Lymphedema limb volume measurement data management system and method

Through AI technology, the feature extraction and fusion of lymphedema limb volume and treatment data is solved, and the problem of insufficient evaluation complexity and accuracy in the existing technology is solved, automated abnormal detection and early warning is realized, and the accuracy and objectiveness of lymphedema evaluation are improved. sex.

CN118658633BActive Publication Date: 2024-11-15JILIN UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411135190.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-15
Estimated Expiration
2044-08-19

AI Technical Summary

Technical Problem

The existing lymphedema evaluation technology has wasted resources, complex operation, insufficient accuracy, potential infection risk, high cost, and reliance on doctors to make subjective judgments, resulting in inaccurate assessment and biased treatment decisions.

Method used

Using AI-based data analysis and processing technology, local timing feature extraction, semantic reasoning and semantic fusion are carried out by obtaining lymphedema limb volume data and treatment data, and abnormal detection and early warning are achieved using a two-way attention network.

Benefits of technology

Automatic detection and early warning of lymphedema limb volume abnormalities has been realized, reducing subjective misjudgment from doctors and improving the objectivity and accuracy of the evaluation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118658633B_ABST
    Figure CN118658633B_ABST
Patent Text Reader

Abstract

The present application provides a lymphedema limb volume measurement data management system and method, which relates to the field of intelligent data analysis. It uses AI-based data analysis and processing technology to extract local temporal features, semantic reasoning and semantic fusion of the lymphedema limb volume data and treatment data of the monitored lymphedema limb patients, so as to automatically judge whether the volume of the patient's lymphedema limb is abnormal according to the semantic interaction fusion features between the limb volume data and the treatment data, and generate corresponding warnings. In this way, the automated limb volume abnormality detection and warning functions can be realized, the misjudgment caused by the doctor's subjective experience can be reduced, and the objectivity and accuracy of the lymphedema limb volume assessment can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent data analysis, and more specifically, to a lymphedema limb volume measurement data management system and method. Background Art

[0002] Lymphedema is a chronic and difficult-to-treat disease with a long and complicated treatment cycle. During this process, regular assessment of limb volume is essential to guide treatment.

[0003] Existing assessment techniques, such as water displacement, waste resources, are cumbersome to operate, lack precision, and have potential infection risks; while lymphoscintigraphy may cause allergic reactions due to the use of contrast agents and is costly. In addition, although magnetic resonance imaging (MRI) and computed tomography (CT) provide detailed images, they are not only expensive, but also not suitable for patients with pacemakers, cochlear implants, or other metal implants. In addition, the current evaluation of the treatment effect of lymphedema often relies on the subjective judgment of doctors, which may be biased due to differences in personal experience, mood swings, or fatigue, affecting the accurate assessment of the condition and treatment decisions. In addition, due to the lack of quantitative assessment tools, doctors may find it difficult to capture subtle changes in the condition, leading to misjudgment of the treatment effect.

[0004] Therefore, an optimized solution for data management of lymphedema limb volumetry is needed. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present application provides a lymphedema limb volume measurement data management system and method.

[0006] According to one aspect of the present application, a lymphedema limb volume measurement data management system is provided, comprising:

[0007] A patient lymphedema limb volume data acquisition module, used to acquire a time series of lymphedema limb volume data of a monitored lymphedema limb patient;

[0008] A patient lymphedema limb treatment data acquisition module, used to acquire treatment data of the monitored lymphedema limb patient, the treatment data including physical therapy data and drug therapy data;

[0009] A limb volume time series feature extraction module, used for arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting the vector into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors;

[0010] A limb volume temporal feature semantic reasoning module, used for passing the sequence of the limb volume local temporal pattern feature vectors through a limb volume temporal node semantic propagation reasoner based on a node energy decay mechanism to obtain a limb volume temporal semantic propagation reasoning feature vector;

[0011] a treatment data multimodal fusion module, configured to perform semantic encoding on the treatment data of the monitored lymphedema limb patient to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector, and then fuse the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector to obtain a treatment multimodal semantic fusion feature vector;

[0012] A treatment multimodal-limb volume semantic feature fusion module, used for fusing the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature;

[0013] An abnormal detection result generation module is used to dynamically select interactive fusion features based on the treatment multimodal semantics-limb volume temporal reasoning semantics to obtain abnormal detection results, and based on the abnormal detection results, generate corresponding early warning prompt signals to medical staff;

[0014] Among them, the limb volume time series feature extraction module includes: a lymphedema limb volume data structuring unit, which is used to arrange the time series of the lymphedema limb volume data according to the time dimension into the lymphedema limb volume time series input vector; a lymphedema limb volume local time series feature generation unit, which is used to input the lymphedema limb volume time series input vector into a lymphedema limb volume time series pattern feature extractor based on a 1D convolutional neural network model to obtain a sequence of the limb volume local time series pattern feature vectors.

[0015] In the above-mentioned lymphedema limb volume measurement data management system, the limb volume temporal feature semantic reasoning module includes: a limb volume local temporal pattern feature energy statistical paradigm value calculation unit, which is used to calculate the energy statistical paradigm value of each limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vectors based on the maximum value, average value and variance of each limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vectors to obtain a sequence of limb volume local temporal pattern energy statistical paradigm values, wherein the energy statistical paradigm value corresponding to the current limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern energy statistical paradigm value is calculated. The energy statistical paradigm value is used as the current node energy statistical paradigm value and other energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of the current limb volume local temporal pattern energy statistical paradigm value and the historical limb volume local temporal pattern energy statistical paradigm value; a limb volume local temporal pattern node propagation space span value calculation unit is used to count the node propagation space span values ​​between each other limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vector and the current limb volume local temporal pattern feature vector to obtain a sequence of limb volume local temporal pattern propagation space span values; the limb volume local temporal pattern energy a propagation attenuation coefficient value calculation unit, for determining the energy propagation attenuation coefficient values ​​of other limb volume local time series pattern feature vectors in the sequence of limb volume local time series pattern feature vectors based on the sequence of propagation space span values ​​of the limb volume local time series pattern and the sequence of energy statistical paradigm values ​​of the historical limb volume local time series pattern, so as to obtain a sequence of limb volume local time series pattern energy propagation attenuation coefficient values, wherein the limb volume local time series pattern energy propagation attenuation coefficient values ​​are inversely correlated with the historical limb volume local time series pattern energy statistical paradigm values; and a limb volume local time series pattern feature weighting unit, for determining the energy propagation attenuation coefficient values ​​of other limb volume local time series pattern feature vectors in the sequence of limb volume local time series pattern feature vectors, so as to obtain a sequence of limb volume local time series pattern energy propagation attenuation coefficient values. A sequence of local temporal pattern energy propagation attenuation coefficient values ​​is used as a weight sequence, and the weighted sum of the sequences of all other limb volume local temporal pattern feature vectors in the sequence of the limb volume local temporal pattern feature vector is calculated to obtain a historical limb volume local temporal pattern energy attenuation temporal aggregation feature vector; a limb volume local temporal pattern propagation reasoning unit is used to fuse the current limb volume local temporal pattern energy statistical paradigm value to calculate the weighted sum of the historical limb volume local temporal pattern energy attenuation temporal aggregation feature vector and the current limb volume local temporal pattern feature vector to obtain the limb volume temporal semantic propagation reasoning feature vector.

