Sleep disorder diagnosis method based on multi-attribute group decision

Abstract

The invention provides a sleep disorder diagnosis method based on multi-attribute group decision, which comprises the following steps: acquiring evaluation information of a plurality of decision makers for a plurality of alternative schemes under different attributes, constructing a decision matrix, determining the weight of each attribute through an entropy weight method, and determining the weight of each attribute; and aggregating the evaluation information of each alternative scheme under all attributes by using a aggregation operator to obtain a comprehensive evaluation value of each alternative scheme corresponding to each decision maker, calculating a score of the comprehensive evaluation value, weighting the score based on the weight of the decision maker to obtain a final score, and sorting the final score. According to the method, a distance measurement method which does not need standardized constraint and retains original data distribution is further provided for a fuzzy set on the basis of a dual hesitant q-run sequence, and an entropy weight mechanism of weight is objectively obtained through information entropy analysis, so that measurement integrity and decision objectivity in sleep disorder diagnosis are ensured.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of medical data analysis, and relates to a sleep disorder diagnosis method based on multi-attribute group decision. BACKGROUND

[0002] Sleep disorders refer to abnormal sleep quantity and abnormal behavior during sleep, and are also a manifestation of the normal rhythm alternation between sleep and wakefulness. The main causes of insomnia are biological factors, such as changes in biological rhythm, physical illness, genetic factors, and body temperature changes. Secondly, psychological reasons, such as sleep expectations, bad sleep habits, and incorrect cognition. Chronic sleep deprivation is associated with an increased risk of various health complications, including cardiovascular disease, metabolic dysfunction, and mental illness. Sleep deprivation has become one of the most common and most pressing medical problems in modern society.

[0003] With the increasing incidence of sleep disorders in the general population, traditional Chinese medicine has played an important role in this regard. Traditional Chinese medicine is one of the best methods for diagnosing sleep disorders. There are many methods for diagnosing sleep disorders in traditional Chinese medicine, and different methods can be chosen for treatment according to the specific circumstances of the patient and the different causes of sleep disorders.

[0004] Traditional Chinese medicine uses a variety of diagnostic methods for sleep disorders and selects different treatment methods based on the specific circumstances of the patient and the underlying causes of sleep disorders. One of the important features of traditional Chinese medicine is individualization. Syndrome differentiation and treatment provide individualized diagnosis and treatment for each patient, and the overall view and emphasis on individualized treatment give traditional Chinese medicine a unique advantage and effectiveness in treating sleep disorders. When developing a treatment plan, traditional Chinese medicine will consider the following aspects:

[0005] 1. Traditional Chinese medicine believes that each disease has its own unique etiology, pathogenesis, and development process, so it is important to evaluate the applicability of treatment methods, including determining whether the selected treatment method is consistent with the pathogenesis of the disease and whether it can effectively alleviate or eliminate symptoms;

[0006] 2. Efficacy is the primary standard for evaluating the effectiveness of traditional Chinese medicine treatment, which mainly includes: short-term efficacy, which refers to the improvement of symptoms in the short term after treatment; long-term efficacy, which refers to the stability of the disease over a longer period of time (such as months or years) after treatment;

[0007] 3. Due to differences in patient constitution and disease, adverse reactions or side effects may occur during treatment, so potential adverse reactions, side effects, and their incidence rates need to be monitored throughout the process of traditional Chinese medicine treatment;

[0008] 4. When evaluating the economic benefits and overall value of different treatment plans, traditional Chinese medicine treatment must consider factors such as treatment cost, efficacy, and patient satisfaction.

[0009] The Chinese medicine diagnosis needs to consider the above-mentioned various factors, which jointly constitute a comprehensive index system for evaluating a treatment scheme, and is helpful for scientifically and comprehensively evaluating the advantages and characteristics of various treatment methods. Through the present application, a more personalized overall treatment scheme can be formulated for a patient. Since the specific conditions and potential causes of syndromes of patients are different from each other, it can be challenging to select a suitable Chinese medicine treatment scheme. In addition, the determination of a Chinese medicine diagnosis scheme is also affected by subjectivity and uncertainty. Therefore, a sleep disorder diagnosis method based on multi-attribute group decision-making is crucial to solve the above-mentioned complex problems.

