Sarcopenia rehabilitation system based on artificial intelligence large model

Abstract

The application relates to a sarcopenia rehabilitation system based on an artificial intelligence large model. According to initial index information of patients, clustering operation is carried out, and generation of a rehabilitation scheme is carried out in units of patient clustering clusters, so that the generation efficiency of an initial rehabilitation scheme is improved; in the rehabilitation scheme execution process, clustering operation is carried out in the patient clustering cluster in units of the patient clustering cluster and all patients in the patient clustering cluster according to respective index recovery conditions of all patients in the patient clustering cluster, representative index recovery deviation information corresponding to all patient clustering subclusters in the patient clustering cluster is obtained, and it is judged whether the rehabilitation scheme corresponding to all patients in the patient clustering subcluster needs to be adjusted, so that the adjustment efficiency of a subsequent rehabilitation scheme is improved. And a new patient clustering cluster is generated in a layer-by-layer progressive sorting mode, the uniformity of all patient information is ensured when all patients in the clustering subcluster are uniformly adjusted in the rehabilitation scheme each time, and the feasibility of adjusting the rehabilitation scheme in units of clusters is ensured.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the field of data processing technology, and specifically to a sarcopenia rehabilitation system based on an artificial intelligence large model. Background Art

[0002] For patients with sarcopenia, a rehabilitation program is often developed to ensure they recover as expected. This is not limited to sarcopenia; other conditions also require a specific rehabilitation program. Numerous studies have explored the development of rehabilitation programs.

[0003] For example, the Chinese invention patent application number CN202411241461.8 discloses a generative AI-assisted method and system for lower back pain rehabilitation. The implementation method includes the following steps: S1, real-time collection of patient health data through a data acquisition module and processing of the health data; S2, the generative AI-assisted system converts the conversation between the patient and the doctor into a structured text record through an automatic speech recognition module and a natural language processing module, and integrates it into the EHR system; S3, the patient's health information is analyzed using a generative AI model to generate a simplified medical language version; S4, based on the health data, a personalized rehabilitation plan is generated; S5, the rehabilitation plan is pushed to the patient through the user interface, and the generative AI-assisted system dynamically adjusts the rehabilitation plan in real time based on patient feedback; S6, the generative AI model is continuously optimized by continuously collecting patient feedback data. This invention can automatically generate personalized rehabilitation plans.

[0004] Another example is the Chinese invention patent with application number CN202411805672.X, which discloses a method and system for intelligently generating a home pelvic floor muscle rehabilitation training program. During the rehabilitation training process, the selection operation of the pelvic floor muscle dynamics model and the local program generation model is introduced. On the one hand, the pelvic floor muscle dynamics model is used to simulate the elastic response to obtain accurate and reliable local stage evaluation data of the target training stage. On the other hand, by selecting a local program generation model that matches the actual situation of the target training stage, a more matching targeted training program can be obtained based on the accurate and reliable local stage evaluation data and the stage test data of the target training stage through the matching local program generation model, providing accurate and high-quality data dependence for adjusting the global rehabilitation training program, which can improve the intelligence level of training program generation, thereby effectively improving the home rehabilitation training effect of the pelvic floor muscles.

[0005] Although the current rehabilitation program formulation method has improved the rationality and effectiveness of program formulation to a certain extent, the efficiency of program formulation has not been given much consideration. Especially in scenarios where rehabilitation programs need to be generated in large quantities, the efficiency of rehabilitation program formulation is particularly important. Summary of the Invention

[0006] The embodiments of this specification provide a sarcopenia rehabilitation system based on a large artificial intelligence model, which can efficiently formulate and adjust rehabilitation plans.

[0007] The technical solution is as follows:

[0008] The embodiments of this specification provide a sarcopenia rehabilitation system based on an artificial intelligence large model, including:

[0009] The first acquisition module obtains the initial indicator information corresponding to all patients;

[0010] A first clustering module performs a first clustering on all patients based on the initial indicator information corresponding to each of the patients to obtain a plurality of patient clusters;

[0011] The first model module outputs a rehabilitation plan corresponding to the patient cluster and indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the implementation of the rehabilitation plan, which can reflect the expected indicator recovery degree based on the first model and the initial indicator information corresponding to all patients in the patient cluster;

[0012] The second data acquisition module, when reaching any rehabilitation time node, regards the rehabilitation time node as an adjustment time node, and obtains the node indicator information corresponding to each of the patients in the patient cluster at the adjustment time node;

[0013] a second clustering module, performing a second clustering on all patients in the patient cluster based on the node indicator information corresponding to each of the patients in the patient cluster when the time nodes are adjusted and the indicator recovery prediction information corresponding to the adjusted time nodes during the execution of the rehabilitation plan corresponding to the patient cluster, to obtain a plurality of patient cluster subclusters;

[0014] The third data acquisition module obtains the representative indicator recovery deviation information corresponding to the patient cluster sub-clusters;

[0015] The redefinition module restores the deviation information based on the representative indicators corresponding to the patient cluster sub-clusters, and redefines the patient cluster sub-clusters to still belong to the original patient cluster or become a new patient cluster;

[0016] The second largest model module obtains the rehabilitation plan corresponding to the new patient clustering cluster and the indicator recovery prediction information that can reflect the expected indicator recovery degree corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan based on the second largest model and the rehabilitation plan corresponding to the patient clustering sub-cluster that becomes the new patient clustering cluster, and the representative indicator recovery deviation information corresponding to the patient clustering sub-cluster that becomes the new patient clustering cluster.

[0017] As a preferred solution, the first model module outputs the rehabilitation plan corresponding to the patient cluster and the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the implementation of the rehabilitation plan, which can reflect the expected indicator recovery degree, based on the first model and the initial indicator information corresponding to all patients in the patient cluster, including:

[0018] Based on the initial indicator information corresponding to all patients in the patient cluster, obtaining representative indicator information corresponding to the patient cluster;

[0019] Based on the first model and the representative indicator information corresponding to the patient cluster, the rehabilitation plan corresponding to the patient cluster and the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan that can reflect the expected indicator recovery degree are output.

