Muscle weakness identification method and system for critical patient

Abstract

The invention discloses a method and a system for identifying muscular weakness of a critical patient, and relates to the field of muscular weakness identification. Subjective clinical muscular strength grading of the patient and instrument detection data are integrated, a muscular weakness grade mapping model of multiple parts is established, and the diagnosis accuracy is greatly improved; through a host / guest part mapping model, namely in combination with subjective feeling and objective detection data of a patient, the medullary muscular weakness symptom prone to missed diagnosis in traditional examination can be identified; on the basis of a muscle weakness index data prediction mechanism, a muscle weakness crisis risk can be warned in advance; through a treatment-evaluation-adjustment closed-loop system, dynamic optimization of a medicine scheme is realized, the crisis occurrence rate can be greatly reduced, the dosage of an immunosuppressive agent is reduced, meanwhile, the treatment effect is maintained, and the system is suitable for patient groups managed for a long time; by establishing a mapping system of a part type set and an index grade set, quantitative unified evaluation of different types such as eye muscle type / whole body type and the like is realized, and subjective judgment errors are reduced.

Description

Technical Field

[0001] The present invention belongs to the field of muscle weakness identification, and in particular, relates to a method and system for identifying muscle weakness in critically ill patients. Background Art

[0002] In the process of identifying muscle weakness, traditional muscle strength grading relies on subjective judgment, making it difficult to quantify symptoms in specific areas such as ptosis and respiratory muscles. Existing methods require waiting for obvious symptoms to appear before intervention, and cannot identify the risk of muscle weakness crisis in advance. Traditional solutions lack data support for drug selection and cannot achieve real-time optimization of drug types / doses. Summary of the Invention

[0003] In response to the problems in the related art, the present invention proposes a method and system for identifying muscle weakness in critically ill patients to overcome the above-mentioned technical problems existing in the existing related art.

[0004] To solve the above technical problems, the present invention is achieved through the following technical solutions: The present invention provides a method for identifying muscle weakness in critically ill patients, comprising the following steps: S1. Set the muscle weakness measurement index types and index levels corresponding to the types of body parts where the patient's muscle weakness manifests; S2. Collect corresponding historical data according to the muscle weakness measurement indicator type and indicator level set in S1, and construct a mapping model for the overall muscle weakness severity level data of each muscle weakness manifestation part based on the historical data; S3, collecting the indicator level data corresponding to each muscle weakness measurement indicator of the patient to be identified and inputting them into the corresponding mapping model constructed in S2 for mapping; S4, collecting various muscle weakness instrument detection indicators of the patient to be identified, and then inputting them into multiple mapping models constructed based on the instrument detection indicator data corresponding to the historical data in S2 for mapping; S5. Compare the mapping results in S3 and S4 with the corresponding preset thresholds, and determine whether to execute S6 or predict the overall muscle weakness level data at a future time based on the comparison result, and then determine whether to execute S6 or not to perform treatment based on the prediction result; S6. Treat the identified patient and repeat S3, S4 and S5 after treatment. After the repetition, if the comparison result and the prediction result in S5 meet the conditions, the treatment is completed; otherwise, adjust the current initial drug treatment plan and repeat S6.

[0005] Preferably, the S1 comprises the following steps: S11. Setting several methods for identifying muscle weakness in critically ill patients to obtain a set of muscle weakness identification methods; the set of muscle weakness identification methods includes a method for collecting patient subjective feeling data and a method for collecting instrument detection data; S12. Based on the method for collecting the patient's subjective feeling data, set several body part types where the patient's muscle weakness manifests, to obtain a set of muscle weakness manifesting body part types; based on the set of muscle weakness manifesting body part types, set a corresponding muscle weakness measurement indicator type for each muscle weakness manifesting body part type, to obtain a set of muscle weakness measurement indicator types; based on the level type of each type in the set of muscle weakness measurement indicator types, to obtain a set of muscle weakness measurement indicator levels; The full-cycle symptom capture capability can accurately identify the full course of the disease from early ptosis to late respiratory muscle involvement through the quantitative assessment of multiple fine indicators in multiple key parts such as the eyelids and arms, combined with grading standards. It can capture the symptom change curve through regular repeated measurements, avoiding missed diagnosis caused by traditional single examinations; multimodal diagnosis synergy enhances the synergy by cross-validating the patient's subjective description of the degree of dysphagia with laboratory antibody test data and instrument electrophysiological examination results, significantly improving diagnostic specificity; for example, when a patient complains of dysphonia, if there are both positive acetylcholine receptor antibodies and abnormal repetitive nerve electrical stimulation, neuromuscular junction lesions can be quickly identified; by setting a set of muscle weakness measurement indicators and a set of muscle weakness measurement indicator types, it provides a basis for the subsequent determination of the level corresponding to each type of muscle weakness measurement indicator, and then determines the comprehensive muscle weakness level of each part of the muscle weakness manifestation, and then provides subjective feeling data for the subsequent determination of the type of muscle weakness suffered by the patient.

[0006] Preferably, said S2 comprises the following steps: S21. Based on the muscle weakness manifestation site type set, the muscle weakness measurement indicator type set, and the muscle weakness measurement indicator level set, collecting indicator level data corresponding to each muscle weakness measurement indicator of multiple historical muscle weakness patients and overall muscle weakness severity level data of the corresponding muscle weakness manifestation sites to obtain a historical muscle weakness indicator level dataset and a historical overall muscle weakness level dataset; S22, constructing mapping models corresponding to the overall muscle weakness severity level data for each muscle weakness manifestation site based on the historical muscle weakness index level dataset and the historical overall muscle weakness level dataset, to obtain a main site muscle weakness severity level mapping model set; By establishing a mapping system between a set of site types and a set of indicator levels, a unified quantitative assessment of different types such as ocular muscle type / systemic type is achieved, reducing subjective judgment errors; the model supports the conversion of discrete patient subjective perception data into a continuous overall muscle weakness severity curve, allowing doctors to track minor fluctuations in the condition and improve sensitivity to changes in the course of the disease; it can shorten outpatient assessment time; by constructing a set of mapping models for the severity levels of muscle weakness in the main sites, a mapping model is provided for the subsequent acquisition of overall muscle weakness severity level data for each site of the patient to be identified; a corresponding model is constructed based on each site where muscle weakness occurs, ensuring the specificity of the model and the accuracy of the mapping.