[0010] If the Chinese medicine diagnosis of sleep disorders is considered as a multi-attribute group decision-making, the multi-attribute group decision-making has considerable fuzziness and hesitation, and the correlation and flexibility of attributes also need to be considered. At present, the dual hesitant q-rung sequence pair fuzzy set can provide a solution to the fuzziness and hesitation existing in the above-mentioned multi-attribute group decision-making, but it has the technical defect of information redundancy caused by the need for normalized distance measurement. Therefore, the present application proposes a sleep disorder diagnosis method based on a dual hesitant q-rung sequence pair fuzzy set, which uses a distance measurement method without normalization and an entropy measurement method to derive the uncertain attribute weights, and obtains comprehensive evaluation information through an aggregation operator to solve the above-mentioned technical problems. SUMMARY

[0011] The present application aims to solve the problems existing in the prior art, and provides a sleep disorder diagnosis method based on multi-attribute group decision-making, which solves the problems of the prior art that: ① the correlation and flexibility of attributes cannot be considered simultaneously, and the score value change rate is low; ② information distortion is caused due to the dependence on data standardization; and ③ decision bias is caused due to subjective attribute weighting.

[0012] To this end, the present application adopts the following technical solutions:

[0013] A sleep disorder diagnosis method based on multi-attribute group decision-making, comprising the following steps:

[0014] Obtaining evaluation information of multiple decision-makers for multiple alternative schemes under different attributes, and constructing multiple dual hesitant q-rung sequence pair fuzzy decision matrices;

[0015] Preprocessing the decision matrix to unify the attribute types;

[0016] Determining the attribute weights by an entropy weight method, including: calculating the entropy value of each attribute evaluation information, and calculating the attribute weights based on the entropy value;

[0017] aggregating the evaluation information of each alternative under all attributes by using an aggregation operator to obtain a comprehensive evaluation value of each alternative corresponding to each decision maker, wherein the aggregation operator comprises a DHq-ROFWDHM operator and a DHq-ROFWDDHM operator;

[0018] calculating a score value of each comprehensive evaluation value;

[0019] weighting the score values of the same alternative given by each decision maker based on decision maker weights to obtain a final score value of each alternative;

[0020] ranking all alternatives according to the final score values and selecting an optimal alternative.

[0021] Specifically, the application comprises the following steps:

[0022] Step 1, collecting evaluation values given by decision makers according to attributes and alternatives to construct a decision matrix , and the expression is:

[0023]

[0024] wherein, ;

[0025] In the formula, is an evaluation matrix of the i th decision maker; is a membership degree of the evaluation value of the i th decision maker about the attribute under the alternative in the evaluation matrix of the i th decision maker; is a non-membership degree of the evaluation value of the i th decision maker about the attribute under the alternative in the evaluation matrix of the i th decision maker.

[0026] Specifically:

[0027] Chinese medicine is used as a decision maker ;

[0028] evaluation elements are used as attributes , and the evaluation elements at least comprise disease adaptability, curative effect, safety and cost-effectiveness;

[0029] diagnosis of sleep disorder schemes are used as alternatives ;

[0030] evaluation information is used for diagnosis of sleep disorders in Chinese medicine, and is used as a decision matrix ;

[0031] ​​​​​​​Set decision-maker weights And satisfy , ;

[0032] Set attribute importance to And satisfy , .

[0033] Step 2: Divide the attributes into cost attributes and benefit attributes, and normalize the decision matrix to obtain a normalized decision matrix;

[0034] The normalized decision matrix expression is as follows:

[0035]

[0036] In the formula, To standardize the decision matrix; For the first The decision-maker in the attribute Below are the alternatives The normalized evaluation value; for The complement, i.e. .

[0037] The attribute evaluation value is represented by the dual hesitant q-rung order pair fuzzy number. The classification criterion is: if the higher the membership degree, the more conducive it is to achieving the decision goal, then it is a benefit-type attribute; if the lower the membership degree, the more conducive it is to achieving the decision goal, then it is a cost-type attribute.

[0038] Step 3: Determine attribute weights using the entropy method;

[0039] The attribute weighting formula is as follows:

[0040]

[0041] in:

[0042]

[0043]

[0044] In the formula, for The weights; for entropy; The number of attributes; For the first Each decision-maker in alternative options The sum of the entropy values ​​of all attribute evaluation values ​​under; For dual hesitant q-order pairs of fuzzy numbers The distance between it and its complement; For the first The decision-maker in the attribute Below are the alternative plans The evaluation value; for The complement; for The number of hesitant elements with high membership degree; for The number of hesitant elements with high membership degree; The number of hesitant elements in the membership degree between China and Africa; for The number of hesitant elements in the membership degree between China and Africa; for Membership degree in; The parameters of the dual hesitant q-order pair of fuzzy sets; for Membership degree in; for Non-membership degree in; for The degree of non-membership in.