[0020] As a preferred solution, the second clustering module performs a second clustering on all patients in the patient cluster based on the node indicator information corresponding to each of the patients in the patient cluster when adjusting the time node and the indicator recovery prediction information corresponding to the adjustment time node during the execution of the rehabilitation plan corresponding to the patient cluster, so as to obtain multiple patient cluster subclusters, including:

[0021] Based on the node indicator information corresponding to each of the patients in the patient cluster when the time nodes are adjusted and the indicator recovery prediction information corresponding to the adjusted time nodes during the execution of the rehabilitation plan corresponding to the patient cluster, the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when the time nodes are adjusted is obtained;

[0022] Based on the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when adjusting the time node, a second clustering is performed on all the patients in the patient cluster to obtain a plurality of patient cluster subclusters.

[0023] As a preferred solution, the node indicator information corresponding to each of the patients in the patient cluster obtained by the second data acquisition module when adjusting the time node includes node indicator values ​​corresponding to each of the multiple indicators;

[0024] The indicator recovery prediction information corresponding to each of the multiple rehabilitation time nodes during the execution of the rehabilitation plan output by the first model module includes the recovery prediction values ​​corresponding to each of the multiple indicators that can reflect the expected recovery degree of the indicators;

[0025] The node indicator recovery deviation information corresponding to each of the patients in the patient cluster obtained in the second clustering module when adjusting the time node includes the recovery deviation values ​​corresponding to each of the multiple indicators.

[0026] As a preferred solution, the third data acquisition module includes a representative deviation acquisition unit and a deviation information acquisition unit;

[0027] The representative deviation obtaining unit obtains representative deviation values ​​corresponding to the plurality of indicators corresponding to the patient cluster sub-cluster based on the node indicator recovery deviation information corresponding to each of the patients in the patient cluster sub-cluster when the time node is adjusted;

[0028] The deviation information acquisition unit acquires representative indicator recovery deviation information corresponding to the patient cluster sub-clusters based on representative deviation values ​​corresponding to the plurality of indicators corresponding to the patient cluster sub-clusters.

[0029] As a preferred solution, the deviation information acquisition unit includes a positive deviation information acquisition subunit and a negative deviation information acquisition subunit;

[0030] The positive deviation information acquisition subunit regards the indicators whose recovery degree exceeds the expected indicator recovery degree among the multiple indicators corresponding to the patient cluster subclusters as positive deviation indicators, and acquires positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators;

[0031] The negative deviation information acquisition subunit regards the indicators whose recovery degree is lower than the expected indicator recovery degree among the multiple indicators corresponding to the patient cluster subclusters as negative deviation indicators, and obtains negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators.

[0032] As a preferred solution, the positive deviation information acquisition subunit acquires positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators, including:

[0033] Normalizing the representative deviation values ​​corresponding to all positive deviation indicators to obtain first normalized deviation values ​​corresponding to all positive deviation indicators;

[0034] Obtaining a first cumulative deviation value based on the absolute values ​​of the first normalized deviation values ​​corresponding to all positive deviation indicators;

[0035] The negative deviation information acquisition subunit acquires negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators, including:

[0036] Normalizing the representative deviation values ​​corresponding to all negative deviation indicators to obtain second normalized deviation values ​​corresponding to all negative deviation indicators;

[0037] A second cumulative deviation value is obtained based on the absolute values ​​of the second normalized deviation values ​​corresponding to all negative deviation indicators.

[0038] As a preferred solution, the redefinition module redefines the patient cluster subcluster into a new patient cluster cluster when the first cumulative deviation value in the representative indicator recovery deviation information corresponding to the patient cluster subcluster is greater than the first preset threshold or the second cumulative deviation value is greater than the second preset threshold; otherwise, the redefined patient cluster subcluster still belongs to the original patient cluster cluster;

[0039] The first preset threshold is greater than the second preset threshold.

[0040] As a preferred solution, the positive deviation information acquisition subunit acquires the positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators, further comprising:

[0041] Obtaining a first deviation maximum value based on the absolute values ​​of the first normalized deviation values ​​corresponding to all positive deviation indicators;

[0042] The negative deviation information acquisition subunit acquires negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators, further comprising:

[0043] Obtaining a second maximum deviation value based on the absolute values ​​of the second normalized deviation values ​​corresponding to all negative deviation indicators;

[0044] The redefinition module includes a first definition unit and a second definition unit;

[0045] The first definition unit redefines the patient cluster subcluster as a new patient cluster subcluster when the first maximum deviation value or the second maximum deviation value in the representative indicator recovery deviation information corresponding to the patient cluster subcluster is greater than a third preset threshold; otherwise, the patient cluster subcluster is regarded as a cluster subcluster to be reconfirmed;

[0046] The second definition unit redefines the cluster subcluster to be confirmed for the second time as a new patient cluster cluster when the first cumulative deviation value in the representative indicator recovery deviation information corresponding to the cluster subcluster to be confirmed for the second time is greater than the first preset threshold or the second cumulative deviation value is greater than the second preset threshold; otherwise, the redefined cluster subcluster to be confirmed for the second time still belongs to the original patient cluster cluster;

[0047] Among them, the first preset threshold is greater than the second preset threshold, and the second preset threshold is greater than the third preset threshold.

[0048] As a preferred solution, the second largest model module, based on the second largest model and all rehabilitation programs historically experienced by the patient cluster sub-cluster that becomes the new patient cluster cluster, all representative indicator recovery deviation information historically related to the patient cluster sub-cluster that becomes the new patient cluster cluster, and the correspondence between all the related representative indicator recovery deviation information and the rehabilitation time nodes in all the rehabilitation programs historically experienced, obtains the rehabilitation program corresponding to the new patient cluster cluster and the indicator recovery prediction information corresponding to each of the multiple rehabilitation time nodes during the execution of the rehabilitation program, which can reflect the expected indicator recovery level.

[0049] The beneficial effects of the technical solutions provided by some embodiments of this specification include at least:

[0050] First, clustering is performed based on the patient's initial indicator information, and rehabilitation plans are generated based on patient clusters, thereby improving the efficiency of initial rehabilitation plan generation. Furthermore, during the rehabilitation plan execution process, clustering is performed within the patient cluster based on the recovery status of the indicators corresponding to all patients within the patient cluster. The representative indicator recovery deviation information corresponding to all patient cluster subclusters within the patient cluster is obtained, and it is determined whether the rehabilitation plans corresponding to all patients in the patient cluster subclusters need to be adjusted uniformly, further improving the efficiency of subsequent rehabilitation plan adjustments.