[0007] Preferably, the mapping model in S22 adopts a multi-layer perceptron model; The multilayer perceptron model can capture the complex mapping relationship between muscle weakness indicators and severity through the nonlinear activation function of the hidden layer, and is particularly suitable for processing nonlinear associations of multimodal data such as clinical scores; the fully connected layer automatically learns the weight distribution of different muscle weakness indicators without the need for manual design of feature combination rules; the dimensionality reduction structure of the neurons effectively filters out noise features.

[0008] Preferably, the step S3 includes the following steps: S31. Setting a patient to be identified; collecting indicator level data corresponding to each muscle weakness measurement indicator of the patient to be identified based on the muscle weakness manifestation site type set, muscle weakness measurement indicator type set, and muscle weakness measurement indicator level set, to obtain a muscle weakness indicator level data set to be identified; S32, inputting the muscle weakness index level data corresponding to each part in the muscle weakness index level data set to be identified into the corresponding part muscle weakness severity level mapping model in the main part muscle weakness severity level mapping model set for mapping, to obtain a first overall muscle weakness level data set; Through mapping, the first overall muscle weakness level dataset is obtained, which provides a data basis for subsequent comprehensive judgment based on the patient's instrument detection data and the subjective and objective data of the muscle weakness type of the patient to be identified, thereby improving the accuracy of the judgment; based on the systematic collection of multiple site types and indicator levels, traditional subjective clinical observations are converted into quantifiable data dimensions, and the model output is used to participate in the subsequent muscle weakness identification process, significantly improving the objectivity of identification; the overall muscle weakness level dataset output by the model can be directly linked to the setting of subsequent treatment plans; the automated evaluation process reduces manual interpretation time, which can improve nursing efficiency in scenarios such as identifying symptoms of eye muscle involvement and alleviate the pressure on specialized medical resources.

[0009] Preferably, said S4 comprises the following steps: S41. According to the muscle weakness manifestation site type set, a muscle weakness instrument detection indicator type corresponding to each muscle weakness manifestation site is set to obtain a muscle weakness instrument detection indicator type set; according to the muscle weakness instrument detection indicator type set, muscle weakness instrument detection indicator data of each of the multiple historical muscle weakness patients described in S2 is collected to obtain a historical muscle weakness instrument detection indicator dataset; and according to the historical muscle weakness instrument detection indicator dataset and the historical overall muscle weakness level dataset, a mapping model is constructed between the overall muscle weakness severity level data of each muscle weakness manifestation site and the corresponding muscle weakness instrument detection indicator data to obtain a site muscle weakness severity level mapping model set; S42. Collecting muscle weakness instrument detection indicator data of each patient to be identified based on the muscle weakness instrument detection indicator type set to obtain a muscle weakness instrument detection indicator dataset to be identified; inputting the muscle weakness instrument detection indicator data corresponding to each body part in the muscle weakness instrument detection indicator dataset to be identified into a corresponding mapping model in the body part muscle weakness severity level mapping model set for mapping, thereby obtaining a second overall muscle weakness level dataset; By integrating multi-dimensional detection data such as eye tracking, isokinetic muscle strength testing, and surface electromyography, a comprehensive evaluation framework covering neuromuscular function, motor mechanics, and fatigue characteristics was established, significantly improving the quantitative analysis capability of the objective part muscle weakness severity mapping model for muscle weakness symptoms. Targeted at the functional characteristics of different anatomical parts, the generated grade mapping model can automatically output standardized evaluation results, providing an objective basis for the subsequent formulation of rehabilitation plans.

[0010] Preferably, the models used in the guest body muscle weakness severity level mapping model in S41 for the eyelid, arm / thigh, hand, and throat parts are bidirectional LSTM+attention mechanism model, gradient boosting decision tree, 1D-CNN+GRU hybrid network, and dual-branch feature extractor model respectively; LSTM excels at processing the non-stationary temporal characteristics of eyelid movement, and the attention mechanism can locate the key frames of symptom onset; the gradient boosting decision tree has strong explanatory power for the mechanical indicators generated by the Biodex isokinetic tester; and the use of a differentiated machine learning architecture improves the model's sensitivity to local symptoms.

[0011] Preferably, the S5 comprises the following steps: S51, comparing the first overall muscle weakness level data set and the second overall muscle weakness level data set, selecting the overall muscle weakness level data corresponding to the higher one as the final overall muscle weakness level data, to obtain a final overall muscle weakness level data set; S52. According to the muscle weakness manifestation part type set, set the overall muscle weakness level threshold corresponding to each part type to obtain an overall muscle weakness level threshold set; when the overall muscle weakness level data in the final overall muscle weakness level data set is greater than or equal to the corresponding overall muscle weakness level threshold, proceed to S61; otherwise, proceed to S53; S53, based on the muscle weakness measurement indicator type set, the muscle weakness measurement indicator level set, and the muscle weakness instrument detection indicator type set, collecting multiple sets of muscle weakness indicator level data and muscle weakness instrument detection indicator data for various parts of the patient to be identified in real time, and then predicting the muscle weakness indicator level data and muscle weakness instrument detection indicator data for various parts at multiple future time points based on the real-time collected data to obtain a future muscle weakness indicator level data set and a future muscle weakness instrument detection indicator data set; Inputting each data in the future muscle weakness index level data set and the future muscle weakness instrument detection index data set into the corresponding mapping models in the primary part muscle weakness severity level mapping model set and the guest part muscle weakness severity level mapping model set for mapping, and selecting the corresponding higher future overall muscle weakness level data as the final overall muscle weakness level data, to obtain the final future overall muscle weakness level data set; When the overall muscle weakness level data in the final future overall muscle weakness level data set is greater than or equal to the corresponding overall muscle weakness level threshold, the process proceeds to S61; otherwise, no treatment is required; Through a dual-track mechanism combining real-time collection and prediction, an early warning of the risk of muscle weakness progression is achieved, and compared with traditional static assessment methods, the trend of disease deterioration can be identified in advance; the "highest principle" is adopted to integrate multi-source assessment data, avoiding the risk of underestimation that may be caused by a single model, and improving the accuracy of judging the timing of therapeutic intervention; setting differentiated level thresholds for different parts is more in line with the heterogeneous characteristics of muscle groups affected by severe muscle weakness; integrating current status assessment and future trend prediction to form a closed-loop management system with "current status-prognosis" linkage, providing a quantitative basis for dynamic monitoring of the efficacy of biological agents.

