weight Matrix of Evaluation Indicators at the Criterion Level W i B The calculation formula is:

[0131]

[0132]

[0133]

[0134] Where i is from 1 to n i Positive integers.

[0135] In this embodiment, step S23 specifically includes:

[0136] S231: The weight vector matrix of the criterion layer is represented as follows:

[0137] Where B represents the criterion layer, i is the criterion indicator number, a and b are both standardized evaluation indicator numbers, and e ab The proportion of standardized evaluation index b obtained by standardized evaluation index a under the selected time scale is the standard index i.

[0138] Specifically, the criterion layer weight vector matrix Its expanded form can be expressed as:

[0139]

[0140] Obviously, the row sum of the matrix is ​​a constant, i.e., the selected time scale; the fuzzy relation synthesis operator is selected, and the evaluation index values ​​of the factor layer are used as the basis for the evaluation of the upper criterion layer, and the item-by-item synthesis operation of each evaluation index is performed; depending on the training time, the evaluation will be completed at different time scales; among them, the fuzzy operators that can be selected according to different needs are shown in Table 5, and the weighted average type operator is used here.

[0141] Table 5 Fuzzy Operators

[0142] serial number Fuzzy operator type illustrate 1 Main factor prominence type: M(Λ,V) More information is used from the R matrix, while less is used from the E matrix. 2 Main factor prominence type: M(.,V) More information is used from the R matrix, while less is used from the E matrix. 3 Weighted average type: M(Λ,+) More information on using the E matrix 4 Weighted average type: M(.,+) Comprehensive use of information from R matrix and E matrix

[0143] S232: Criterion Layer Evaluation Vector S i B The calculation formula is:

[0144]

[0145] in, s1 to The membership degree of criterion index i, that is, from s1 to s2. This represents the proportion of different comments in the comment set V for the criterion indicator.

[0146] In this embodiment, step S3 is similar to the criterion layer. Obtaining the target layer evaluation requires constructing the target layer fuzzy judgment matrix and the target layer weight vector matrix. The construction of the target layer fuzzy judgment matrix and the weight calculation are exactly the same as described above. The target layer weight vector matrix cannot be directly measured, so it is composed of the criterion layer evaluation vector of the previous layer.

[0147] Step S3 is as follows:

[0148] S31: The target layer is scored by the criteria indicators in the criteria layer, and the final rehabilitation evaluation result is standardized by the scoring to obtain the standardized final rehabilitation evaluation result. The standardized target layer is composed of the standardized final rehabilitation evaluation result.

[0149] S32: Construct a standardized target layer fuzzy judgment matrix for the target layer, and calculate the target layer evaluation index weight matrix through the target layer fuzzy judgment matrix;

[0150] S33: Obtain the target layer weight vector matrix by calculating the criterion layer evaluation vector, and obtain the target layer evaluation vector by calculating the target layer evaluation index weight matrix and the target layer weight vector matrix.

[0151] In this embodiment, step S32 specifically includes:

[0152] S321: The target layer fuzzy judgment matrix is ​​represented as: R A =(r′) df ) c×c ;

[0153] Where A represents the evaluation layer, d and f are the index numbers of the criteria indicators, and r' ab Let c represent the relative importance of criterion indicator a to criterion indicator b, c represent the total number of criterion indicators, and d and f both fall within the range of c.

[0154] S322: R AConvert to matrix K = (k′) df ) c×c ,in Perform a consistency check on matrix K. If the consistency check passes, proceed to step S325; otherwise, proceed to step S323.