[0051] After performing a second clustering operation on all patients in the patient cluster, the representative index recovery deviation information corresponding to the patient cluster sub-cluster is obtained, and based on the representative index recovery deviation information corresponding to the patient cluster sub-cluster, the patient cluster sub-cluster is redefined to still belong to the original patient cluster or become a new patient cluster, that is, the patients in the patient cluster sub-cluster whose representative index recovery deviation information meets the preset requirements will continue to be rehabilitated according to the original corresponding rehabilitation plan, while the patients in the patient cluster sub-cluster whose representative index recovery deviation information does not meet the preset requirements will need to be rehabilitated according to the adjusted rehabilitation plan. In the process of subsequent execution of the new rehabilitation plan, it is necessary to perform a second clustering operation as a new independent patient cluster, and determine whether it is necessary to further adjust the rehabilitation plan for some or all patients in the cluster. That is, a new patient cluster is generated by a layer-by-layer progressive sorting method to ensure the uniformity of all patient information when the rehabilitation plan of all patients in the cluster sub-cluster is uniformly adjusted each time, and to ensure the feasibility of adjusting the rehabilitation plan on a cluster basis. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 This is a structural diagram of a sarcopenia rehabilitation system based on an artificial intelligence large model provided in an embodiment of this specification.

[0054] Figure 2 It is a flowchart of the rehabilitation program adjustment using the layer-by-layer progressive sorting method described in the embodiments of this specification.

[0055] Figure 3 This is a further flowchart of the rehabilitation program adjustment using the layer-by-layer progressive sorting method described in the embodiments of this specification.

[0056] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of this specification will be described clearly and completely below in conjunction with the drawings in the embodiments of this specification.

[0058] Throughout this specification, the claims, and the accompanying drawings, the terms "first," "second," "third," and the like are used to distinguish between different items, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may include other steps or elements inherent to the process, method, product, or apparatus.

[0059] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of this specification. Various examples may appropriately omit, replace, or add various processes or components. For example, the described methods may be performed in an order different from the order described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined in other examples.

[0060] Reference Figure 1 As shown, Figure 1The structural diagram of a sarcopenia rehabilitation system based on an artificial intelligence large model provided in the embodiments of this specification may at least include:

[0061] The first acquisition module obtains the initial indicator information corresponding to all patients.

[0062] It is understandable that when formulating a rehabilitation plan for sarcopenia, the initial indicator information may include but is not limited to age, gender, past medical history, weight, muscle mass, fat mass, water mass, bone mass, protein content, blood sugar content, fatty acid content, amino acid content, basal metabolic rate, activity metabolic rate, etc.

[0063] Gender can be quantified using 1 or 0, for example, male is quantified as 1 and female is quantified as 0. Past medical history can also be quantified using 1 or 0, for example, rheumatism is quantified as 1 and no rheumatism is quantified as 0, to facilitate subsequent clustering operations.

[0064] The system may further include at least a first clustering module, which performs a first clustering on all patients based on the initial indicator information corresponding to all patients to obtain a plurality of patient clusters.

[0065] It should be noted that clustering here involves clustering all patients based on the initial indicator information described above. This clustering can be performed using, but is not limited to, the K-means clustering algorithm. Constraints can be set during the clustering process to ensure the rationality of subsequent rehabilitation plans. For example, patients of different genders must be assigned to different clusters, and patients with and without rheumatic disease must be assigned to different clusters. Specific constraints can be set based on actual needs.

[0066] The system may also include at least a first large model module, which, based on the first large model and the initial indicator information corresponding to all patients in the patient cluster, outputs the rehabilitation plan corresponding to the patient cluster and the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan, which can reflect the expected indicator recovery level.

[0067] Furthermore, the first model module outputs, based on the first model and the initial indicator information corresponding to all patients in the patient cluster, a rehabilitation plan corresponding to the patient cluster and indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan that can reflect the expected indicator recovery degree, including:

[0068] Based on the initial indicator information corresponding to all patients in the patient cluster, obtaining representative indicator information corresponding to the patient cluster;

[0069] Based on the first model and the representative indicator information corresponding to the patient cluster, the rehabilitation plan corresponding to the patient cluster and the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan that can reflect the expected indicator recovery degree are output.

[0070] It can be understood that the first large model is a model obtained by pre-training with the corresponding data set (Note: the training of the model is a conventional technology and will not be elaborated here). The model can output the corresponding rehabilitation plan and the indicator recovery prediction information that can reflect the expected indicator recovery degree corresponding to multiple rehabilitation time nodes during the implementation of the rehabilitation plan based on the input representative indicator information, that is, representative indicator values ​​such as muscle mass, fat mass, water mass, and bone mass.

[0071] It needs to be explained that the multiple rehabilitation time nodes each correspond to indicator recovery prediction information that can reflect the expected degree of recovery of the indicators, that is, during the implementation of the rehabilitation program, there are multiple time nodes, and each time node includes the expected recovery values ​​corresponding to multiple indicators. For example, after the first week of the rehabilitation program, the muscle mass should be restored to 25 kg.

[0072] It should be noted that the representative indicator information corresponding to the patient cluster can be obtained by averaging each indicator in the initial indicator information. For example, if the patient cluster includes patients 1, 2, and 3, and their muscle mass is 20 kg, 22 kg, and 24 kg respectively, the representative indicator information of the patient cluster will include a muscle mass representative indicator value of 22 kg.

[0073] The system may also include at least a second data acquisition module, which, when reaching any rehabilitation time node during the execution of the rehabilitation plan corresponding to the patient cluster, regards the rehabilitation time node as an adjustment time node, and obtains the node indicator information corresponding to each of the patients in the patient cluster at the adjustment time node.

[0074] That is, obtain the node indicator value corresponding to each indicator (note: the actual value).

[0075] The system may also include at least a second clustering module, which performs a second clustering on all patients in the patient cluster based on the node indicator information corresponding to each patient in the patient cluster when adjusting the time node and the indicator recovery prediction information corresponding to the adjusted time node during the execution of the rehabilitation plan corresponding to the patient cluster, so as to obtain multiple patient cluster subclusters.

[0076] Furthermore, the second clustering module performs a second clustering on all patients in the patient cluster based on the node indicator information corresponding to each of the patients in the patient cluster when adjusting the time node and the indicator recovery prediction information corresponding to the adjustment time node during the execution of the rehabilitation plan corresponding to the patient cluster, so as to obtain multiple patient cluster subclusters, including:

[0077] Based on the node indicator information corresponding to each of the patients in the patient cluster when the time nodes are adjusted and the indicator recovery prediction information corresponding to the adjusted time nodes during the execution of the rehabilitation plan corresponding to the patient cluster, the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when the time nodes are adjusted is obtained;

[0078] Based on the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when adjusting the time node, a second clustering is performed on all the patients in the patient cluster to obtain a plurality of patient cluster subclusters.