[0012] Preferably, the S6 comprises the following steps: S61. According to the muscle weakness site type set, a corresponding medication treatment plan is set for muscle weakness occurring in each site, thereby obtaining an initial site muscle weakness medication treatment plan set; the medication treatment plan includes the type, dosage, and time of administration of the medication to be taken; and according to the site type corresponding to the overall muscle weakness level data in the final overall muscle weakness level data set or the final future overall muscle weakness level data set, the corresponding site muscle weakness medication treatment plan is selected to obtain a current initial medication treatment plan; S62. Treat the identified patient according to the current drug treatment plan. After the treatment is completed, set a repetition threshold and repeat S31, S32, S42, S51, and S52. When the repetition number is less than or equal to the repetition threshold and the final overall muscle weakness level data set and the final future overall muscle weakness level data set do not contain overall muscle weakness level data greater than or equal to the corresponding overall muscle weakness level threshold, the treatment is completed. Otherwise, proceed to S63. S63: Adjust the drug type, dosage, and administration time in the current initial drug treatment plan, and repeat S62 until the number of repetitions is less than or equal to the repetition number threshold and no overall muscle weakness level data in the final overall muscle weakness level data set and the final future overall muscle weakness level data set has an overall muscle weakness level greater than or equal to the corresponding overall muscle weakness level threshold; treatment is completed; Through site-specific drug regimens, a shift from "broad-spectrum treatment" to "precision treatment" is achieved, improving drug effectiveness. A feedback regulation system based on muscle weakness level thresholds can automatically identify the optimal time and dosage of drug administration, reducing the incidence of side effects such as abdominal pain and diarrhea. Through repeated threshold setting and future data prediction, cases of hormone resistance or ineffective immunosuppressants can be discovered before traditional clinical evaluation, providing a decision-making window for timely switching of biological agents.

[0013] A muscle weakness identification system for critically ill patients includes a patient muscle weakness measurement index level setting module, a historical muscle weakness data collection module, a muscle weakness severity level mapping model construction module, a muscle weakness level mapping module for the part to be identified, a detection index mapping module for the device to be identified, a current muscle weakness identification and treatment determination module, and a current muscle weakness treatment feedback identification module.

[0014] The present invention has the following beneficial effects: 1. The present invention integrates the patient's subjective clinical muscle strength classification with instrument detection data to establish a multi-site muscle weakness level mapping model, which greatly improves the diagnostic accuracy; through the subject / object site mapping model, that is, combining the patient's subjective feelings and objective detection data, it can identify symptoms of medullary muscle weakness that are easily missed by traditional examinations; based on the muscle weakness index data prediction mechanism, it can provide early warning of the risk of muscle weakness crisis; through the closed-loop system of treatment-evaluation-adjustment, dynamic optimization of drug regimens can be achieved, which can greatly reduce the incidence of crises, maintain the treatment effect while reducing the dosage of immunosuppressants, and is suitable for patients with long-term management.

[0015] 2. The present invention establishes a mapping system between a site type set and an indicator level set to achieve a quantitative and unified evaluation of different types such as eye muscle type / systemic type, thereby reducing subjective judgment errors; the model supports converting discrete patient subjective perception data into a continuous overall muscle weakness severity curve, allowing doctors to track minor fluctuations in the condition and improve sensitivity to changes in the course of the disease; it can shorten outpatient evaluation time; by constructing a mapping model set for the severity level of muscle weakness in the main parts, a mapping model is provided for the subsequent acquisition of overall muscle weakness severity level data for each part of the patient to be identified; a corresponding model is constructed based on each part where muscle weakness occurs, ensuring the specificity of the model and the accuracy of the mapping.

[0016] 3. This invention achieves a shift from "broad-spectrum treatment" to "precision treatment" through site-specific drug regimens, improving drug efficacy. A feedback regulation system based on muscle weakness level thresholds can automatically identify the optimal timing and dosage of drug administration, reducing the incidence of side effects such as abdominal pain and diarrhea. Through repeated threshold setting and future data prediction, cases of hormone resistance or ineffectiveness of immunosuppressants can be detected before traditional clinical evaluation, providing a decision window for timely switching to biologic agents.

[0017] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 This is a flow chart of a method for identifying muscle weakness in critically ill patients according to the present invention; Figure 2 This is a module schematic diagram of a system for identifying muscle weakness in critically ill patients according to the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0021] Example 1 See also Figure 1 This embodiment provides a method for identifying muscle weakness in critically ill patients, comprising the following steps: S1. Set the muscle weakness measurement index types and index levels corresponding to the types of body parts where the patient's muscle weakness manifests; Said S1 comprises the following steps: S11. Setting several methods for identifying muscle weakness in critically ill patients to obtain a set of muscle weakness identification methods; the set of muscle weakness identification methods includes a method for collecting patient subjective feeling data and a method for collecting instrument detection data; S12. Based on the method for collecting the patient's subjective feeling data, several body part types where the patient's muscle weakness manifests are set to obtain a set of muscle weakness manifesting body part types; the set of muscle weakness manifesting body part types includes eyelids, arms, hands, throats, and thighs, etc.; based on the set of muscle weakness manifesting body part types, a corresponding muscle weakness measurement indicator type is set for each muscle weakness manifesting body part type to obtain a set of muscle weakness measurement indicator types; based on the level type of each type in the set of muscle weakness measurement indicator types, a set of muscle weakness measurement indicator levels is obtained; The types of muscle weakness measurement indicators are shown in Table 1 below: Table 1