[0045] Step 4: Use the first aggregation operator or the second aggregation operator to aggregate the evaluation values ​​of each alternative in the decision matrix under all attributes to obtain the comprehensive evaluation value of each alternative by all decision-makers, and thus obtain the comprehensive decision matrix.

[0046] Specifically, the first aggregation operator is the DHq-ROFWDHM operator; the second aggregation operator is the DHq-ROFWDDHM operator; where:

[0047] The first aggregation operator focuses more on the aggregation of membership degree and is relatively conservative in its treatment of non-membership degree. It is suitable for situations where decision-makers value positive factors (such as treatment effect and safety) and there is a positive correlation or synergistic effect between attributes, and can be used to highlight the influence of high membership degree values.

[0048] The second aggregation operator focuses more on aggregation of non-membership degrees and is relatively conservative in its treatment of membership degrees. It is suitable for decision-makers who are more concerned about negative factors (such as side effects and costs) and for situations where there are negative correlations or offsetting effects between attributes, and can be used to highlight the impact of high non-membership degree values.

[0049] When the first aggregation operator is used, the expression for the comprehensive decision matrix is ​​as follows:

[0050]

[0051] in:

[0052]

[0053] In the formula, An additive aggregation operator for dual hesitant q-rung order pairs of fuzzy sets, used to fuse membership and non-membership information of multiple fuzzy numbers; It is a set of dual hesitant q-rung order pair fuzzy numbers that need to be aggregated; n is the total number of attributes; This is an aggregation parameter that indicates how many attributes to consider for each interaction. Traversal All Combinations of tuples; Indicates from Select from the attributes The number of all combinations of ; It is an alternative plan The following attributes The weights; It is an alternative plan The following attributes The evaluation value; A multiplicative aggregation operator for dual hesitant q-rung order pairs of fuzzy sets, used to integrate membership and non-membership information of multiple fuzzy numbers.

[0054] When the second aggregation operator is used, the expression for the comprehensive decision matrix is ​​as follows:

[0055]

[0056] in:

[0057] .

[0058] Step 5: Calculate the score for each evaluation value in the comprehensive decision matrix;

[0059] The formula for calculating the score of the evaluation value is as follows:

[0060]

[0061] In the formula, This represents the number of membership degrees. For the first Each decision-maker considers alternative options The following attributes The membership degree raised to the power of q; The parameters of the dual hesitant q-rung order pair fuzzy set; This represents the number of non-membership degrees; For the first Each decision-maker considers alternative options The following attributes The non-membership degree raised to the power of q.

[0062] Step 6: Calculate the final score for each alternative based on the decision-maker's weight;

[0063] The formula for the final score is as follows:

[0064]

[0065] In the formula, As an alternative plan The overall score; Let p be the weight of the decision-maker. To aggregate the p-th decision-maker's opinions on alternatives Evaluation values ​​across all attributes.

[0066] Step 7: Consider the alternative solutions. Sort the options and determine the final score based on the alternatives. Rank all alternatives and select the best one; the higher the score, the better the solution.

[0067] The beneficial effects of this invention are as follows:

[0068] This invention, based on dual-hesitant q-rung order pairs of fuzzy sets, further proposes a distance metric method that does not require standardization constraints and preserves the original data distribution. It also employs an entropy weighting mechanism, through information entropy analysis, to objectively derive weights, thereby ensuring the integrity of the metric and the objectivity of decision-making in the diagnosis of sleep disorders. Specifically:

[0069] 1. This invention captures the inherent ambiguity and hesitation of data, making it suitable for the diagnosis of sleep disorders in Traditional Chinese Medicine (TCM). Its dual representation capability can simultaneously represent the ambiguity of patients' symptoms and the diagnostic hesitation of decision-makers, providing the most suitable data model for the diagnosis of sleep disorders in TCM. Furthermore, it can combine ambiguous symptoms with clinical expertise to generate personalized treatment plans, thereby effectively linking the theoretical framework with actual diagnostic requirements.

[0070] 2. This invention utilizes the interrelationships between aggregation operators to fuse parameters, providing flexibility and increasing the rate of change of score values, thereby offering stronger distinguishing ability for ranking alternative solutions.

[0071] 3. This invention utilizes a distance metric method that does not require standardization constraints and preserves the original data distribution. It uses an entropy weighting mechanism to objectively derive weights through information entropy analysis, thereby ensuring the integrity of the measurement and the objectivity of the decision in the diagnosis of sleep disorders. Detailed Implementation

[0072] The technical solution of the present invention will be described below in conjunction with the implementation method.