[0155] S323: Calculate the eigenvectors P′=[p′1,p′2,…,p′] of matrix K. c ] T Constructing a matrix Where β∈(0,1);

[0156] S324: Calculate the fuzzy judgment matrix of the criterion layer that satisfies consistency. The calculation formula is:

[0157] in use Replace R A Return to step S322;

[0158] S325: Weight Matrix of Evaluation Indicators at the Criterion Level W A The calculation formula is:

[0159] W A =[w′1,w′2,…w′ a …, w′ c ]

[0160]

[0161]

[0162] In this embodiment, step S33 specifically includes:

[0163] S331: Target layer weight vector matrix E A The calculation formula is:

[0164]

[0165] Where U is the criterion-level evaluation vector S i B The maximum value of the index number i in the standard;

[0166] S332: Target layer evaluation vector S A The calculation formula is:

[0167] S A =W A E A

[0168] Among them, W AThis is the weight matrix of the evaluation indicators for the target layer.

[0169] In this embodiment, step S4 specifically includes:

[0170] A quantitative score vector G is obtained based on rehabilitation assessment needs;

[0171] Specifically, for example, you can choose G = [5,4,3,2,1];

[0172] The formula for calculating the final rehabilitation evaluation result Y is:

[0173] Y = S A G T

[0174] Among them, S A This is the evaluation vector for the target layer.

[0175] This invention provides a multi-timescale rehabilitation evaluation system based on fuzzy hierarchical analysis, comprising:

[0176] The structural construction module is used to build a hierarchical structure for rehabilitation evaluation, which includes: information layer, factor layer, criterion layer and goal layer.

[0177] The first correlation module is used to construct the correlation between the factor layer and the criterion layer to obtain the criterion layer evaluation vector;

[0178] The second association module is used to construct the association between the criterion layer and the target layer through the criterion layer evaluation vector, and obtain the target layer evaluation vector.

[0179] The evaluation module is used to obtain the final rehabilitation evaluation result through the target layer evaluation vector.

[0180] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0181] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In the unit claims listing several devices, several of these devices may be embodied by the same hardware item. The use of the terms first, second, and third, etc., does not indicate any order and can be interpreted as identifiers.

[0182] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis, characterized in that, include: S1: Construct a hierarchical structure for rehabilitation assessment, which includes: information layer, factor layer, criterion layer, and goal layer; Step S1 is as follows: Acquire multi-source individual information in lower limb rehabilitation scenarios, and construct an information layer based on this multi-source individual information; Set evaluation indicators and construct a factor layer based on these indicators; Set criteria indicators and construct a criteria layer based on these indicators; The hierarchical structure of rehabilitation assessment, from bottom to top, is as follows: information layer - factor layer - criterion layer - goal layer; S2: Construct the relationship between the factor layer and the criterion layer to obtain the criterion layer evaluation vector; Step S2 is as follows: S21: Standardize the evaluation indicators to obtain standardized evaluation indicators. The standardized evaluation indicators constitute a standardized factor layer. The standardized evaluation indicators are scored by individual multi-source information in the information layer. S22: Construct a fuzzy judgment matrix for the criteria indicators of the criteria layer, and calculate the weight matrix of the evaluation indicators of the criteria layer through the fuzzy judgment matrix of the criteria layer. S23: Construct the criterion layer weight vector matrix, and calculate the criterion layer evaluation vector through the criterion layer evaluation index weight matrix and the criterion layer weight vector matrix; Step S23 is as follows: S231: The weight vector matrix of the criterion layer is represented as follows: Where B represents the criterion layer, i is the criterion indicator number, a and b are both standardized evaluation indicator numbers, and e ab The proportion of standardized evaluation index b obtained by standardized evaluation index a under the selected time scale is the standard index i. S232: Criterion Layer Evaluation Vector S i B The calculation formula is: in, s1 to The membership degree of criterion index i; S3: Construct the relationship between the criterion layer and the target layer through the criterion layer evaluation vector to obtain the target layer evaluation vector; S4: Obtain the final rehabilitation evaluation result through the target layer evaluation vector.