[0079] It should be noted that the node indicator recovery deviation information may include the deviation value between the node indicator value corresponding to each indicator and the expected recovery indicator value, which can be understood as:

[0080] The node indicator information corresponding to each of the patients in the patient cluster obtained by the second data acquisition module when adjusting the time node includes node indicator values ​​corresponding to each of the multiple indicators;

[0081] The indicator recovery prediction information corresponding to each of the multiple rehabilitation time nodes during the execution of the rehabilitation plan output by the first model module includes the recovery prediction values ​​corresponding to each of the multiple indicators that can reflect the expected recovery degree of the indicators;

[0082] The node indicator recovery deviation information corresponding to each of the patients in the patient cluster obtained in the second clustering module when adjusting the time node includes the recovery deviation values ​​corresponding to each of the multiple indicators.

[0083] The calculation of the recovery deviation value can be:

[0084] Node indicator value of the indicator - recovery prediction value of the indicator;

[0085] For example:

[0086] The node index value of the muscle mass index is 23kg, and the recovery prediction value of the muscle mass index is 25kg. The recovery deviation value of the muscle mass index is 23kg-25kg=-2kg.

[0087] The node index value of the muscle mass index is 27kg, and the recovery prediction value of the muscle mass index is 25kg. Then the recovery deviation value of the muscle mass index is 27kg-25kg=+2kg.

[0088] It can be understood that the second clustering module is similar to the first clustering module and can also use, but is not limited to, the K-means clustering algorithm for clustering, which will not be elaborated herein.

[0089] The system may further include at least a third data acquisition module for acquiring representative indicator recovery deviation information corresponding to the patient cluster sub-clusters. Furthermore, the representative indicator recovery deviation information corresponding to the patient cluster sub-clusters may be acquired based on the node indicator recovery deviation information corresponding to each patient in the patient cluster sub-clusters when the time node is adjusted.

[0090] It should be noted that the representative indicator recovery deviation information for a patient cluster subcluster can be obtained by averaging the node indicator recovery deviation information corresponding to each patient in the patient cluster subcluster at the time of adjustment. For example, if a patient cluster subcluster includes patients 1, 2, and 3, and their muscle mass indicator recovery deviation values ​​are 2kg, -2kg, and 0kg, respectively, then the representative indicator recovery deviation information for the patient cluster subcluster includes a representative deviation value of 0kg for the muscle mass indicator.

[0091] The system may further include at least a redefinition module, which restores deviation information based on representative indicators corresponding to the patient cluster subclusters, and redefines the patient cluster subclusters to still belong to the original patient cluster cluster or become a new patient cluster cluster.

[0092] The system may also include at least a second largest model module, which obtains the rehabilitation plan corresponding to the new patient clustering cluster and the indicator recovery prediction information that can reflect the expected indicator recovery degree corresponding to each of the multiple rehabilitation time nodes during the execution of the rehabilitation plan based on the second largest model and the rehabilitation plan corresponding to the patient clustering sub-cluster that becomes the new patient clustering cluster and the representative indicator recovery deviation information corresponding to the patient clustering sub-cluster that becomes the new patient clustering cluster.

[0093] It should be noted that the rehabilitation program may include, but is not limited to, an exercise training program, a diet program, a drug treatment program, and the like.

[0094] The following Figure 2 Explain the redefined module (Note: For the repeated explanation of the corresponding module, please refer to the above explanation, and no repeated explanation will be given below):

[0095] First, all patients are clustered using the first clustering module to obtain patient cluster 1, patient cluster 2, and patient cluster 3.

[0096] Furthermore, the rehabilitation plans corresponding to patient cluster 1, patient cluster 2, and patient cluster 3, as well as the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the implementation of the rehabilitation plan, which can reflect the expected degree of recovery of the indicators, are obtained through the first large model module.

[0097] Furthermore, for any patient cluster, when reaching any rehabilitation time node in the execution process of its corresponding rehabilitation program, the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when adjusting the time node is obtained.

[0098] Furthermore, based on the node indicator recovery deviation information corresponding to all patients in the patient cluster when adjusting the time node, a second clustering operation is performed on all patients in the patient cluster to obtain multiple patient cluster subclusters, such as patient cluster subcluster 1, patient cluster subcluster 2, and patient cluster subcluster 3 in patient cluster cluster 1, such as patient cluster subcluster 4, patient cluster subcluster 5, and patient cluster subcluster 6 in patient cluster cluster 2, and patient cluster subcluster 7, patient cluster subcluster 8, and patient cluster subcluster 9 in patient cluster cluster 3.

[0099] Furthermore, based on the node indicator recovery deviation information corresponding to all patients in the patient cluster sub-cluster when adjusting the time node, the representative indicator recovery deviation information corresponding to the patient cluster sub-cluster is obtained.

[0100] Furthermore, based on the representative index corresponding to the patient cluster sub-cluster, the deviation information is restored and the patient cluster sub-cluster is redefined to still belong to the original patient cluster or become a new patient cluster. Figure 2 As shown in , if patient cluster subcluster 1 does not meet the preset requirements, it will be redefined as a new patient cluster 4, while patient cluster subcluster 2 and patient cluster subcluster 3 will be redefined as still belonging to patient cluster 1 because they meet the preset requirements, and the rehabilitation plan corresponding to the new patient cluster 4 and the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan that can reflect the expected indicator recovery degree are re-obtained.

[0101] It is important to note that at this time, patient cluster subcluster 1 becomes the new patient cluster cluster 4, and patient cluster subcluster 2 and patient cluster subcluster 3 are redefined as still belonging to the original patient cluster cluster. Therefore, patient cluster subcluster 1, patient cluster subcluster 2, and patient cluster subcluster 3 will no longer be used subsequently. Subsequently, the second data acquisition module and the second clustering module should be used to re-perform data acquisition, clustering, and other operations on the patients in patient cluster cluster 1 and the new patient cluster cluster 4 respectively, so as to realize the progressive sorting of patients at each level.