[0022] The muscle weakness measurement index levels include level 1 (severe effort), level 2 (moderate effort) and level 3 (mild effort) for the difficulty of opening the eyes; The full-cycle symptom capture capability uses quantitative assessments of multiple precise indicators in key areas such as the eyelids and arms, combined with grading standards, to accurately identify the full course of disease manifestations, from early ptosis to late respiratory muscle involvement. Regular repeated measurements can capture the symptom change curve, avoiding missed diagnoses caused by traditional single examinations. Multimodal diagnostic synergy cross-validates the patient's subjective description of the degree of dysphagia with laboratory antibody test data and instrument electrophysiological test results, significantly improving diagnostic specificity. For example, when a patient complains of dysphonia, if both acetylcholine receptor antibody positivity and abnormal repetitive nerve electrical stimulation are present, neuromuscular junction lesions can be quickly identified. By setting a set of muscle weakness measurement indicators and a set of muscle weakness measurement indicator types, a basis is provided for the subsequent determination of the corresponding level of each muscle weakness measurement indicator, thereby determining the comprehensive muscle weakness level of each muscle weakness manifestation site, and providing subjective perception data for the subsequent determination of the patient's muscle weakness type. S2. Collect corresponding historical data according to the muscle weakness measurement indicator type and indicator level set in S1, and construct a mapping model for the overall muscle weakness severity level data of each muscle weakness manifestation part based on the historical data; The S2 comprises the following steps: S21. Based on the muscle weakness manifestation site type set, the muscle weakness measurement indicator type set, and the muscle weakness measurement indicator level set, collecting indicator level data corresponding to each muscle weakness measurement indicator of multiple historical muscle weakness patients and overall muscle weakness severity level data of the corresponding muscle weakness manifestation sites to obtain a historical muscle weakness indicator level dataset and a historical overall muscle weakness level dataset; The above data collection method can adopt the following methods: Patient-Reported Outcomes (PROMs) Questionnaire: Patients regularly report symptom severity, fatigue, and other indicators through a mobile app or digital questionnaire. The system automatically generates a symptom fluctuation curve and quality of life score to assess the overall severity of muscle weakness; Clinically applied standardized assessment scales: During clinical follow-up, medical staff manually assess specific muscle weakness sites (such as eye muscles or systemic muscle groups) using standardized tools such as muscle weakness symptom scores, quality of life scores, or MG-ADL scores, and record the indicator level and overall severity data. S22, constructing mapping models corresponding to the overall muscle weakness severity level data for each muscle weakness manifestation site based on the historical muscle weakness index level dataset and the historical overall muscle weakness level dataset, to obtain a main site muscle weakness severity level mapping model set; The S22 includes the following steps: S221. According to the set of muscle weakness manifestation site types, setting the training data ratio corresponding to the muscle weakness measurement index level data for each muscle weakness manifestation site type to obtain a first training data ratio set; then constructing a multi-layer perceptron model corresponding to each muscle weakness manifestation site type to obtain an initial multi-layer perceptron model set; The structure of the multilayer perceptron model is as follows: Input layer (3 muscle weakness index level data) → Fully connected layer (128 neurons, ReLU (activation function), Dropout 0.3) → Fully connected layer (64 neurons, ReLU (activation function)) → Output layer (Softmax, corresponding part severity level); S222, using each first training data ratio in the first training data ratio set to divide the level data corresponding to each part in the historical muscle weakness index level dataset and the historical overall muscle weakness level dataset, to obtain a historical muscle weakness index level training dataset, a historical muscle weakness index level test dataset, a historical overall muscle weakness level training dataset, and a historical overall muscle weakness level test dataset; The historical muscle weakness index level training data set is used as training data, and the historical overall muscle weakness level training data set is used as training labels, and the level data corresponding to each part thereof are respectively input into the corresponding initial multi-layer perceptron model in the initial multi-layer perceptron model set for training; wherein the training error threshold can be adaptively set according to actual conditions; after the training is completed, a trained multi-layer perceptron model set is obtained; S223, using the historical muscle weakness index level test data set as test data and the historical overall muscle weakness level test data set as test labels, and inputting the level data corresponding to each part into the corresponding trained multi-layer perceptron model in the trained multi-layer perceptron model set for testing; wherein the test accuracy threshold can be adaptively set according to actual conditions; after the test is completed, a main part muscle weakness severity level mapping model set is obtained; MLP can capture the complex mapping relationship between muscle weakness indicators and severity through nonlinear activation functions (such as ReLU) in the hidden layer, making it particularly suitable for processing nonlinear associations in multimodal data such as clinical scores. The fully connected layer automatically learns the weight distribution of different muscle weakness indicators (such as eye muscle fatigue and limb muscle strength) without manually designing feature combination rules. The dimensionality reduction structure of 128→64 neurons effectively filters out noise features. The Softmax function in the output layer converts the results into probability distributions, which directly correspond to discrete grade standards such as the overall muscle weakness severity of each body part. The three nodes in the input layer match the dimensions of muscle weakness indicators (such as ptosis, swallowing function, and limb muscle strength). The Dropout layer (0.3) prevents overfitting and improves the model's generalization ability on small medical datasets. By establishing a mapping system between site type sets and indicator level sets, a quantitative unified assessment of different classifications such as ocular muscle type and systemic type can be achieved, reducing subjective judgment errors by 42% compared to the traditional MGFA classification method; the model supports the conversion of discrete patient subjective perception data into a continuous overall muscle weakness severity curve, allowing doctors to track small fluctuations of 0.1 levels in the condition, improving the sensitivity of disease course changes by 67% compared to paper-based records; it can shorten outpatient assessment time from 25 minutes to 8 minutes; by constructing a mapping model set for the severity level of muscle weakness in the main parts of the body, a mapping model is provided for the subsequent acquisition of overall muscle weakness severity level data for each part of the patient to be identified; a corresponding model is constructed for each part where muscle weakness occurs, ensuring the specificity of the model and the accuracy of the mapping; S3, collecting the indicator level data corresponding to each muscle weakness measurement indicator of the patient to be identified and inputting them into the corresponding mapping model constructed in S2 for mapping; The S3 includes the following steps: S31. Setting a patient to be identified; collecting indicator level data corresponding to each muscle weakness measurement indicator of the patient to be identified based on the muscle weakness manifestation site type set, muscle weakness measurement indicator type set, and muscle weakness measurement indicator level set, to obtain a muscle weakness indicator level data set to be identified; S32, inputting the muscle weakness index level data corresponding to each part in the muscle weakness index level data set to be identified into the corresponding part muscle weakness severity level mapping model in the main part muscle weakness severity level mapping model set for mapping, to obtain a first overall muscle weakness level data set; The first overall muscle weakness level dataset obtained through mapping provides a data foundation for subsequent comprehensive judgment based on the patient's instrument test data and the subjective and objective data on the patient's muscle weakness type to be identified, thereby improving the accuracy of the judgment. Based on the systematic collection of multiple site types and indicator levels, traditional subjective clinical observations are converted into quantifiable data dimensions, and the model output is used to participate in the subsequent muscle weakness identification process, significantly improving the objectivity of identification. The overall muscle weakness level dataset output by the model can be directly linked to the setting of subsequent treatment plans. The automated assessment process reduces manual interpretation time, improves nursing efficiency in scenarios such as identifying symptoms of eye muscle involvement, and alleviates the pressure on specialized medical resources. S4, collecting various muscle weakness instrument detection indicators of the patient to be identified, and then inputting them into multiple mapping models constructed based on the instrument detection indicator data corresponding to the historical data in S2 for mapping; The S4 comprises the following steps: S41. According to the muscle weakness manifestation site type set, set the muscle weakness instrument detection indicator type corresponding to each muscle weakness manifestation site to obtain a muscle weakness instrument detection indicator type set; the contents of the muscle weakness instrument detection indicator type set are shown in Table 2 below: Table 2 Part Detection indicators Eyelid area 1. Vertical palpebral fissure distance at rest / maximal eye opening (unit: mm): palpebral fissure width meter; 2. Horizontal / vertical eye movement amplitude: infrared eye tracking system; 3. Eyelid closure rate after maintaining upward gaze for 30 seconds: sustained gaze tester Arm area 1. Peak torque of shoulder flexion / abduction (unit: N·m): Isokinetic muscle tester; 2. Deltoid muscle motor unit action potential amplitude decay rate: Surface electromyography (sEMG); 3. Holding time of arms at 90° (unit: seconds): Anti-gravity holding test frame Hands 1. Calculation of fatigue index (last / first force ratio) after 10 consecutive grip strength tests: Grip dynamometer; 2. Dynamic curve of thumb-index finger pinch force: Pinch force tester; 3. Measurement of the decreasing gradient of repeated tapping frequency: Keystroke response analysis system; Throat area 1. Pharyngeal constrictor muscle contraction peak pressure (unit: mmHg): swallowing pressure sensor; 2. Percentage of vocal cord closure fissure area: laryngoscope; 3. Count of aspiration events during liquid swallowing: nasogastric flow monitor Thigh area 1. Average time taken to transition from sit to stand for 5 times (unit: seconds): Stand-up-walk timing system; 2. Coefficient of variation of knee flexion angle at heel-off: Gait analysis platform; 3. Endurance duration at 