2. The multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis according to claim 1, characterized in that, Step S22 is as follows: S221: The fuzzy judgment matrix of the criterion layer is represented as follows: Where B represents the criterion layer, i is the criterion indicator number, a and b are both standardized evaluation indicator numbers, and r ab To determine the relative importance of standardized evaluation indicator a to standardized evaluation indicator b, n i Let a and b be the total number of standardized evaluation indicators for criterion indicator i, where a and b are both in the range of n. i Within the range; S222: Will Convert to matrix in For matrix K i Perform a consistency check. If the consistency check passes, proceed to step S225; otherwise, proceed to step S223. S223: Calculate and obtain matrix K i eigenvector P i =[p1,p2,…,p ni ] T Constructing a matrix Where α∈(0,1); S224: Calculate the fuzzy judgment matrix of the criterion layer that satisfies consistency. The calculation formula is: in use replace Return to step S222; S225: Weight Matrix of Evaluation Indicators at the Criterion Level W i B The calculation formula is: Where i is from 1 to n i Positive integers.

3. The multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis according to claim 1, characterized in that, Step S3 is as follows: S31: The target layer is scored by the criteria indicators in the criteria layer, and the final rehabilitation evaluation result is standardized by the scoring to obtain the standardized final rehabilitation evaluation result. The standardized target layer is composed of the standardized final rehabilitation evaluation result. S32: Construct a standardized target layer fuzzy judgment matrix for the target layer, and calculate the target layer evaluation index weight matrix through the target layer fuzzy judgment matrix; S33: Obtain the target layer weight vector matrix by calculating the criterion layer evaluation vector, and obtain the target layer evaluation vector by calculating the target layer evaluation index weight matrix and the target layer weight vector matrix.

4. The multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis according to claim 3, characterized in that, Step S32 is as follows: S321: The target layer fuzzy judgment matrix is ​​represented as: R A =(r′) df ) c×c ; Where A represents the evaluation layer, d and f are the index numbers of the criteria indicators, and r' ab Let c represent the relative importance of criterion indicator a to criterion indicator b, c represent the total number of criterion indicators, and d and f both fall within the range of c. S322: R A Convert to matrix K = (k′) df ) c×c ,in Perform a consistency check on matrix K. If the consistency check passes, proceed to step S325; otherwise, proceed to step S323. S323: Calculate the eigenvectors P′=[p′1,p′2,…,p′] of matrix K. c ] T Constructing a matrix Where β∈(0,1); S324: Calculate the fuzzy judgment matrix of the criterion layer that satisfies consistency. The calculation formula is: in use Replace R A Return to step S322; S325: Weight Matrix of Evaluation Indicators at the Criterion Level W A The calculation formula is: IN A =[w′1,w′2,…w′ a …,In' c ] 5. The multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis according to claim 3, characterized in that, Step S33 is as follows: S331: Target layer weight vector matrix E A The calculation formula is: Where U is the criterion-level evaluation vector S i B The maximum value of the index number i in the standard; S332: Target layer evaluation vector S A The calculation formula is: S A =W A E A Among them, W A This is the weight matrix of the evaluation indicators for the target layer.

6. The multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis according to claim 1, characterized in that, Step S4 is as follows: A quantitative score vector G is obtained based on rehabilitation assessment needs; The formula for calculating the final rehabilitation evaluation result Y is: Y=S A G T Among them, S A This is the evaluation vector for the target layer.

7. A multi-timescale rehabilitation evaluation system based on fuzzy hierarchical analysis, characterized in that, The system applies the multi-timescale rehabilitation evaluation method based on fuzzy hierarchical analysis as described in claim 1, including: The structural construction module is used to build a hierarchical structure for rehabilitation evaluation, which includes: information layer, factor layer, criterion layer and goal layer. The first correlation module is used to construct the correlation between the factor layer and the criterion layer to obtain the criterion layer evaluation vector; The second association module is used to construct the association between the criterion layer and the target layer through the criterion layer evaluation vector, and obtain the target layer evaluation vector. The evaluation module is used to obtain the final rehabilitation evaluation result through the target layer evaluation vector.

Images