[0016] In the above-mentioned lymphedema limb volume measurement data management system, the limb volume local time series pattern feature energy statistical paradigm value calculation unit is used to: extract the maximum value of the limb volume local time series pattern feature vector to obtain the maximum value of the limb volume local time series pattern feature; respectively calculate the average value and variance of the limb volume local time series pattern feature vector to obtain the limb volume local time series pattern feature average value and the limb volume local time series pattern feature variance; add the limb volume local time series pattern feature variance and the preset hyperparameter to obtain a value obtained by multiplying by four to obtain a first energy statistical paradigm value of the limb volume local time series pattern feature; calculate the limb volume local time series pattern feature vector. The square of the difference between the maximum value of the local timing pattern feature and the average value of the local timing pattern feature of the limb volume is calculated to obtain the difference value of the local timing pattern feature of the limb volume; the modulated limb volume local timing pattern feature variance obtained by multiplying the variance of the local timing pattern feature of the limb volume by two is added to the value obtained by multiplying the preset hyperparameter by two and the difference value of the local timing pattern feature of the limb volume to obtain the second energy statistical paradigm value of the local timing pattern feature of the limb volume; the first energy statistical paradigm value of the local timing pattern feature of the limb volume is divided by the second energy statistical paradigm value of the local timing pattern feature of the limb volume to obtain the energy statistical paradigm value of the local timing pattern feature of the limb volume.

[0017] In the above-mentioned lymphedema limb volume measurement data management system, the limb volume local time series pattern energy propagation attenuation coefficient value calculation unit is used to: multiply each limb volume local time series pattern propagation space span value in the sequence of limb volume local time series pattern propagation space span value by a first weight hyperparameter to obtain a sequence of first weighted propagation space span value; use each limb volume local time series pattern propagation space span value in the sequence of limb volume local time series pattern propagation space span value as an exponent of a natural constant to calculate an exponential function value with base e according to position to obtain a sequence of limb volume local time series pattern propagation class support space span value; and multiply each limb volume local time series pattern propagation space span value in the sequence of limb volume local time series pattern propagation space span value by a first weight hyperparameter to obtain a sequence of first weighted propagation space span value. Each limb volume local temporal pattern propagation class support space span value in the sequence of class support space span values ​​is multiplied by the second weight hyperparameter to obtain a sequence of second weighted propagation space span values; the positional sum of the first weighted propagation space span value sequence and the second weighted propagation space span value sequence is calculated to obtain a sequence of weighted sum propagation space span values; the sequence of historical limb volume local temporal pattern energy statistical paradigm values ​​and the sequence of weighted sum propagation space span values ​​are divided by each group of corresponding historical limb volume local temporal pattern energy statistical paradigm values ​​and the weighted sum propagation space span value to obtain a sequence of limb volume local temporal pattern energy propagation attenuation coefficient values.

[0018] In the above-mentioned lymphedema limb volume measurement data management system, the treatment multimodal-limb volume semantic feature fusion module includes: a limb volume temporal semantic enhancement unit, which is used to perform feature semantic interaction enhancement on the limb volume temporal semantic propagation reasoning feature vector based on the treatment multimodal semantic fusion feature vector to obtain an enhanced limb volume temporal semantic propagation reasoning feature vector containing treatment multimodal semantic-limb volume temporal reasoning semantic interaction information; a treatment multimodal semantic enhancement unit, which is used to perform feature semantic interaction enhancement on the treatment multimodal semantic fusion feature vector based on the limb volume temporal semantic propagation reasoning feature vector to obtain an enhanced treatment multimodal semantic fusion feature vector containing limb volume temporal reasoning semantic-treatment multimodal semantic interaction information; an enhanced treatment multimodal semantic-limb volume interaction fusion unit, which is used to perform cascade processing on the enhanced limb volume temporal semantic propagation reasoning feature vector and the enhanced treatment multimodal semantic fusion feature vector to obtain an enhanced treatment multimodal semantic-limb volume interaction fusion feature vector. ; An interactive information fusion response gating value calculation unit, used to input the enhanced treatment multimodal semantics-limb volume interactive fusion feature vector into a gating response function to obtain an enhanced treatment multimodal semantics-limb volume interactive information fusion response gating value; a treatment multimodal semantic gating adjustment unit, used to calculate the positional product between the treatment multimodal semantic fusion feature vector and the enhanced treatment multimodal semantics-limb volume interactive information fusion response gating value to obtain a treatment multimodal semantic gating adjustment vector; a limb volume temporal semantic gating adjustment unit, used to calculate the positional product between the limb volume temporal semantic propagation reasoning feature vector and one minus the enhanced treatment multimodal semantics-limb volume interactive information fusion response gating value to obtain a limb volume temporal semantic gating adjustment vector; a semantic dynamic selection interactive fusion unit, used to add the treatment multimodal semantic gating adjustment vector and the limb volume temporal semantic gating adjustment vector by position to obtain the treatment multimodal semantics-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector.

[0019] In the above-mentioned lymphedema limb volume measurement data management system, the limb volume temporal semantics enhancement unit is used to: multiply the first weight matrix with the treatment multimodal semantics fusion feature vector to obtain the treatment multimodal semantics first weight vector; multiply the limb volume temporal semantics propagation reasoning feature vector with the second weight matrix and the third weight matrix respectively to obtain the limb volume temporal semantics propagation reasoning second weight vector and the limb volume temporal semantics propagation reasoning third weight vector; calculate the product between the transposed vector of the treatment multimodal semantics first weight vector and the limb volume temporal semantics propagation reasoning second weight vector to obtain the treatment multimodal semantics. -limb volume association value, divide the treatment multimodal-limb volume association value by the square root of the length of the limb volume temporal semantic propagation reasoning feature vector to obtain the treatment multimodal-limb volume weight value; input the treatment multimodal-limb volume weight value into the softmax function to obtain the treatment multimodal-limb volume activation weight value; use the treatment multimodal-limb volume activation weight value as the weight value, and weight the transposed vector of the third weight vector of the limb volume temporal semantic propagation reasoning to obtain the enhanced limb volume temporal semantic propagation reasoning feature vector containing the treatment multimodal semantics-limb volume temporal reasoning semantic interaction information.