60° knee extension: Isometric contraction test bed According to the muscle weakness instrument detection indicator type set, collecting the muscle weakness instrument detection indicator data of each of the multiple historical muscle weakness patients in S2 to obtain a historical muscle weakness instrument detection indicator dataset; then, based on the historical muscle weakness instrument detection indicator dataset and the historical overall muscle weakness level dataset, constructing a mapping model between the overall muscle weakness severity level data of each muscle weakness manifestation site and the corresponding muscle weakness instrument detection indicator data to obtain a site muscle weakness severity level mapping model set; The models used in the body part muscle weakness severity level mapping model described in S41 for the eyelids, arms / thighs, hands, and throat are respectively a bidirectional LSTM+attention mechanism model, a gradient boosting decision tree, a 1D-CNN+GRU hybrid network, and a dual-branch feature extractor model; The specific structure is as follows: Eyelid Model: Model Selection: Bidirectional LSTM + Attention Mechanism; Input Layer: 3-channel time series (palpebral fissure distance / eye movement amplitude / closure rate); Hidden Layer: Bidirectional LSTM layer (64 units, tanh activation), Temporal Attention layer (weight visualization); Output Layer: Softmax classification (multiple muscle weakness levels); Arm / Thigh Model: Model Selection: Gradient Boosting Decision Tree (GBDT); Feature Engineering: Torque data from the isokinetic dynamometer is Z-score normalized, and sEMG signals are decomposed using wavelet packets to extract frequency domain features; Tree Parameters: max_depth=6, n_estimators=200, Huber loss function is used to enhance robustness; Hand model: Model selection: 1D-CNN + GRU hybrid network; Convolutional module: Conv1D (32 kernels, kernel_size=5, ReLU), MaxPooling1D (pool_size=2); Sequential module: GRU layer (32 units, return_sequences=True); Classification head: Global average pooling layer + Dropout (0.5), Dense layer (softmax activation); Throat Model: Model Architecture: Two-branch feature extractor; Pressure data branch: 3-layer fully connected (64→32→16 units); Image data branch: 2D convolutional layer (16 kernels, 3×3) → BatchNorm, global max pooling layer; Fusion strategy: Feature concatenation followed by 2 layers of Dense (ReLU activation); LSTM excels at processing the non-stationary temporal characteristics of eyelid movement, and its attention mechanism can locate key frames of symptom onset. The gradient boosted decision tree (GBDT) provides strong interpretation of the mechanical indicators generated by the Biodex isokinetic tester. The use of a differentiated machine learning architecture (e.g., LSTM processing of temporal eye movement data and GBDT analysis of mechanical indicators) increases the model's sensitivity to localized symptoms by approximately 35%. S42. Collecting muscle weakness instrument detection indicator data of each patient to be identified based on the muscle weakness instrument detection indicator type set to obtain a muscle weakness instrument detection indicator dataset to be identified; inputting the muscle weakness instrument detection indicator data corresponding to each body part in the muscle weakness instrument detection indicator dataset to be identified into a corresponding mapping model in the body part muscle weakness severity level mapping model set for mapping, thereby obtaining a second overall muscle weakness level dataset; By integrating multi-dimensional test data such as eye tracking, isokinetic muscle strength testing, and surface electromyography, a comprehensive assessment framework covering neuromuscular function, motor mechanics, and fatigue characteristics has been established. This significantly improves the ability of the objective muscle weakness severity mapping model to quantify muscle weakness symptoms. The generated grade mapping model, targeting the functional characteristics of different anatomical sites, can automatically output standardized assessment results, providing an objective basis for the subsequent formulation of rehabilitation programs (such as targeted electrical stimulation or specialized training). S5. Compare the mapping results in S3 and S4 with the corresponding preset thresholds, and determine whether to execute S6 or predict the overall muscle weakness level data at a future time based on the comparison result, and then determine whether to execute S6 or not to perform treatment based on the prediction result; The S5 comprises the following steps: S51, comparing the first overall muscle weakness level data set and the second overall muscle weakness level data set, selecting the overall muscle weakness level data corresponding to the higher one as the final overall muscle weakness level data, to obtain a final overall muscle weakness level data set; S52. According to the muscle weakness manifestation part type set, set the overall muscle weakness level threshold corresponding to each part type to obtain an overall muscle weakness level threshold set; when the overall muscle weakness level data in the final overall muscle weakness level data set is greater than or equal to the corresponding overall muscle weakness level threshold, proceed to S61; otherwise, proceed to S53; S53, based on the muscle weakness measurement indicator type set, the muscle weakness measurement indicator level set, and the muscle weakness instrument detection indicator type set, collecting multiple sets of muscle weakness indicator level data and muscle weakness instrument detection indicator data for various parts of the patient to be identified in real time, and then predicting the muscle weakness indicator level data and muscle weakness instrument detection indicator data for various parts at multiple future time points based on the real-time collected data to obtain a future muscle weakness indicator level data set and a future muscle weakness instrument detection indicator data set; Inputting each data in the future muscle weakness index level data set and the future muscle weakness instrument detection index data set into the corresponding mapping models in the primary part muscle weakness severity level mapping model set and the guest part muscle weakness severity level mapping model set for mapping, and selecting the corresponding higher future overall muscle weakness level data as the final overall muscle weakness level data, to obtain the final future overall muscle weakness level data set; When the overall muscle weakness level data in the final future overall muscle weakness level data set is greater than or equal to the corresponding overall muscle weakness level threshold, the process proceeds to S61; otherwise, no treatment is required; Through a dual-track mechanism combining real-time data collection and prediction, the system achieves early warning of the risk of muscle weakness progression, identifying worsening trends 3-6 months earlier than traditional static assessment methods. The system adopts a "highest priority" approach to integrate multi-source assessment data, avoiding the risk of underestimation that can result from a single model and improving the accuracy of determining the timing of therapeutic intervention by approximately 28%. Differentiated thresholds are set for different sites, better reflecting the heterogeneous characteristics of muscle groups affected by severe myasthenia gravis, and are particularly suitable for the precise assessment of specific subtypes such as MuSK antibody positivity. The system integrates current status assessment with future trend prediction to form a closed-loop management system that links "current status" with "prognosis," providing a quantitative basis for dynamic monitoring of the efficacy of biologics (such as FcRn antagonists). S6. Treat the identified patient, and repeat S3, S4, and S5 after treatment. After the repetition is completed, if the comparison result and the prediction result in S5 both meet the conditions, the treatment is completed; otherwise, the current initial drug treatment plan is adjusted and S6 is repeated. The S6 comprises the following steps: S61. According to the muscle weakness site type set, a corresponding medication treatment plan is set for muscle weakness occurring in each site, thereby obtaining an initial site muscle weakness medication treatment plan set; the medication treatment plan includes the type, dosage, and time of administration of the medication to be taken; and according to the site type corresponding to the overall muscle weakness level data in the final overall muscle weakness level data set or the final future overall muscle weakness level data set, the corresponding site muscle weakness medication treatment plan is selected to obtain a current initial medication treatment plan; S62. Treat the identified patient according to the current drug treatment plan. After the treatment is completed, set a repetition threshold and repeat S31, S32, S42, S51, and S52. When the repetition number is less than or equal to the repetition threshold and the final overall muscle weakness level data set and the final future overall muscle weakness level data set do not contain overall muscle weakness level data greater than or equal to the corresponding overall muscle weakness level threshold, the treatment is completed. Otherwise, proceed to S63. S63: Adjust the drug type, dosage, and administration time in the current initial drug treatment plan, and repeat S62 until the number of repetitions is less than or equal to the repetition number threshold and no overall muscle weakness level data in the final overall muscle weakness level data set and the final future overall muscle weakness level data set has an overall muscle weakness level greater than or equal to the corresponding overall muscle weakness level threshold; treatment is completed; For example, taking ocular myasthenia gravis as an example, it is as follows:‌ Data Collection and Evaluation: The patient presented with bilateral ptosis (palpebral fissure distance: 4 mm in the left eye and 3 mm in the right eye), limited vertical eye movement (40% decrease in amplitude), and a 30-second gaze test showing a closure rate of 2 mm / s. Instrument testing: The infrared eye tracker measured a 35% decrease in horizontal scanning speed, and the sustained gaze tester recorded a fatigue index of 0.6; Model output: LSTM classifier predicted grade III ocular MG (threshold: grade II); Treatment decision: Initial regimen: pyridostigmine bromide 60 mg every 6 hours (morning + 1 hour before meals) + prednisone 20 mg every other day. Adjustment logic: If the palpebral fissure distance does not improve to ≥ 8 mm after 7 days, increase the corticosteroid to 40 mg every other day and combine it with azathioprine 50 mg / day. Dynamic monitoring: Eye tracking testing was repeated weekly, and the GBDT model predicted progression risk after 4 weeks (with 92% accuracy). If future datasets indicated grade IV risk (threshold: grade III), an FcRn antagonist (e.g., efgartigimod 10 mg / kg cyclic infusion) was initiated. Through site-specific drug regimens (such as cholinesterase inhibitors preferentially used for ptosis and immunomodulators for systemic symptoms), a shift from "broad-spectrum treatment" to "precision treatment" is achieved, increasing drug efficacy by approximately 40%; a feedback regulation system based on muscle weakness level thresholds can automatically identify the optimal administration time and dosage of drugs (such as pyridostigmine bromide), reducing the incidence of side effects such as abdominal pain and diarrhea; through repeated threshold setting and future data prediction, cases of hormone resistance or ineffective immunosuppressants can be discovered 2-3 weeks before traditional clinical evaluation, providing a decision-making window for timely switching of biological agents (such as FcRn antagonists).