[0102] It can be seen that in the rehabilitation system provided by multiple embodiments of this specification, clustering operations are first performed based on the initial indicator information of the patient, and rehabilitation plans are generated based on patient clusters, thereby improving the efficiency of generating the initial rehabilitation plan. Furthermore, during the execution of the rehabilitation plan, clustering operations are still performed within the patient cluster based on the recovery status of the indicators corresponding to all patients within the patient cluster, using the patient cluster as a unit. The representative indicator recovery deviation information corresponding to all patient cluster subclusters within the patient cluster is obtained, and it is determined whether it is necessary to uniformly adjust the rehabilitation plans corresponding to all patients within the patient cluster subclusters, further improving the efficiency of adjusting subsequent rehabilitation plans.

[0103] In the rehabilitation system provided by multiple embodiments of this specification, representative index recovery deviation information corresponding to patient cluster subclusters is obtained, and based on the representative index recovery deviation information corresponding to the patient cluster subclusters, the patient cluster subclusters are redefined to still belong to the original patient cluster cluster or become a new patient cluster cluster, that is, the patients in the patient cluster subclusters whose representative index recovery deviation information meets the preset requirements will continue to be rehabilitated according to the original corresponding rehabilitation plan, while the patients in the patient cluster subclusters whose representative index recovery deviation information does not meet the preset requirements will need to be rehabilitated according to the adjusted rehabilitation plan, and in the process of subsequent execution of the new rehabilitation plan, it is necessary to perform a second clustering operation as a new independent patient cluster cluster, and determine whether it is necessary to further adjust the rehabilitation plan for some or all patients in the cluster. That is, a new patient cluster cluster is generated by a layer-by-layer progressive sorting method to ensure the uniformity of all patient information when the rehabilitation plan for all patients in the cluster subclusters is uniformly adjusted each time, and to ensure the feasibility of adjusting the rehabilitation plan on a cluster basis.

[0104] In some embodiments of this specification, the third data acquisition module includes a representative deviation acquisition unit and a deviation information acquisition unit;

[0105] The representative deviation obtaining unit obtains representative deviation values ​​corresponding to the plurality of indicators corresponding to the patient cluster sub-cluster based on the node indicator recovery deviation information corresponding to each of the patients in the patient cluster sub-cluster when the time node is adjusted;

[0106] The deviation information acquisition unit acquires representative indicator recovery deviation information corresponding to the patient cluster sub-clusters based on representative deviation values ​​corresponding to the plurality of indicators corresponding to the patient cluster sub-clusters.

[0107] That is, consistent with the above explanation, the representative indicator recovery deviation information corresponding to the patient cluster subcluster is obtained based on the representative deviation values ​​corresponding to multiple indicators. However, it should be noted that in addition to directly including the representative deviation values ​​corresponding to multiple indicators, the representative indicator recovery deviation information can also include other information obtained from the representative deviation values ​​corresponding to multiple indicators. The specific information can be set according to actual needs.

[0108] It is understandable that, in order to determine whether the rehabilitation program needs to be adjusted and how to adjust it, the indicator data that exceeds the expected recovery level and the indicator data that falls below the expected recovery level should be analyzed accordingly to provide data support for the rehabilitation program adjustment decision. Therefore, in some embodiments of this specification, the deviation information acquisition unit includes a positive deviation information acquisition subunit and a negative deviation information acquisition subunit;

[0109] The positive deviation information acquisition subunit regards the indicators whose recovery degree exceeds the expected indicator recovery degree among the multiple indicators corresponding to the patient cluster subclusters as positive deviation indicators, and acquires positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators;

[0110] The negative deviation information acquisition subunit regards the indicators whose recovery degree is lower than the expected indicator recovery degree among the multiple indicators corresponding to the patient cluster subclusters as negative deviation indicators, and obtains negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators.

[0111] It should be noted here that among the multiple indicators corresponding to the patient cluster subclusters, the indicators whose recovery degree exceeds the expected indicator recovery degree are regarded as positive deviation indicators, and among the multiple indicators corresponding to the patient cluster subclusters, the indicators whose recovery degree is lower than the expected indicator recovery degree are regarded as negative deviation indicators, rather than those representing positive deviation values ​​being regarded as positive deviation indicators and those representing negative deviation values ​​being regarded as positive deviation indicators. The following is an explanation:

[0112] The expected direction of recovery for indicators varies. For example, if the expected direction of recovery for blood sugar is an increase, a representative deviation value of +2 for the blood sugar indicator is considered a positive deviation indicator. On the other hand, a representative deviation value of -2 for the blood sugar indicator is considered a negative deviation indicator.

[0113] For example, if the expected recovery direction of blood sugar is a decrease, then even if the representative deviation value of the blood sugar index is -2, it is still considered a positive deviation indicator. However, if the representative deviation value of the blood sugar index is +2, it is considered a negative deviation indicator.

[0114] Therefore, we cannot directly judge whether the indicator is a positive deviation indicator or a negative deviation indicator based on the positive or negative value of the indicator.

[0115] It is understandable that the corresponding analysis of indicator data that exceeds the expected recovery level and indicator data that falls below the expected recovery level may include, but is not limited to, the calculation of the cumulative deviation. In addition, due to the differences in the numerical ranges between the data, normalization is required before calculating the cumulative amount. In addition, based on the above, the representative deviation value of the positive deviation indicator may be positive or negative, and the representative deviation value of the negative deviation indicator may also be positive or negative. Therefore, when performing the accumulation, it is necessary to perform it based on the absolute value.

[0116] Therefore, in some embodiments of the present specification, the positive deviation information obtaining subunit obtains positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators, including:

[0117] Normalizing the representative deviation values ​​corresponding to all positive deviation indicators to obtain first normalized deviation values ​​corresponding to all positive deviation indicators;

[0118] Obtaining a first cumulative deviation value based on the absolute values ​​of the first normalized deviation values ​​corresponding to all positive deviation indicators;

[0119] The negative deviation information acquisition subunit acquires negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators, including:

[0120] Normalizing the representative deviation values ​​corresponding to all negative deviation indicators to obtain second normalized deviation values ​​corresponding to all negative deviation indicators;

[0121] A second cumulative deviation value is obtained based on the absolute values ​​of the second normalized deviation values ​​corresponding to all negative deviation indicators.