[0020] In the above-mentioned lymphedema limb volume measurement data management system, the interactive information fusion response gating value calculation unit is used to: calculate the matrix product of the enhanced treatment multimodal semantics-limb volume interactive fusion feature vector and the parameter matrix, and then add the obtained feature vector and the bias vector by position to obtain the enhanced treatment multimodal semantics-limb volume interactive fusion bias feature vector; input the enhanced treatment multimodal semantics-limb volume interactive fusion bias feature vector into the sigmoid activation function to obtain the enhanced treatment multimodal semantics-limb volume interactive information fusion response gating value.

[0021] In the above-mentioned lymphedema limb volume measurement data management system, the abnormal detection result generation module is used to: pass the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector through the classifier-based lymphedema limb volume abnormality identifier to obtain the abnormal detection result, and the abnormal detection result is used to indicate whether the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal; in response to the abnormal detection result that the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal, an early warning prompt signal is generated to medical staff.

[0022] According to another aspect of the present application, a method for managing lymphedema limb volume measurement data is provided, comprising:

[0023] Obtaining a time series of lymphedema limb volume data of a patient with a monitored lymphedema limb;

[0024] Acquiring treatment data of the monitored lymphedema limb patient, wherein the treatment data includes physical therapy data and drug therapy data;

[0025] Arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting the vector into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors;

[0026] Passing the sequence of the limb volume local temporal pattern feature vectors through a limb volume temporal node semantic propagation reasoner based on a node energy decay mechanism to obtain a limb volume temporal semantic propagation reasoning feature vector;

[0027] After semantically encoding the treatment data of the monitored lymphedema limb patient to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector, the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector are fused to obtain a treatment multimodal semantic fusion feature vector;

[0028] The treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector are passed through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature;

[0029] Based on the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion features, an abnormal detection result is obtained, and based on the abnormal detection result, a corresponding early warning prompt signal is generated to the medical staff;

[0030] The method of arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting the vector into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors includes:

[0031] Arranging the time series of the lymphedema limb volume data according to the time dimension into the lymphedema limb volume time series input vector;

[0032] The lymphedema limb volume temporal input vector is input into a lymphedema limb volume temporal pattern feature extractor based on a 1D convolutional neural network model to obtain a sequence of the limb volume local temporal pattern feature vectors.

[0033] Compared with the prior art, the lymphedema limb volume measurement data management system and method provided by the present application adopts AI-based data analysis and processing technology to extract local temporal features, semantic reasoning and semantic fusion of the lymphedema limb volume data and treatment data of the monitored lymphedema limb patients, so as to automatically judge whether the volume of the patient's lymphedema limb is abnormal according to the semantic interaction fusion features between the limb volume data and the treatment data, and generate corresponding warnings. In this way, the automated limb volume abnormality detection and warning functions can be realized, the misjudgment caused by the doctor's subjective experience can be reduced, and the objectivity and accuracy of the lymphedema limb volume assessment can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art description are briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0035] Figure 1 4 is a system block diagram of a lymphedema limb volume measurement data management system according to an embodiment of the present application.

[0036] Figure 2 Schematic diagram of the architecture of a lymphedema limb volume measurement data management system according to an embodiment of the present application.

[0037] Figure 3 Flow chart of a method for managing lymphedema limb volume measurement data according to an embodiment of the present application. DETAILED DESCRIPTION

[0038] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0039] Lymphedema is a chronic disease that is difficult to cure, and its treatment requires long-term and complex evaluation. Existing evaluation methods such as water replacement method have the problems of waste of resources, complex operation, low precision and infection risk; lymphoscintigraphy is costly and may cause allergies due to the use of contrast agents. Although MRI and CT have clear images, they are expensive and not suitable for patients with metal implants. In addition, the evaluation of treatment effect often relies on the subjective judgment of doctors, which may be biased due to personal differences among doctors. At the same time, the lack of quantitative tools may lead to misjudgment of changes in the condition.

[0040] Therefore, in response to the above problems, the technical concept of the present application is to obtain the time series of the volume data of the lymphedema limb of the patient with monitored lymphedema limb and the treatment data (physical therapy and drug therapy) of the patient with monitored lymphedema limb, and use AI-based data analysis and processing technology to extract local temporal features, semantic reasoning and semantic fusion of the lymphedema limb volume data and the treatment data respectively, so as to automatically judge whether there is an abnormality in the volume of the lymphedema limb of the patient with monitored lymphedema limb according to the semantic interaction fusion features between the limb volume data and the treatment data, and generate a corresponding warning. In this way, the automatic limb volume abnormality detection and warning function can be realized, and the misjudgment caused by the subjective experience of doctors can be reduced. At the same time, the temporal information and semantic associations in the volume data and treatment data can be effectively captured, helping medical staff to fully understand the patient's condition and realize more intelligent lymphedema limb volume measurement data management.

[0041] Figure 1 4 is a system block diagram of a lymphedema limb volume measurement data management system according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of a lymphedema limb volume measurement data management system according to an embodiment of the present application. Figure 1 and Figure 2As shown, in the lymphedema limb volume measurement data management system 100, it includes: a patient lymphedema limb volume data acquisition module 110, which is used to obtain the time series of lymphedema limb volume data of the monitored lymphedema limb patient; a patient lymphedema limb treatment data acquisition module 120, which is used to obtain the treatment data of the monitored lymphedema limb patient, and the treatment data includes physical therapy data and drug treatment data; a limb volume time series feature extraction module 130, which is used to arrange the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then input it into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors; a limb volume time series feature semantic reasoning module 140, which is used to pass the sequence of the limb volume local time series pattern feature vectors through a limb volume time series node semantic propagation reasoner based on a node energy attenuation mechanism to obtain a limb volume time series semantic propagation reasoning feature vector; a treatment data multimodal fusion module 150, which is used After semantic encoding of the treatment data of the monitored lymphedema limb patient to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector, the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector are fused to obtain a treatment multimodal semantic fusion feature vector; a treatment multimodal-limb volume semantic feature fusion module 160 is used to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature by passing the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector through a dynamic feature selection interactive fusion module based on a bidirectional attention network; an abnormality detection result generation module 170 is used to obtain an abnormality detection result based on the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature, and generate a corresponding early warning prompt signal to medical staff based on the abnormality detection result.

[0042] In the embodiment of the present application, the patient lymphedema limb volume data acquisition module 110 and the patient lymphedema limb treatment data acquisition module 120 are respectively used to obtain the time series of lymphedema limb volume data of the monitored lymphedema limb patient and obtain the treatment data of the monitored lymphedema limb patient, and the treatment data includes physical therapy data and drug therapy data. It should be understood that the time series of lymphedema limb volume data can reflect the change of limb volume over time, and by analyzing the change trend of limb volume, the potential abnormal state of limb volume can be identified. The patient's physical therapy data and drug therapy data can help understand the specific situation of the physical therapy and drug therapy received by the patient, so as to analyze the impact of these treatment measures on the change of limb volume. By combining and analyzing the lymphedema limb volume data and treatment data, a more complete limb volume abnormality recognition model can be constructed, thereby improving the accuracy of limb volume abnormality detection. In particular, in a specific implementation of the embodiment of the present application, the lymphedema limb volume data and treatment data can be obtained through the patient's clinical treatment records.