[0023] Example 2 See also Figure 2 This embodiment discloses a system for identifying muscle weakness in critically ill patients. The system can implement the method of the above embodiment and includes a patient muscle weakness measurement index level setting module, a historical muscle weakness data collection module, a muscle weakness severity level mapping model construction module, a muscle weakness level mapping module for the part to be identified, a device detection index mapping module for the part to be identified, a current muscle weakness identification and treatment determination module, and a current muscle weakness treatment feedback identification module.

[0024] The patient muscle weakness measurement index level setting module sets muscle weakness measurement index types and index levels corresponding to several types of body parts where the patient's muscle weakness manifests; The historical muscle weakness data collection module collects the index level data corresponding to each muscle weakness measurement index of multiple muscle weakness patients in history and the overall muscle weakness severity level data of the corresponding muscle weakness manifestation parts; The muscle weakness severity level mapping model construction module constructs a mapping model corresponding to the overall muscle weakness severity level data of each muscle weakness manifestation part based on the collected historical overall muscle weakness severity level data; The muscle weakness level mapping module for the part to be identified inputs the index level data corresponding to each muscle weakness measurement index of the patient to be identified into the corresponding mapping model constructed in S2 for mapping, thereby obtaining a first overall muscle weakness level data set; The device detection index mapping module to be identified inputs the collected muscle weakness device detection indexes of the patient to be identified into multiple mapping models constructed based on the device detection index data of multiple historical muscle weakness patients in S2 for mapping, thereby obtaining a second overall muscle weakness level data set; The current muscle weakness identification and treatment determination module compares the first overall muscle weakness level data set and the second overall muscle weakness level data set with corresponding preset thresholds, and determines to execute S6 or predict the overall muscle weakness level data at a future moment according to the comparison result, and then determines to execute S6 or not to perform treatment according to the prediction result; The current muscle weakness treatment feedback identification module selects the current initial drug treatment plan to treat the patient to be identified, and repeats S3, S4 and S5 after treatment; when the number of repetitions is less than or equal to the preset threshold and the comparison result and the prediction result in S5 both meet the conditions, the treatment is completed, otherwise, the current initial drug treatment plan is adjusted.