[0073] Example

[0074] Traditional Chinese medicine (TCM) employs various diagnostic methods for sleep disorders and selects different treatment methods based on the patient's specific condition and the root cause of the sleep disorder. TCM provides individualized diagnosis and treatment tailored to each patient through syndrome differentiation and treatment. Its holistic view and emphasis on personalized treatment give TCM advantages and effectiveness in treating sleep disorders.

[0075] In this example, the alternative treatment plan for sleep disorders diagnosed by traditional Chinese medicine is: traditional Chinese medicine therapy. Acupuncture therapy Massage therapy Dietary therapy Cupping therapy Assessment attributes include disease fit. Therapeutic effects Safety Cost-effectiveness Three traditional Chinese medicine practitioners assessed one patient, with each assessment having a weighted average. ;set up Dombi parameters HM parameters .

[0076] This embodiment specifically includes the following steps:

[0077] Step 1: Based on the patient's specific condition, the three TCM doctors constructed three decision matrices using attributes and alternative treatment options, as shown in Table 1-3.

[0078] Table 1. Decision Matrix of the First Traditional Chinese Medicine Practitioner

[0079]

[0080] Table 2. Decision Matrix of the Second Traditional Chinese Medicine Practitioner

[0081]

[0082] Table 3. Decision Matrix of the Third-Rank Traditional Chinese Medicine Practitioner

[0083]

[0084] Step 2: Since all five attributes are profit-oriented, there is no need to normalize the three decision matrices mentioned above.

[0085] Step 3: Calculate all alternative solutions Entropy weight of each attribute As shown in Table 4-6.

[0086] Table 4. Attribute Weights of the First Traditional Chinese Medicine Practitioner

[0087]

[0088] Table 5. Attribute Weights of the Second-Ranked Traditional Chinese Medicine Practitioner

[0089]

[0090] Table 6. Attribute Weights of the Third-Ranked Traditional Chinese Medicine Practitioner

[0091]

[0092] Step 4: Aggregate each alternative solution using the first aggregation operator. The evaluation value of the corresponding attribute in the decision matrix .

[0093] Step 5: For each alternative in the decision matrix, calculate the score. As shown in Table 7.

[0094] Table 7 Evaluation Information Score for Each Alternative Solution

[0095]

[0096] Step 6: Use decision-maker weights Calculate the final score; where: , , , , .

[0097] Step 7: Based on the final score, the priority of the alternative solutions is as follows: Therefore, the best treatment option for this patient with sleep disorder is .

[0098] If the second aggregation operator is used for calculation in step 4, the scores of the alternative solutions are as follows: 0.1237 0.0872 0.6351 0.1558 0.2984, the final sorting result is .

[0099] Therefore, this further validates that the optimal treatment plan for this sleep disorder patient is... That is, using either the first aggregation operator or the second aggregation operator has no impact on the decision result, which reflects the stability of the two operators provided in the embodiment.

[0100] The following comparison of this embodiment with other decision-making methods will verify the technical effectiveness of this embodiment.

[0101] The method proposed by Gayen et al. in 2023, "A method for multi-criteria group decision-making using DHq-ROFAAWA and DHq-ROFAAWG operators," is used as comparison method 1; the method proposed by Gayen et al. in 2023, "A method for multi-criteria group decision-making using DHq-ROFSSWA and DHq-ROFSSWG operators," is used as comparison method 2. Since the attribute weights in both methods are subjectively assigned, the attribute weight vector is assumed to be... .

[0102] Table 7 shows the diagnostic results obtained by three different diagnostic methods for patients with sleep disorders.

[0103] Table 7 Evaluation Information Score for Each Alternative Solution

[0104] As shown in Table 7, the optimal result of this embodiment ( Consistent with existing methods, although other rankings differ slightly, the best diagnostic results are consistent, thus confirming the reliability of the invention.

[0105] This embodiment is compared with other decision-making methods to verify that the operator in this embodiment has a strong discriminative ability.

[0106] This embodiment uses the DHq-ROFWDHM and DHq-ROFWDDHM operators. The calculation results of these two operators are compared with those of the DHq-ROFAAWA and DHq-ROFAAWG operators. Since the operators are used to aggregate fuzzy numbers and obtain scores, the greater the difference in scores, the stronger the discriminative power of the operator. Therefore, the comparison process is as follows:

[0107] First, calculate the overall change in score (OVSV):

[0108]

[0109] In the formula, For the change of the i-th score,

[0110] Secondly, calculate the percentage increase in the overall rate of change of the score ( ):

[0111]

[0112] In the formula, This represents the overall change in the score of this invention; This is to compare the overall changes in scores using different methods.