[0043] In the embodiment of the present application, the limb volume temporal feature extraction module 130 is used to arrange the time series of the lymphedema limb volume data into a lymphedema limb volume temporal input vector and then input it into a lymphedema limb volume temporal pattern feature extractor to obtain a sequence of limb volume local temporal pattern feature vectors. Specifically, in the embodiment of the present application, the limb volume temporal feature extraction module includes: a lymphedema limb volume data structuring unit, which is used to arrange the time series of the lymphedema limb volume data into the lymphedema limb volume temporal input vector according to the time dimension; a lymphedema limb volume local temporal feature generation unit, which is used to input the lymphedema limb volume temporal input vector into a lymphedema limb volume temporal pattern feature extractor based on a 1D convolutional neural network model to obtain a sequence of limb volume local temporal pattern feature vectors. It should be understood that the lymphedema limb volume data has time series characteristic information. Therefore, in order to better understand the time series change fluctuation trend of the limb volume, in the technical solution of the present application, the time series of the lymphedema limb volume data is arranged according to the time dimension as a lymphedema limb volume time series input vector. Then, in order to capture and mine the local time series characteristic information in the lymphedema limb volume time series input vector, in the technical solution of the present application, the lymphedema limb volume time series input vector is input into a lymphedema limb volume time series pattern feature extractor based on a 1D convolutional neural network model to obtain a sequence of limb volume local time series pattern feature vectors.

[0044] In the embodiment of the present application, the limb volume temporal feature semantic reasoning module 140 is used to pass the sequence of the limb volume local temporal pattern feature vectors through the limb volume temporal node semantic propagation reasoner based on the node energy decay mechanism to obtain the limb volume temporal semantic propagation reasoning feature vector. Accordingly, considering that each limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vectors expresses the temporal change information of the limb volume in a local time period, therefore, the sequence of the limb volume local temporal pattern feature vectors is used as a benchmark to better identify the more critical and significant limb volume change feature points in the entire temporal sequence, so as to accurately infer the entire lymphedema limb volume feature. In the technical solution of the present application, the sequence of the limb volume local temporal pattern feature vectors is passed through the limb volume temporal node semantic propagation reasoner based on the node energy decay mechanism to obtain the limb volume temporal semantic propagation reasoning feature vector. It is worth mentioning that the limb volume temporal node semantic propagation reasoner based on the node energy decay mechanism uses the information decay process that changes over time in the time series data to help the model better understand the dynamic changes and development trends in the time series data. In other words, by dynamically evaluating the energy level of the characteristic vector of the local time series pattern of each limb volume and capturing the topological structure between each node, the energy propagation attenuation and the propagation of time-space characteristics can be achieved. This can improve the sensitivity of the model to the dynamic changes in the time series of node features, thereby effectively capturing the trends and patterns of limb volume features changing over time, and thus better inferring the evolution process of limb volume time series data.

[0045] Specifically, in the embodiment of the present application, the limb volume temporal feature semantic reasoning module 140 includes: a limb volume local temporal pattern feature energy statistical paradigm value calculation unit, which is used to calculate the energy statistical paradigm value of each limb volume local temporal pattern feature vector based on the maximum value, average value and variance of each limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vector to obtain a sequence of limb volume local temporal pattern energy statistical paradigm values, wherein the energy statistical paradigm value corresponding to the current limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern energy statistical paradigm values ​​is calculated. The energy statistical paradigm value is used as the current node energy statistical paradigm value and other energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of the current limb volume local time series pattern energy statistical paradigm value and the historical limb volume local time series pattern energy statistical paradigm value; a limb volume local time series pattern node propagation space span value calculation unit is used to count the node propagation space span values ​​between each other limb volume local time series pattern feature vector in the sequence of the limb volume local time series pattern feature vector and the current limb volume local time series pattern feature vector to obtain a sequence of limb volume local time series pattern propagation space span values; limb volume local time series pattern energy propagation an attenuation coefficient value calculation unit, for determining the energy propagation attenuation coefficient values ​​of other limb volume local time series pattern feature vectors in the sequence of limb volume local time series pattern feature vectors based on the sequence of propagation spatial span values ​​of the limb volume local time series pattern and the sequence of energy statistical paradigm values ​​of the historical limb volume local time series pattern, so as to obtain a sequence of limb volume local time series pattern energy propagation attenuation coefficient values, wherein the limb volume local time series pattern energy propagation attenuation coefficient values ​​are inversely correlated with the historical limb volume local time series pattern energy statistical paradigm values; and a limb volume local time series pattern feature weighting unit, for determining the energy propagation attenuation coefficient values ​​of other limb volume local time series pattern feature vectors in the sequence of limb volume local time series pattern feature vectors, so as to obtain a sequence of limb volume local time series pattern energy propagation attenuation coefficient values. A sequence of local temporal pattern energy propagation attenuation coefficient values ​​is used as a weight sequence, and the weighted sum of the sequences of all other limb volume local temporal pattern feature vectors in the sequence of the limb volume local temporal pattern feature vector is calculated to obtain a historical limb volume local temporal pattern energy attenuation temporal aggregation feature vector; a limb volume local temporal pattern propagation inference unit is used to fuse the current limb volume local temporal pattern energy statistical paradigm value to calculate the weighted sum of the historical limb volume local temporal pattern energy attenuation temporal aggregation feature vector and the current limb volume local temporal pattern feature vector to obtain the limb volume temporal semantic propagation inference feature vector.

[0046] More specifically, in an embodiment of the present application, the limb volume local time series pattern feature energy statistical paradigm value calculation unit is used to: extract the maximum value of the limb volume local time series pattern feature vector to obtain the maximum value of the limb volume local time series pattern feature; respectively calculate the mean value and variance of the limb volume local time series pattern feature vector to obtain the limb volume local time series pattern feature mean value and the limb volume local time series pattern feature variance; multiply the value obtained by adding the limb volume local time series pattern feature variance and a preset hyperparameter by four to obtain the first energy statistical paradigm value of the limb volume local time series pattern feature; calculate the limb volume local time series pattern feature The square of the difference between the maximum value of the characteristic and the average value of the local timing pattern characteristic of the limb volume is obtained to obtain the difference value of the local timing pattern characteristic of the limb volume; the modulated limb volume local timing pattern characteristic variance obtained by multiplying the variance of the local timing pattern characteristic of the limb volume by two is added to the value obtained by multiplying the preset hyperparameter by two and the difference value of the local timing pattern characteristic of the limb volume to obtain the second energy statistical paradigm value of the local timing pattern characteristic of the limb volume; the first energy statistical paradigm value of the local timing pattern characteristic of the limb volume is divided by the second energy statistical paradigm value of the local timing pattern characteristic of the limb volume to obtain the energy statistical paradigm value of the local timing pattern characteristic of the limb volume.