[0122] In some embodiments of the present specification, the redefinition module redefines the patient cluster subcluster into a new patient cluster cluster when the first cumulative deviation value in the representative indicator recovery deviation information corresponding to the patient cluster subcluster is greater than a first preset threshold or the second cumulative deviation value is greater than a second preset threshold; otherwise, the redefined patient cluster subcluster still belongs to the original patient cluster cluster;

[0123] The first preset threshold is greater than the second preset threshold.

[0124] It can be understood that if the first cumulative deviation value is greater than the first preset threshold or the absolute value of the second cumulative deviation value is greater than the second preset threshold, it indicates that when the rehabilitation plan is executed to this rehabilitation time node, it is already significantly different from the expected recovery situation. Therefore, it is necessary to separate from the original patient cluster cluster and redefine the patient cluster sub-cluster into a new patient cluster cluster in order to independently carry out subsequent rehabilitation plan adjustment operations.

[0125] It should also be noted that, ideally, the patient's recovery indicators fully align with the expected development during the rehabilitation program. However, deviations are inevitable during implementation. Therefore, the first and second preset thresholds are set to allow for these deviations. Furthermore, the tolerance for exceeding the expected recovery level during rehabilitation should be greater than the tolerance for falling short of the expected recovery level (Note: It is understandable that the actual recovery rate may be faster than the expected recovery rate to a certain extent, which is generally permitted). Therefore, the first preset threshold is greater than the second preset threshold.

[0126] It is further understood that to avoid excessive deviation of a single indicator, the following two situations can be avoided:

[0127] Case 1:

[0128] The expected direction of blood sugar recovery is upward, and the representative deviation value of the blood sugar index is +3. At this time, it is considered a positive deviation indicator. However, +3 is too different from the expected recovery level, which affects the normal progress of the rehabilitation program.

[0129] Case 2:

[0130] The expected recovery direction of blood sugar is upward, and the representative deviation value of the blood sugar index is +5. At this time, it is considered to be a positive deviation indicator. However, +5 is not only too different from the expected degree of recovery, affecting the normal progress of the rehabilitation program, but +5 is also likely to cause blood sugar to exceed the normal value range after complete recovery, that is, there is over-recovery.

[0131] Therefore, analysis of indicator data exceeding or falling below the expected recovery level can also include determining whether the absolute value of each indicator's corresponding representative deviation exceeds a preset amount. Furthermore, based on the aforementioned normalization of all positive and negative deviation indicators, whether the absolute value of each indicator's corresponding representative deviation exceeds a preset amount can be determined using a unified preset threshold to improve program execution efficiency.

[0132] Therefore, in some embodiments of the present specification, the positive deviation information obtaining subunit obtains the positive deviation information based on the representative deviation values ​​corresponding to all the positive deviation indicators, further comprising:

[0133] Based on the absolute values ​​of the first normalized deviation values ​​corresponding to all positive deviation indicators, a first deviation maximum value (i.e., the maximum value among the absolute values ​​of all first normalized deviation values) is obtained;

[0134] The negative deviation information acquisition subunit acquires negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators, further comprising:

[0135] Based on the absolute values ​​of the second normalized deviation values ​​corresponding to all negative deviation indicators, a second deviation maximum value (ie, the maximum value among the absolute values ​​of all second normalized deviation values) is obtained.

[0136] Furthermore, the redefinition module includes a first definition unit and a second definition unit;

[0137] The first definition unit redefines the patient cluster subcluster as a new patient cluster subcluster when the first maximum deviation value or the second maximum deviation value in the representative indicator recovery deviation information corresponding to the patient cluster subcluster is greater than a third preset threshold; otherwise, the patient cluster subcluster is regarded as a cluster subcluster to be reconfirmed;

[0138] The second definition unit redefines the cluster subcluster to be confirmed for the second time as a new patient cluster cluster when the first cumulative deviation value in the representative indicator recovery deviation information corresponding to the cluster subcluster to be confirmed for the second time is greater than the first preset threshold or the second cumulative deviation value is greater than the second preset threshold; otherwise, the redefined cluster subcluster to be confirmed for the second time still belongs to the original patient cluster cluster;

[0139] Among them, the first preset threshold is greater than the second preset threshold, and the second preset threshold is greater than the third preset threshold.

[0140] In some embodiments of the present specification, the second largest model module obtains the rehabilitation plan corresponding to the new patient clustering cluster and the indicator recovery prediction information that can reflect the expected indicator recovery degree corresponding to each of the multiple rehabilitation time nodes in the execution of the rehabilitation plan based on the second largest model and all the rehabilitation plans corresponding to the patient clustering subcluster that becomes the new patient clustering cluster in history, all the representative indicator recovery deviation information related to the patient clustering subcluster that becomes the new patient clustering cluster in history, and the correspondence between all the related representative indicator recovery deviation information and the rehabilitation time nodes in all the rehabilitation plans experienced in history.

[0141] Reference Figure 2 、 Figure 3As shown, it can be understood that, for the patient clustering sub-cluster 10, all the rehabilitation programs it has experienced in history include the rehabilitation program corresponding to the patient clustering cluster 1 and the rehabilitation program corresponding to the patient clustering cluster 4. All the representative index recovery deviation information related to history includes not only the representative index recovery deviation information corresponding to the patient clustering sub-cluster 10 itself, but also the representative index recovery deviation information corresponding to the patient clustering sub-cluster 1 (note: because all the patients in the patient clustering sub-cluster 10 exist in the patient clustering sub-cluster 1). Therefore, by comprehensively considering the rehabilitation program corresponding to the patient clustering cluster 1, the rehabilitation program corresponding to the patient clustering cluster 4, the representative index recovery deviation information corresponding to the patient clustering sub-cluster 10 itself, the representative index recovery deviation information corresponding to the patient clustering sub-cluster 1, and the corresponding relationship between the relevant all representative index recovery deviation information and the rehabilitation time node in the history experienced all the rehabilitation programs (note: that is, the representative index recovery deviation information of the patient clustering sub-cluster 10 corresponds to which rehabilitation time node in the process of executing the rehabilitation program corresponding to the patient clustering cluster 4, and the representative index recovery deviation information of the patient clustering sub-cluster 1 corresponds to which rehabilitation time node in the process of executing the rehabilitation program corresponding to the patient clustering cluster 1), the index recovery prediction information that can reflect the expected index recovery degree corresponding to the rehabilitation program corresponding to the new patient clustering cluster 5 and the respective rehabilitation time node in the process of executing the rehabilitation program can be more reasonably obtained.