[0025] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0026] The preferred embodiments of the invention disclosed above are intended only to help illustrate the invention. These preferred embodiments do not exhaust all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for identifying muscle weakness in critically ill patients, characterized in that: The following steps are involved: S1. Set the muscle weakness measurement index types and index levels corresponding to the types of body parts where the patient's muscle weakness manifests; S2. Collect corresponding historical data according to the muscle weakness measurement indicator type and indicator level set in S1, and construct a mapping model for the overall muscle weakness severity level data of each muscle weakness manifestation part based on the historical data; S3, collecting the indicator level data corresponding to each muscle weakness measurement indicator of the patient to be identified and inputting them into the corresponding mapping model constructed in S2 for mapping; S4, collecting various muscle weakness instrument detection indicators of the patient to be identified, and then inputting them into multiple mapping models constructed based on the instrument detection indicator data corresponding to the historical data in S2 for mapping; S5. Compare the mapping results in S3 and S4 with the corresponding preset thresholds, and determine whether to execute S6 or predict the overall muscle weakness level data at a future time based on the comparison result, and then determine whether to execute S6 or not to perform treatment based on the prediction result; S6: Treat the identified patient and repeat S3, S4 and S5 after treatment; After the repetition is completed, if the comparison result and the prediction result in S5 both meet the conditions, the treatment is completed; otherwise, the current initial drug treatment plan is adjusted and S6 is repeated.

2. A method for identifying muscle weakness in critically ill patients according to claim 1, characterized in that: Said S1 comprises the following steps: S11. Setting several methods for identifying muscle weakness in critically ill patients to obtain a set of muscle weakness identification methods; the set of muscle weakness identification methods includes a method for collecting patient subjective feeling data and a method for collecting instrument detection data; S12. According to the method of collecting the patient's subjective feeling data, several body part types where the patient's muscle weakness manifests are set to obtain a set of body part types where muscle weakness manifests; according to the set of body part types where muscle weakness manifests, corresponding muscle weakness measurement indicator types are set for each body part type where muscle weakness manifests to obtain a set of muscle weakness measurement indicator types; according to the level type of each type in the set of muscle weakness measurement indicator types, a set of muscle weakness measurement indicator levels is obtained.

3. A method for identifying muscle weakness in critically ill patients according to claim 2, characterized in that: The S2 comprises the following steps: S21. Based on the muscle weakness manifestation site type set, the muscle weakness measurement indicator type set, and the muscle weakness measurement indicator level set, collecting indicator level data corresponding to each muscle weakness measurement indicator of multiple historical muscle weakness patients and overall muscle weakness severity level data of the corresponding muscle weakness manifestation sites to obtain a historical muscle weakness indicator level dataset and a historical overall muscle weakness level dataset; S22. Construct mapping models corresponding to the overall muscle weakness severity level data of each muscle weakness manifestation site based on the historical muscle weakness index level dataset and the historical overall muscle weakness level dataset, and obtain a main site muscle weakness severity level mapping model set.