[0047] More specifically, in an embodiment of the present application, the limb volume local temporal pattern energy propagation attenuation coefficient value calculation unit is used to: multiply each limb volume local temporal pattern propagation spatial span value in the sequence of limb volume local temporal pattern propagation spatial span values ​​by a first weight hyperparameter to obtain a sequence of first weighted propagation spatial span values; use each limb volume local temporal pattern propagation spatial span value in the sequence of limb volume local temporal pattern propagation spatial span values ​​as an exponent of a natural constant to calculate an exponential function value with base e according to position to obtain a sequence of limb volume local temporal pattern propagation class support spatial span values; use the limb volume local temporal pattern propagation class support space span value as an exponent of a natural constant Each limb volume local temporal pattern propagation class supports multiplying the spatial span value with the second weight hyperparameter in the sequence of span values ​​to obtain a sequence of second weighted propagation spatial span values; calculating the positional sum of the first weighted propagation spatial span value sequence and the second weighted propagation spatial span value sequence to obtain a sequence of weighted sum propagation spatial span values; dividing each group of corresponding historical limb volume local temporal pattern energy statistical paradigm values ​​and the weighted sum propagation spatial span value in the sequence of historical limb volume local temporal pattern energy statistical paradigm values ​​and the sequence of weighted sum propagation spatial span values ​​to obtain a sequence of limb volume local temporal pattern energy propagation attenuation coefficient values.

[0048] In the embodiment of the present application, specifically, the limb volume temporal feature semantic reasoning module 140 is used to: process the sequence of the limb volume local temporal pattern feature vectors through a limb volume temporal node semantic propagation reasoner based on a node energy decay mechanism, using the following propagation reasoning formula to obtain the limb volume temporal semantic propagation reasoning feature vector; wherein the propagation reasoning formula is:

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] ;

[0054] in, is the sequence of the limb volume local temporal pattern feature vectors, and are the first and The local temporal pattern feature vector of the limb volume, represents the number of feature vectors in the sequence of feature vectors of the local temporal pattern of the limb volume, and are the first The mean value of the local temporal pattern characteristics of the limb volume and the variance of the local temporal pattern characteristics of the limb volume, For the The maximum value of the eigenvector of the local temporal pattern of the limb volume, is the preset hyperparameter, It is The energy statistical norm value of the local temporal pattern feature vector of the limb volume, It is The limb volume local temporal pattern feature vector The eigenvalues ​​at the positions, is the number of eigenvalues ​​in the eigenvector of the local temporal pattern of each limb volume, and They are and The energy statistical norm value of the local temporal pattern feature vector of the limb volume, express and The number of eigenvectors between , , and is the weight hyperparameter, is the limb volume temporal semantic propagation reasoning feature vector.

[0055] In an embodiment of the present application, the treatment data multimodal fusion module 150 is used to semantically encode the treatment data of the monitored lymphedema limb patient to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector, and then fuse the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector to obtain a treatment multimodal semantic fusion feature vector. It should be understood that the treatment data of the monitored lymphedema limb patient contains semantic associations between respective category information. Therefore, the treatment data of the monitored lymphedema limb patient is semantically encoded to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector. Next, in order to more comprehensively and comprehensively display the semantic information of the treatment data, the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector are further fused to obtain a treatment multimodal semantic fusion feature vector.

[0056] In an embodiment of the present application, the treatment multimodal-limb volume semantic feature fusion module 160 is used to pass the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature. Accordingly, in order to achieve the interaction and fusion of the features of the two, the semantic relevance between the treatment data and the limb volume temporal reasoning data is fully considered, so as to more comprehensively capture the complex relationship between the two. In the technical solution of the present application, the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector are passed through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector.

[0057] Specifically, the dynamic feature selection interactive fusion module based on the bidirectional attention network dynamically selects and focuses on the most relevant features by respectively calculating the interactive semantic feature information between the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector. Then, the interactive information of the two is fused and gated to obtain a gate value. Finally, the importance and weight of the two features are flexibly adjusted based on the gate value to better capture the semantic association between the two data, thereby achieving more accurate and effective feature fusion, and finally obtaining a comprehensive treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector.

[0058] Specifically, in an embodiment of the present application, the treatment multimodal-limb volume semantic feature fusion module includes: a limb volume temporal semantic reinforcement unit, which is used to perform feature semantic interaction reinforcement on the limb volume temporal semantic propagation reasoning feature vector based on the treatment multimodal semantic fusion feature vector to obtain an enhanced limb volume temporal semantic propagation reasoning feature vector containing treatment multimodal semantic-limb volume temporal reasoning semantic interaction information; a treatment multimodal semantic reinforcement unit, which is used to perform feature semantic interaction reinforcement on the treatment multimodal semantic fusion feature vector based on the limb volume temporal semantic propagation reasoning feature vector to obtain an enhanced treatment multimodal semantic fusion feature vector containing limb volume temporal reasoning semantic-treatment multimodal semantic interaction information; an enhanced treatment multimodal semantic-limb volume interaction fusion unit, which is used to cascade the enhanced limb volume temporal semantic propagation reasoning feature vector and the enhanced treatment multimodal semantic fusion feature vector to obtain an enhanced treatment multimodal semantic-limb volume interaction fusion feature vector; interaction signal The information fusion response gating value calculation unit is used to input the enhanced treatment multimodal semantics-limb volume interaction fusion feature vector into the gating response function to obtain the enhanced treatment multimodal semantics-limb volume interaction information fusion response gating value; the treatment multimodal semantic gating adjustment unit is used to calculate the positional product between the treatment multimodal semantic fusion feature vector and the enhanced treatment multimodal semantics-limb volume interaction information fusion response gating value to obtain the treatment multimodal semantic gating adjustment vector; the limb volume temporal semantic gating adjustment unit is used to calculate the positional product between the limb volume temporal semantic propagation reasoning feature vector and one minus the enhanced treatment multimodal semantics-limb volume interaction information fusion response gating value to obtain the limb volume temporal semantic gating adjustment vector; the semantic dynamic selection interaction fusion unit is used to add the treatment multimodal semantic gating adjustment vector and the limb volume temporal semantic gating adjustment vector by position to obtain the treatment multimodal semantics-limb volume temporal reasoning semantic dynamic selection interaction fusion feature vector.