[0113] Set the attribute weight vector to As shown in Tables 8 and 9, Table 8 lists the ranking and scores of the alternative solutions obtained by comparing Method 1 with those of this embodiment, and Table 9 lists the overall rate of change of the scores.

[0114] Table 8. Scores obtained by the Gayen method and this invention

[0115] Table 9. Rate of change of scores for different methods

[0116]

[0117] As shown in Table 8, although the rankings of the two methods are slightly different, the selection of the best treatment option is consistent. As shown in Table 9, the DHq-ROFWDHM operator is 29% better than the DHq-ROFAAWA operator, and the DHq-ROFWDDHM operator is 28% better than the DHq-ROFAAWG operator. Therefore, this method is more effective in selecting the best alternative.

[0118] We will continue to compare this embodiment with other decision-making methods to verify the advantages of the distance formula proposed in this embodiment.

[0119] Distance formulas are a suitable metric for measuring the closeness of two fuzzy numbers; however, existing distance metrics for DHq-ROFNs have certain limitations. For example, some existing distance formulas require the number of memberships and non-memberships to be equal. However, in many real-world scenarios, the number of memberships and non-memberships are not equal. Therefore, a supplementary strategy must be adopted to ensure that the number of memberships and the number of N memberships are the same in the compared DHq-ROFNs. Unfortunately, this adjustment may lead to information loss.

[0120] Given several DHq-ROFNs: , , , .

[0121] The parameters proposed in this embodiment The distance formula, and The distance formula proposed in the "A Novel q-Order Double Hesitant Fuzzy Multi-Attribute Decision-Making Method Based on Entropy Weights" published in 2015 serves as a comparison method 3, and The q-order orthogonal hesitant fuzzy Euclidean distance formula introduced in the example is used as comparison method 4 to calculate the distance between fuzzy numbers in this embodiment. The comparison calculation results are shown in Table 10.

[0122] Table 10 Distance between fuzzy numbers calculated using different methods

[0123]

[0124] As shown in Table 10, the results calculated by comparison method 4 are... and The distance is 0, but, and They are clearly not equal; both methods 3 and 4 use different normalization methods to ensure that the lengths of DHq-ROFN are the same, which introduces information loss; in contrast, the distance formula proposed in this embodiment does not require normalization, thus avoiding any information loss.

[0125] To further highlight the advantages of this embodiment, Table 11 presents a comparison of the features of several existing decision-making methods with those of the present invention. Among them, the "Hesitant Fermatean fuzzy Bonferroni averaging operator for multi-attribute decision-making" proposed by Wang et al. in 2024 is used as comparison method 5; the "double hesitant q-cascaded orthogonal pairing fuzzy Hammarskjöld aggregation operator and its application in project scheme selection" proposed by Wang et al. in 2019 is used as comparison method 6; and the "hesitant q-rung orthogonal fuzzy clustering operator and its application in multi-criteria decision-making" proposed by Hussain in 2020 is used as comparison method 7.

[0126] Table 11 Comparison of the characteristics of different decision-making methods

[0127]

[0128] As shown in Table 11, this embodiment has the following advantages:

[0129] First, this embodiment utilizes dual hesitant q-rung orthogonal fuzzy sets (DHq-ROFS), which can represent membership and non-membership degrees with hesitant values. By adjusting the q value, the representation range of the fuzzy information space is expanded. In contrast, comparison method 5 uses hesitant Fermat fuzzy sets as its basic framework. When q is set to 3, this fuzzy set is a special case of DHq-ROFS. Hesitant Fermat fuzzy sets cannot effectively solve certain situations, while this embodiment can adapt to a wider fuzzy information space by adjusting the q value. Therefore, compared with comparison method 5, this embodiment has stronger effectiveness.

[0130] Secondly, this embodiment is flexible and can consider the interrelationships between various attributes. This embodiment is based on the Dombi norm, where the parameters are related to the decision-maker's risk preference. The decision-maker can adaptively select appropriate Dombi parameters according to their personal risk tolerance. In addition, in the TCM diagnosis of sleep disorders, different attributes are interrelated, and the HM operator can effectively consider these interrelationships. In contrast, comparison methods 1, 2, 5, 6, and 7 all lack the above features. Therefore, compared with the above methods, this embodiment is more reliable and more flexible.