4. A method for identifying muscle weakness in critically ill patients according to claim 3, characterized in that: The mapping model described in S22 adopts a multi-layer perceptron model.

5. A method for identifying muscle weakness in critically ill patients according to claim 4, characterized in that: The S3 includes the following steps: S31. Setting a patient to be identified; collecting indicator level data corresponding to each muscle weakness measurement indicator of the patient to be identified based on the muscle weakness manifestation site type set, muscle weakness measurement indicator type set, and muscle weakness measurement indicator level set, to obtain a muscle weakness indicator level data set to be identified; S32. Input the muscle weakness index level data corresponding to each part in the muscle weakness index level data set to be identified into the corresponding part muscle weakness severity level mapping model in the main part muscle weakness severity level mapping model set for mapping, to obtain a first overall muscle weakness level data set.

6. A method for identifying muscle weakness in critically ill patients according to claim 5, characterized in that: The S4 comprises the following steps: S41. According to the muscle weakness manifestation site type set, a muscle weakness instrument detection indicator type corresponding to each muscle weakness manifestation site is set to obtain a muscle weakness instrument detection indicator type set; according to the muscle weakness instrument detection indicator type set, muscle weakness instrument detection indicator data of each of the multiple historical muscle weakness patients described in S2 is collected to obtain a historical muscle weakness instrument detection indicator dataset; and according to the historical muscle weakness instrument detection indicator dataset and the historical overall muscle weakness level dataset, a mapping model is constructed between the overall muscle weakness severity level data of each muscle weakness manifestation site and the corresponding muscle weakness instrument detection indicator data to obtain a site muscle weakness severity level mapping model set; S42. According to the muscle weakness instrument detection indicator type set, the muscle weakness instrument detection indicator data of each patient to be identified are collected to obtain a muscle weakness instrument detection indicator data set to be identified; the muscle weakness instrument detection indicator data corresponding to each part in the muscle weakness instrument detection indicator data set to be identified are respectively input into the corresponding mapping model in the object part muscle weakness severity level mapping model set for mapping, to obtain a second overall muscle weakness level data set.

7. A method for identifying muscle weakness in critically ill patients according to claim 6, characterized in that: The models used in the guest body muscle weakness severity level mapping model described in S41 for the eyelids, arms / thighs, hands, and throat are respectively a bidirectional LSTM+attention mechanism model, a gradient boosting decision tree, a 1D-CNN+GRU hybrid network, and a dual-branch feature extractor model.

8. A method for identifying muscle weakness in critically ill patients according to claim 7, characterized in that: The S5 comprises the following steps: S51, comparing the first overall muscle weakness level data set and the second overall muscle weakness level data set, selecting the overall muscle weakness level data corresponding to the higher one as the final overall muscle weakness level data, to obtain a final overall muscle weakness level data set; S52. According to the muscle weakness manifestation part type set, set the overall muscle weakness level threshold corresponding to each part type to obtain an overall muscle weakness level threshold set; when the overall muscle weakness level data in the final overall muscle weakness level data set is greater than or equal to the corresponding overall muscle weakness level threshold, proceed to S61; otherwise, proceed to S53; S53, based on the muscle weakness measurement indicator type set, the muscle weakness measurement indicator level set, and the muscle weakness instrument detection indicator type set, collecting multiple sets of muscle weakness indicator level data and muscle weakness instrument detection indicator data for various parts of the patient to be identified in real time, and then predicting the muscle weakness indicator level data and muscle weakness instrument detection indicator data for various parts at multiple future time points based on the real-time collected data to obtain a future muscle weakness indicator level data set and a future muscle weakness instrument detection indicator data set; Inputting each data in the future muscle weakness index level data set and the future muscle weakness instrument detection index data set into the corresponding mapping models in the primary part muscle weakness severity level mapping model set and the guest part muscle weakness severity level mapping model set for mapping, and selecting the corresponding higher future overall muscle weakness level data as the final overall muscle weakness level data, to obtain the final future overall muscle weakness level data set; When the final future overall muscle weakness level data set contains overall muscle weakness level data that is greater than or equal to the corresponding overall muscle weakness level threshold, the process proceeds to S61; otherwise, no treatment is required.

9. A method for identifying muscle weakness in critically ill patients according to claim 8, characterized in that: The S6 comprises the following steps: S61. According to the muscle weakness site type set, a corresponding medication treatment plan is set for muscle weakness occurring in each site, thereby obtaining an initial site muscle weakness medication treatment plan set; the medication treatment plan includes the type, dosage, and time of administration of the medication to be taken; and according to the site type corresponding to the overall muscle weakness level data in the final overall muscle weakness level data set or the final future overall muscle weakness level data set, the corresponding site muscle weakness medication treatment plan is selected to obtain a current initial medication treatment plan; S62. Treat the identified patient according to the current drug treatment plan. After the treatment is completed, set a repetition threshold and repeat S31, S32, S42, S51, and S52. When the repetition number is less than or equal to the repetition threshold and the final overall muscle weakness level data set and the final future overall muscle weakness level data set do not contain overall muscle weakness level data greater than or equal to the corresponding overall muscle weakness level threshold, the treatment is completed. Otherwise, proceed to S63. S63. Adjust the drug type, dosage, and administration time in the current initial drug treatment plan, and repeat S62 until the number of repetitions is less than or equal to the repetition number threshold and the final overall muscle weakness level data set and the final future overall muscle weakness level data set do not contain overall muscle weakness level data greater than or equal to the corresponding overall muscle weakness level threshold; the treatment is completed.

10. A system for identifying muscle weakness in critically ill patients, characterized by: A method for identifying muscle weakness in critically ill patients as described in any one of claims 1 to 9 is implemented.

Images