[0059] More specifically, in an embodiment of the present application, the limb volume temporal semantics reinforcement unit is used to: multiply the first weight matrix by the treatment multimodal semantics fusion feature vector to obtain the treatment multimodal semantics first weight vector; multiply the limb volume temporal semantics propagation reasoning feature vector by the second weight matrix and the third weight matrix respectively to obtain the limb volume temporal semantics propagation reasoning second weight vector and the limb volume temporal semantics propagation reasoning third weight vector; calculate the product between the transposed vector of the treatment multimodal semantics first weight vector and the limb volume temporal semantics propagation reasoning second weight vector to obtain the treatment multimodal-limb After the volume association value is obtained, the treatment multimodal-limb volume association value is divided by the square root of the length of the limb volume temporal semantic propagation reasoning feature vector to obtain the treatment multimodal-limb volume weight value; the treatment multimodal-limb volume weight value is input into the softmax function to obtain the treatment multimodal-limb volume activation weight value; the treatment multimodal-limb volume activation weight value is used as the weight value, and the transposed vector of the third weight vector of the limb volume temporal semantic propagation reasoning is weighted to obtain the enhanced limb volume temporal semantic propagation reasoning feature vector containing the treatment multimodal semantics-limb volume temporal reasoning semantic interaction information.

[0060] More specifically, in an embodiment of the present application, the interactive information fusion response gating value calculation unit is used to: calculate the matrix product of the enhanced therapy multimodal semantic-limb volume interactive fusion feature vector and the parameter matrix, and then add the obtained feature vector and the bias vector by position to obtain the enhanced therapy multimodal semantic-limb volume interactive fusion bias feature vector; input the enhanced therapy multimodal semantic-limb volume interactive fusion bias feature vector into the sigmoid activation function to obtain the enhanced therapy multimodal semantic-limb volume interactive information fusion response gating value.

[0061] In the embodiment of the present application, specifically, the treatment multimodal-limb volume semantic feature fusion module 160 is used to: process the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector through a dynamic feature selection interactive fusion module based on a bidirectional attention network, using the following interactive fusion formula to obtain the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector; wherein the interactive fusion formula is:

[0062] ;

[0063] ;

[0064] ;

[0065] ;

[0066] in, and are respectively the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, , and represent the first weight matrix, the second weight matrix, and the third weight matrix of the treatment multimodal semantic fusion feature vector respectively, represents the transpose of a vector, , and are respectively the first weight matrix, the second weight matrix and the third weight matrix of the limb volume temporal semantic propagation reasoning feature vector, is the length of the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, yes function, is the enhanced limb volume temporal semantic propagation reasoning feature vector containing the treatment multimodal semantic-limb volume temporal reasoning semantic interaction information, is the enhanced treatment multimodal semantic fusion feature vector containing the limb volume temporal reasoning semantics-treatment multimodal semantic interaction information, It is cascade processing. is the parameter matrix, is the bias vector, is the sigmoid function, is the information fusion gating value between the enhanced limb volume temporal semantic propagation reasoning feature vector and the enhanced treatment multimodal semantic fusion feature vector, The treatment is multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector.

[0067] In an embodiment of the present application, the abnormal detection result generation module 170 is used to obtain an abnormal detection result based on the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion features, and based on the abnormal detection result, generate a corresponding early warning prompt signal to medical staff. Specifically, in an embodiment of the present application, the abnormal detection result generation module 170 is used to: pass the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector through a classifier-based lymphedema limb volume abnormality identifier to obtain the abnormal detection result, and the abnormal detection result is used to indicate whether the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal; in response to the abnormal detection result that the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal, generate an early warning prompt signal to medical staff. That is, the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion features obtained by bidirectional interactive fusion of the treatment multimodal semantics fusion feature vector and the limb volume temporal semantics propagation reasoning feature vector are used for classification processing, so as to automatically determine whether the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal, and generate corresponding warnings. In this way, the automated limb volume abnormality detection and warning functions can be realized, reducing misjudgments caused by the doctor's subjective experience, and effectively capturing the temporal information and semantic associations in the volume data and treatment data, helping medical staff to fully understand the patient's condition.

[0068] In particular, considering that the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector respectively represent the multimodal semantic encoding features of the treatment data of the monitored lymphedema limb patients and the propagation semantic features of the one-dimensional temporal local correlation of the lymphedema limb volume data based on the temporal node energy attenuation, when performing dynamic feature selection interactive fusion based on the bidirectional attention network, it is necessary to compensate for the imbalance in dynamic selection interactive fusion caused by the imbalance in the correspondence ratio under the fine-grained distribution of different feature expression dimension sequences, so as to improve the expression effect of the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector, thereby improving the accuracy of the classification results.

[0069] Therefore, preferably, the present application corrects the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector when passing the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector through a classifier-based lymphedema limb volume abnormality identifier to obtain an abnormality detection result, including the following steps: calculating the first norm and second norm of the mean vector of the treatment multimodal semantics fusion feature vector and the limb volume temporal semantics propagation reasoning feature vector; calculating the weighted sum of the square root of the second norm and the first norm, and performing point multiplication with the point addition sum vector of the treatment multimodal semantics fusion feature vector and the limb volume temporal semantics propagation reasoning feature vector to obtain a first treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction subvector; performing the point product vector of the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction subvector with the point product vector of the treatment multimodal semantics fusion feature vector and the limb volume temporal semantics propagation reasoning feature vector. Select the square root of the length of the interactive fusion feature vector for point multiplication to obtain the second therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction subvector; calculate the weighted sum of the first therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction subvector and the second therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction subvector to obtain the therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction vector; perform point multiplication of the therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction vector and the therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector to obtain the corrected therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector, and pass the corrected therapeutic multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector through a classifier-based lymphedema limb volume abnormality identifier to obtain an abnormality detection result.

[0070] The treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction vector is expressed as:

[0071] ;

[0072] in, and are respectively the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, is the mean vector of the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, is the one-norm of the eigenvector, is the reciprocal of the square root of the two-norm of the eigenvector, It is added by location point. It is the point multiplication by position. is the length of the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector, and and is the weighted sum weight as a hyperparameter, It is the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion correction vector.

[0073] Therefore, based on the mean vector norm between the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, the constrained representation of the structured foreground and background distinction of the supermanifold of the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector is used to model the feature-level key correspondence between the feature vectors of the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, and the global correlation relationship between the corresponding features is adjusted, so as to make positive fine-grained correspondence suggestions through the unbalanced proportion control between the corresponding eigenvalues ​​of the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector, so as to avoid the imbalance of fusion domain offset between the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector by focusing on the focus.

[0074] In this way, by correcting the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector through the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion correction vector, the expression effect of the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector can be improved, thereby improving the accuracy of the abnormality detection result obtained by the classifier-based lymphedema limb volume abnormality identifier.

[0075] In summary, the lymphedema limb volume measurement data management system 100 based on the embodiment of the present application is explained, which uses AI-based data analysis and processing technology to perform local temporal feature extraction, semantic reasoning and semantic fusion on the lymphedema limb volume data and treatment data of the monitored lymphedema limb patient, so as to automatically judge whether the volume of the patient's lymphedema limb is abnormal according to the semantic interaction fusion features between the limb volume data and the treatment data, and generate corresponding warnings. In this way, the automated limb volume abnormality detection and warning functions can be realized, the misjudgment caused by the doctor's subjective experience can be reduced, and the objectivity and accuracy of the lymphedema limb volume assessment can be improved.