[0142] The embodiment of the present specification also provides a sarcopenia rehabilitation method based on an artificial intelligence large model, which can at least include:

[0143] Step 102, obtaining the initial index information corresponding to each of all the patients;

[0144] Step 104, performing a first clustering operation on all the patients based on the initial index information corresponding to each of all the patients, to obtain a plurality of patient clustering clusters;

[0145] Step 106, obtaining the representative index information corresponding to the patient clustering cluster based on the initial index information corresponding to each of all the patients in the patient clustering cluster;

[0146] Step 108, outputting the rehabilitation program corresponding to the patient clustering cluster and the index recovery prediction information that can reflect the expected index recovery degree corresponding to each of a plurality of rehabilitation time nodes in the process of executing the rehabilitation program based on the first large model and the representative index information corresponding to the patient clustering cluster;

[0147] Step 110, when any rehabilitation time node in the process of executing the rehabilitation program corresponding to the patient clustering cluster is reached, regarding the rehabilitation time node as an adjustment time node, and obtaining the node index information corresponding to each of all the patients in the patient clustering cluster at the adjustment time node;

[0148] In step 112, node index recovery deviation information of each patient in the patient cluster at the adjustment time node is obtained based on the node index information of each patient in the patient cluster at the adjustment time node and the index recovery prediction information of the adjustment time node in the process of executing the rehabilitation scheme corresponding to the patient cluster.

[0149] In step 114, a second clustering operation is performed on all patients in the patient cluster based on the node index recovery deviation information of each patient in the patient cluster at the adjustment time node, to obtain a plurality of patient clustering sub-clusters.

[0150] In step 116, representative index recovery deviation information of the patient clustering sub-cluster is obtained based on the node index recovery deviation information of each patient in the patient clustering sub-cluster at the adjustment time node.

[0151] In step 118, the patient clustering sub-cluster is redefined to belong to the original patient cluster or become a new patient cluster based on the representative index recovery deviation information of the patient clustering sub-cluster. When there is no newly established patient cluster, the process returns to step 110, otherwise, step 120 is performed.

[0152] In step 120, the rehabilitation scheme corresponding to the new patient cluster and the index recovery prediction information corresponding to each rehabilitation time node in the process of executing the rehabilitation scheme are obtained based on the second model, the rehabilitation scheme corresponding to the patient clustering sub-cluster of the new patient cluster, and the representative index recovery deviation information corresponding to the patient clustering sub-cluster of the new patient cluster.

[0153] In step 122, the process returns to step 110.

[0154] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to. Each embodiment focuses on the difference from other embodiments. In particular, for the sarcopenia rehabilitation method embodiment, it is basically similar to the sarcopenia rehabilitation system embodiment, so the description is relatively simple, and the relevant parts can be referred to the part of the sarcopenia rehabilitation system embodiment.

[0155] Please refer to Figure 4 The electronic device provided by the embodiment of the specification is shown in a structural schematic diagram.

[0156] As Figure 4 shown, the electronic device 400 can include at least one processor 401, at least one network interface 404, a user interface 403, a memory 405, and at least one communication bus 402.

[0157] The communication bus 402 can be used to realize the connection and communication of the above-mentioned components.

[0158] The user interface 403 can include a button, and the optional user interface can also include a standard wired interface, a wireless interface.

[0159] The network interface 404 can include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0160] The processor 401 can include one or more processing cores. The processor 401 connects various parts in the entire electronic device 400 through various interfaces and lines, executes various functions of the electronic device 400 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one of the hardware forms of DSP, FPGA, PLA. The processor 401 can integrate one or a combination of CPU, GPU and modem. Among them, the CPU mainly processes the operating system, user interface and application program; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used for processing wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 401, but be realized by a separate chip.

[0161] The memory 405 can include RAM and ROM. Optionally, the memory 405 includes a non-transitory computer readable medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 405 can also be at least one storage device located away from the above-mentioned processor 401. The memory 405 as a computer storage medium can include an operating system, a network communication module, a user interface module and a sarcopenia rehabilitation application. The processor 401 can be used to call the sarcopenia rehabilitation application stored in the memory 405, and execute the steps of the sarcopenia rehabilitation method mentioned in the above-mentioned embodiments.

[0162] The embodiments of this specification also provide a computer-readable storage medium containing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps of the sarcopenia rehabilitation method embodiments described above. If the components of the electronic device described above are implemented as software functional units and sold or used as independent products, they may be stored in the computer-readable storage medium.

[0163] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state drive (SSD)).

[0164] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks. The technical features of this embodiment and the implementation scheme can be combined in any manner unless they conflict.

[0165] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A sarcopenia rehabilitation system based on an artificial intelligence large model, characterized by: include: The first acquisition module obtains the initial indicator information corresponding to all patients; A first clustering module performs a first clustering on all patients based on the initial indicator information corresponding to each of the patients to obtain a plurality of patient clusters; The first model module outputs a rehabilitation plan corresponding to the patient cluster and indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the implementation of the rehabilitation plan, which can reflect the expected indicator recovery degree based on the first model and the initial indicator information corresponding to all patients in the patient cluster; The second data acquisition module, when reaching any rehabilitation time node, regards the rehabilitation time node as an adjustment time node, and obtains the node indicator information corresponding to each of the patients in the patient cluster at the adjustment time node; a second clustering module, performing a second clustering on all patients in the patient cluster based on the node indicator information corresponding to each of the patients in the patient cluster when the time nodes are adjusted and the indicator recovery prediction information corresponding to the adjusted time nodes during the execution of the rehabilitation plan corresponding to the patient cluster, to obtain a plurality of patient cluster subclusters; The third data acquisition module obtains the representative indicator recovery deviation information corresponding to the patient cluster sub-clusters; The redefinition module restores the deviation information based on the representative indicators corresponding to the patient cluster sub-clusters, and redefines the patient cluster sub-clusters to still belong to the original patient cluster or become a new patient cluster; The second largest model module obtains the rehabilitation plan corresponding to the new patient clustering cluster and the indicator recovery prediction information that can reflect the expected indicator recovery degree corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan based on the second largest model and the rehabilitation plan corresponding to the patient clustering sub-cluster that becomes the new patient clustering cluster and the representative indicator recovery deviation information corresponding to the patient clustering sub-cluster that becomes the new patient clustering cluster.

2. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 1 is characterized in that: The first model module outputs, based on the first model and the initial indicator information corresponding to all patients in the patient cluster, a rehabilitation plan corresponding to the patient cluster and indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the implementation of the rehabilitation plan that can reflect the expected indicator recovery degree, including: Based on the initial indicator information corresponding to all patients in the patient cluster, obtaining representative indicator information corresponding to the patient cluster; Based on the first model and the representative indicator information corresponding to the patient cluster, the rehabilitation plan corresponding to the patient cluster and the indicator recovery prediction information corresponding to multiple rehabilitation time nodes during the execution of the rehabilitation plan that can reflect the expected indicator recovery degree are output.

3. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 1 is characterized in that: The second clustering module performs a second clustering on all patients in the patient cluster based on the node indicator information corresponding to each of the patients in the patient cluster when adjusting the time node and the indicator recovery prediction information corresponding to the adjustment time node during the execution of the rehabilitation plan corresponding to the patient cluster, to obtain multiple patient cluster subclusters, including: Based on the node indicator information corresponding to each of the patients in the patient cluster when the time nodes are adjusted and the indicator recovery prediction information corresponding to the adjusted time nodes during the execution of the rehabilitation plan corresponding to the patient cluster, the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when the time nodes are adjusted is obtained; Based on the node indicator recovery deviation information corresponding to each of the patients in the patient cluster when adjusting the time node, a second clustering is performed on all the patients in the patient cluster to obtain a plurality of patient cluster subclusters.

4. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 3 is characterized in that: The node indicator information corresponding to each of the patients in the patient cluster obtained by the second data acquisition module when adjusting the time node includes node indicator values ​​corresponding to each of the multiple indicators; The indicator recovery prediction information corresponding to each of the multiple rehabilitation time nodes during the execution of the rehabilitation plan output by the first model module includes the recovery prediction values ​​corresponding to each of the multiple indicators that can reflect the expected recovery degree of the indicators; The node indicator recovery deviation information corresponding to each of the patients in the patient cluster obtained in the second clustering module when adjusting the time node includes the recovery deviation values ​​corresponding to each of the multiple indicators.

5. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 4 is characterized in that: The third data acquisition module includes a representative deviation acquisition unit and a deviation information acquisition unit; The representative deviation obtaining unit obtains representative deviation values ​​corresponding to the plurality of indicators corresponding to the patient cluster sub-cluster based on the node indicator recovery deviation information corresponding to each of the patients in the patient cluster sub-cluster when the time node is adjusted; The deviation information acquisition unit acquires representative indicator recovery deviation information corresponding to the patient cluster sub-clusters based on representative deviation values ​​corresponding to the plurality of indicators corresponding to the patient cluster sub-clusters.

6. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 5 is characterized in that: The deviation information acquisition unit includes a positive deviation information acquisition subunit and a negative deviation information acquisition subunit; The positive deviation information acquisition subunit regards the indicators whose recovery degree exceeds the expected indicator recovery degree among the multiple indicators corresponding to the patient cluster subclusters as positive deviation indicators, and acquires positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators; The negative deviation information acquisition subunit regards the indicators whose recovery degree is lower than the expected indicator recovery degree among the multiple indicators corresponding to the patient cluster subclusters as negative deviation indicators, and obtains negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators.

7. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 6 is characterized in that: The positive deviation information acquisition subunit acquires positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators, including: Normalizing the representative deviation values ​​corresponding to all positive deviation indicators to obtain first normalized deviation values ​​corresponding to all positive deviation indicators; Obtaining a first cumulative deviation value based on the absolute values ​​of the first normalized deviation values ​​corresponding to all positive deviation indicators; The negative deviation information acquisition subunit acquires negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators, including: Normalizing the representative deviation values ​​corresponding to all negative deviation indicators to obtain second normalized deviation values ​​corresponding to all negative deviation indicators; A second cumulative deviation value is obtained based on the absolute values ​​of the second normalized deviation values ​​corresponding to all negative deviation indicators.

8. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 7 is characterized in that: The redefinition module redefines the patient cluster subcluster into a new patient cluster cluster when the first cumulative deviation value in the representative indicator recovery deviation information corresponding to the patient cluster subcluster is greater than the first preset threshold or the second cumulative deviation value is greater than the second preset threshold; otherwise, the redefined patient cluster subcluster still belongs to the original patient cluster cluster; The first preset threshold is greater than the second preset threshold.

9. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 7, characterized in that: The positive deviation information acquisition subunit acquires positive deviation information based on the representative deviation values ​​corresponding to all positive deviation indicators, further comprising: Obtaining a first deviation maximum value based on the absolute values ​​of the first normalized deviation values ​​corresponding to all positive deviation indicators; The negative deviation information acquisition subunit acquires negative deviation information based on the representative deviation values ​​corresponding to all negative deviation indicators, further comprising: Obtaining a second maximum deviation value based on the absolute values ​​of the second normalized deviation values ​​corresponding to all negative deviation indicators; The redefinition module includes a first definition unit and a second definition unit; The first definition unit redefines the patient cluster subcluster as a new patient cluster subcluster when the first maximum deviation value or the second maximum deviation value in the representative indicator recovery deviation information corresponding to the patient cluster subcluster is greater than a third preset threshold; otherwise, the patient cluster subcluster is regarded as a cluster subcluster to be reconfirmed; The second definition unit redefines the cluster subcluster to be confirmed for the second time as a new patient cluster cluster when the first cumulative deviation value in the representative indicator recovery deviation information corresponding to the cluster subcluster to be confirmed for the second time is greater than the first preset threshold or the second cumulative deviation value is greater than the second preset threshold; otherwise, the redefined cluster subcluster to be confirmed for the second time still belongs to the original patient cluster cluster; Among them, the first preset threshold is greater than the second preset threshold, and the second preset threshold is greater than the third preset threshold.

10. The sarcopenia rehabilitation system based on artificial intelligence large model according to claim 1 is characterized in that: The second largest model module, based on the second largest model and all rehabilitation programs historically experienced by the patient cluster sub-cluster that becomes the new patient cluster cluster, all representative indicator recovery deviation information historically related to the patient cluster sub-cluster that becomes the new patient cluster cluster, and the correspondence between all the related representative indicator recovery deviation information and the rehabilitation time nodes in all the rehabilitation programs historically experienced, obtains the rehabilitation program corresponding to the new patient cluster cluster and the indicator recovery prediction information corresponding to each of the multiple rehabilitation time nodes during the execution of the rehabilitation program, which can reflect the expected indicator recovery level.

Images