[0131] Third, this embodiment minimizes information loss by introducing a distance metric that does not require normalization. Based on this, an entropy weighting method is proposed to objectively calculate attribute weights within the DHq-ROFS range, thereby improving the reliability of the diagnosis results for patients with sleep disorders. However, compared with methods 1, 2, 5, 6, and 7, all of which rely on decision-maker evaluation, the credibility of the diagnosis results is weakened because decision-makers will subjectively allocate weights.

[0132] Finally, since group decision-making is inherently complex, the participation of multiple decision-makers can provide different perspectives and experiences, thereby improving the scientific nature of the decision. In contrast, methods 5, 6, and 7 all focus on the decision-making of a single decision-maker. This embodiment can accommodate multiple decision-makers, effectively overcoming the limitations of individual decision-making, and is proven to be more suitable for complex real-world decision-making.

Claims

1. A sleep disorder diagnosis method based on multi-attribute group decision, characterized in that, The method comprises the following steps: obtaining evaluation information of multiple decision makers on multiple alternatives under different attributes, and constructing a decision matrix; preprocessing the decision matrix to unify attribute types; determining attribute weights by an entropy weight method; aggregating evaluation information of each alternative under all attributes by using a first aggregation operator or a second aggregation operator to obtain a comprehensive evaluation value of each alternative corresponding to each decision maker; calculating score values of each comprehensive evaluation value; weighting the score values based on decision maker weights to obtain final score values of each alternative; ranking all alternatives according to the final score values, and the higher the score value, the better the alternative. 2.The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, characterized in that, The entropy weight method comprises calculating an entropy value of each attribute evaluation information, and calculating attribute weights based on the entropy value. 3.The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, characterized in that, The first aggregation operator refers to a DHq-ROFWDHM operator, and the second aggregation operator refers to a DHq-ROFWDDHM operator. 4.The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, characterized in that, The decision matrix The expression is as follows: ; wherein is the evaluation matrix of the decision maker in the th position; is the evaluation matrix of the decision maker in the th position; is the membership of the evaluation value of the alternative under the attribute is the non-membership of the evaluation value of the alternative under the attribute for the decision maker in the th position.

5. The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, characterized in that, The normalized decision matrix expression is as follows: ; wherein is the first decision maker's decision on the normalized evaluation values of the alternatives; is the complement of .​​​ 6. The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, characterized in that, The attribute weight formula is as follows: ； wherein: ； ； where, is the weight of the attribute; is the entropy of the attribute; is the number of attributes; is the sum of the entropy values of all attribute evaluation values of the alternative under the i-th decision maker; is the distance between the alternative and its complement; is the evaluation value of the alternative by the i-th decision maker under the attribute is the number of membership hesitant elements in is the number of non-membership hesitant elements in is the membership in is the parameter of the dual hesitant q-rung orthopair fuzzy set; is the membership in is the non-membership in is the non-membership in​​​​​​​​​​​​​​​​​ 7. The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, characterized in that, When the first aggregation operator is used, the comprehensive decision matrix expression is as follows: ； wherein: ； wherein is the comprehensive decision matrix; is the additive aggregation operator of dual hesitant q-rung pair-wise fuzzy sets, which is used to fuse the membership and non-membership information of multiple fuzzy numbers; is the aggregation parameter, which represents how many attributes are considered for interaction at a time, iterates all tuple combinations; represents the number of all combinations of selecting from attributes; is the weight of attribute under alternative ; is the evaluation value of attribute under alternative ; is the multiplicative aggregation operator of dual hesitant q-rung pair-wise fuzzy sets, which is used to fuse the membership and non-membership information of multiple fuzzy numbers; When the second aggregation operator is used, the comprehensive decision matrix expression is as follows: ； wherein: 。 8.The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, wherein, The evaluation value score formula is as follows: ； In the formula, The number of membership degrees; For the first Each decision-maker considers alternative options The following attributes The membership degree raised to the power of q; The parameters of the dual hesitant q-rung order pair fuzzy set; This represents the number of non-membership degrees; For the first Each decision-maker considers alternative options The following attributes The non-membership degree raised to the power of q. 9.The sleep disorder diagnosis method based on multi-attribute group decision according to claim 1, wherein, The final score formula is as follows: ； wherein the overall score for the alternative is the pth decision maker weight is the aggregate pth decision maker evaluation value for the alternative under all attributes.​

Images