[0076] As described above, the lymphedema limb volume measurement data management system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a lymphedema limb volume measurement data management algorithm. In a possible implementation, the lymphedema limb volume measurement data management system 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the lymphedema limb volume measurement data management system 100 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the lymphedema limb volume measurement data management system 100 can also be one of the many hardware modules of the wireless terminal.

[0077] Alternatively, in another example, the lymphedema limb volume measurement data management system 100 and the wireless terminal may also be separate devices, and the lymphedema limb volume measurement data management system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0078] Figure 3 FIG. 1 is a flow chart of a method for managing lymphedema limb volume measurement data according to an embodiment of the present application. Figure 3As shown, in the lymphedema limb volume measurement data management method, it includes: S110, obtaining the time series of lymphedema limb volume data of the monitored lymphedema limb patient; S120, obtaining the treatment data of the monitored lymphedema limb patient, and the treatment data includes physical therapy data and drug treatment data; S130, arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting it into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors; S140, passing the sequence of limb volume local time series pattern feature vectors through a limb volume time series node semantic propagation reasoner based on a node energy attenuation mechanism to obtain a limb volume time series semantic propagation reasoning feature vector; S150, semantically encoding the treatment data of the monitored lymphedema limb patient. After obtaining the semantic coding feature vector of physical therapy data and the semantic coding feature vector of drug therapy data by coding, the semantic coding feature vector of physical therapy data and the semantic coding feature vector of drug therapy data are fused to obtain the treatment multimodal semantic fusion feature vector; S160, the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector are passed through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature; S170, based on the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature, an abnormality detection result is obtained, and based on the abnormality detection result, a corresponding early warning prompt signal is generated to medical staff.

[0079] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned lymphedema limb volume measurement data management method have been described in the above reference. Figure 1 to Figure 2 The lymphedema limb volume measurement data management system has been described in detail, and therefore, its repeated description will be omitted.

[0080] The above descriptions of various implementations of the present disclosure are exemplary and not exhaustive. They are not limited to the disclosed implementations, and many modifications and changes are obvious to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The terms used herein are selected to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other ordinary skill in the art to understand the various practical methods disclosed herein.

Claims

1. A lymphedema limb volume measurement data management system, characterized in that: include: A patient lymphedema limb volume data acquisition module, used to acquire a time series of lymphedema limb volume data of a monitored lymphedema limb patient; A patient lymphedema limb treatment data acquisition module, used to acquire treatment data of the monitored lymphedema limb patient, the treatment data including physical therapy data and drug therapy data; A limb volume time series feature extraction module, used for arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting the vector into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors; A limb volume temporal feature semantic reasoning module, used for passing the sequence of the limb volume local temporal pattern feature vectors through a limb volume temporal node semantic propagation reasoner based on a node energy decay mechanism to obtain a limb volume temporal semantic propagation reasoning feature vector; a treatment data multimodal fusion module, configured to perform semantic encoding on the treatment data of the monitored lymphedema limb patient to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector, and then fuse the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector to obtain a treatment multimodal semantic fusion feature vector; A treatment multimodal-limb volume semantic feature fusion module, used for fusing the treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature; An abnormal detection result generation module is used to dynamically select interactive fusion features based on the treatment multimodal semantics-limb volume temporal reasoning semantics to obtain abnormal detection results, and based on the abnormal detection results, generate corresponding early warning prompt signals to medical staff; Wherein, the limb volume temporal feature extraction module includes: A lymphedema limb volume data structuring unit, used for arranging the time series of the lymphedema limb volume data into the lymphedema limb volume time series input vector according to the time dimension; A lymphedema limb volume local temporal feature generating unit, used for inputting the lymphedema limb volume temporal input vector into a lymphedema limb volume temporal pattern feature extractor based on a 1D convolutional neural network model to obtain a sequence of the limb volume local temporal pattern feature vectors; The limb volume temporal feature semantic reasoning module includes: A limb volume local temporal pattern feature energy statistical paradigm value calculation unit, used for calculating the energy statistical paradigm value of each limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vectors based on the maximum value, average value and variance of each limb volume local temporal pattern feature vector in the sequence of the limb volume local temporal pattern feature vectors to obtain a sequence of limb volume local temporal pattern energy statistical paradigm values, wherein the energy statistical paradigm value corresponding to the current limb volume local temporal pattern feature vector in the sequence of limb volume local temporal pattern energy statistical paradigm values ​​is used as the current node energy statistical paradigm value and other energy statistical paradigm values ​​are used as historical node energy statistical paradigm values ​​to obtain a sequence of current limb volume local temporal pattern energy statistical paradigm values ​​and historical limb volume local temporal pattern energy statistical paradigm values; A limb volume local time series pattern node propagation space span value calculation unit, used for counting the node propagation space span values ​​between each other limb volume local time series pattern feature vector in the sequence of the limb volume local time series pattern feature vector and the current limb volume local time series pattern feature vector to obtain a sequence of limb volume local time series pattern propagation space span values; A limb volume local time series pattern energy propagation attenuation coefficient value calculation unit is used to determine the energy propagation attenuation coefficient values ​​of other limb volume local time series pattern feature vectors in the sequence of the limb volume local time series pattern feature vectors based on the sequence of the limb volume local time series pattern propagation space span values ​​and the sequence of the historical limb volume local time series pattern energy statistical paradigm values ​​to obtain a sequence of limb volume local time series pattern energy propagation attenuation coefficient values, wherein the limb volume local time series pattern energy propagation attenuation coefficient values ​​are in an inverse correlation with the historical limb volume local time series pattern energy statistical paradigm values; A limb volume local time series pattern feature weighting unit, used to calculate the weighted sum of all other limb volume local time series pattern feature vector sequences in the sequence of the limb volume local time series pattern feature vector, using the sequence of the limb volume local time series pattern energy propagation attenuation coefficient values ​​as a weight sequence, to obtain a historical limb volume local time series pattern energy attenuation time series aggregation feature vector; A limb volume local temporal pattern propagation reasoning unit, used for fusing the current limb volume local temporal pattern energy statistical paradigm value to calculate the weighted sum of the historical limb volume local temporal pattern energy attenuation temporal aggregation feature vector and the current limb volume local temporal pattern feature vector to obtain the limb volume temporal semantic propagation reasoning feature vector; Wherein, the limb volume local time series pattern characteristic energy statistical paradigm value calculation unit is used to: Extracting the maximum value of the limb volume local time series pattern feature vector to obtain the limb volume local time series pattern feature maximum value; Respectively calculating the average value and variance of the limb volume local time series pattern feature vector to obtain the limb volume local time series pattern feature average value and the limb volume local time series pattern feature variance; The value obtained by adding the variance of the local temporal pattern feature of the limb volume and the preset hyperparameter is multiplied by four to obtain the first energy statistical paradigm value of the local temporal pattern feature of the limb volume; Calculating the square of the difference between the maximum value of the local temporal pattern feature of the limb volume and the average value of the local temporal pattern feature of the limb volume to obtain a difference value of the local temporal pattern feature of the limb volume; The second energy statistical paradigm value of the limb volume local temporal pattern feature is obtained by adding the modulated limb volume local temporal pattern feature variance obtained by multiplying the limb volume local temporal pattern feature variance by two, the value obtained by multiplying the preset hyperparameter by two, and the limb volume local temporal pattern feature difference value; Dividing the first energy statistical paradigm value of the local temporal pattern characteristic of the limb volume by the second energy statistical paradigm value of the local temporal pattern characteristic of the limb volume to obtain the energy statistical paradigm value of the local temporal pattern of the limb volume; Wherein, the limb volume local time series mode energy propagation attenuation coefficient value calculation unit is used for: Multiplying each limb volume local temporal pattern propagation spatial span value in the sequence of limb volume local temporal pattern propagation spatial span values ​​by a first weight hyperparameter to obtain a sequence of first weighted propagation spatial span values; Using each limb volume local time series pattern propagation space span value in the sequence of limb volume local time series pattern propagation space span values ​​as an exponent of a natural constant to calculate an exponential function value with base e according to position to obtain a sequence of limb volume local time series pattern propagation class support space span values; Multiplying each limb volume local temporal pattern propagation class support space span value in the sequence of limb volume local temporal pattern propagation class support space span values ​​by a second weight hyperparameter to obtain a sequence of second weighted propagation space span values; Calculating a position-wise sum of the sequence of the first weighted propagation spatial span values ​​and the sequence of the second weighted propagation spatial span values ​​to obtain a sequence of weighted sum propagation spatial span values; Dividing each corresponding group of historical limb volume local time series pattern energy statistical paradigm values ​​and weighted sum propagation space span values ​​in the sequence of the historical limb volume local time series pattern energy statistical paradigm values ​​and the sequence of the weighted sum propagation space span values ​​to obtain the sequence of the limb volume local time series pattern energy propagation attenuation coefficient values; Wherein, the treatment multimodal-limb volume semantic feature fusion module includes: A limb volume temporal semantics reinforcement unit, configured to perform feature semantics interaction reinforcement on the limb volume temporal semantics propagation reasoning feature vector based on the treatment multimodal semantics fusion feature vector to obtain a reinforced limb volume temporal semantics propagation reasoning feature vector containing treatment multimodal semantics-limb volume temporal reasoning semantics interaction information; A treatment multimodal semantic enhancement unit, configured to perform feature semantic interaction enhancement on the treatment multimodal semantic fusion feature vector based on the limb volume temporal semantic propagation reasoning feature vector to obtain an enhanced treatment multimodal semantic fusion feature vector containing limb volume temporal reasoning semantics-treatment multimodal semantic interaction information; An intensive therapy multimodal semantic-limb volume interactive fusion unit, used for cascading the intensive limb volume temporal semantic propagation reasoning feature vector and the intensive therapy multimodal semantic fusion feature vector to obtain an intensive therapy multimodal semantic-limb volume interactive fusion feature vector; An interactive information fusion response gating value calculation unit, used for inputting the intensive treatment multimodal semantics-limb volume interactive fusion feature vector into a gating response function to obtain an intensive treatment multimodal semantics-limb volume interactive information fusion response gating value; a treatment multimodal semantic gating adjustment unit, used for calculating the positional product between the treatment multimodal semantic fusion feature vector and the enhanced treatment multimodal semantic-limb volume interaction information fusion response gating value to obtain a treatment multimodal semantic gating adjustment vector; A limb volume temporal semantic gating adjustment unit, used for calculating the positional product between the limb volume temporal semantic propagation reasoning feature vector and one minus the intensive treatment multimodal semantic-limb volume interaction information fusion response gating value to obtain a limb volume temporal semantic gating adjustment vector; A semantic dynamic selection interactive fusion unit, used for adding the treatment multimodal semantic gating adjustment vector and the limb volume temporal semantic gating adjustment vector by position to obtain the treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector; Wherein, the limb volume temporal semantic reinforcement unit is used to: Multiplying the first weight matrix by the treatment multimodal semantic fusion feature vector to obtain a treatment multimodal semantic first weight vector; Multiplying the limb volume temporal semantic propagation reasoning feature vector with the second weight matrix and the third weight matrix respectively to obtain a limb volume temporal semantic propagation reasoning second weight vector and a limb volume temporal semantic propagation reasoning third weight vector; After calculating the product between the transposed vector of the treatment multimodal semantic first weight vector and the limb volume temporal semantic propagation reasoning second weight vector to obtain the treatment multimodal-limb volume association value, the treatment multimodal-limb volume association value is divided by the square root of the length of the limb volume temporal semantic propagation reasoning feature vector to obtain the treatment multimodal-limb volume weight value; Inputting the treatment multimodality-limb volume weight value into a softmax function to obtain a treatment multimodality-limb volume activation weight value; Using the treatment multimodal-limb volume activation weight value as a weight value, weighting the transposed vector of the third weight vector of the limb volume temporal semantic propagation reasoning to obtain the enhanced limb volume temporal semantic propagation reasoning feature vector containing the treatment multimodal semantic-limb volume temporal reasoning semantic interaction information; Wherein, the interactive information fusion response gate value calculation unit is used to: After calculating the matrix product of the intensive therapy multimodal semantic-limb volume interaction fusion feature vector and the parameter matrix, the obtained feature vector and the bias vector are added according to the position to obtain the intensive therapy multimodal semantic-limb volume interaction fusion bias feature vector; The intensive therapy multimodal semantic-limb volume interaction fusion bias feature vector is input into a sigmoid activation function to obtain the intensive therapy multimodal semantic-limb volume interaction information fusion response gating value.

2. The lymphedema limb volume measurement data management system according to claim 1, characterized in that: The abnormality detection result generating module is used to: The treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion feature vector is passed through a classifier-based lymphedema limb volume abnormality identifier to obtain the abnormality detection result, and the abnormality detection result is used to indicate whether the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal; In response to the abnormal detection result that the volume of the lymphedema limb of the monitored lymphedema limb patient is abnormal, an early warning signal is generated to the medical staff.

3. A method for managing lymphedema limb volume measurement data, using the lymphedema limb volume measurement data management system according to claim 1, characterized in that: include: Obtaining a time series of lymphedema limb volume data of a patient with a monitored lymphedema limb; Acquiring treatment data of the monitored lymphedema limb patient, wherein the treatment data includes physical therapy data and drug therapy data; Arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting the vector into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors; Passing the sequence of the limb volume local temporal pattern feature vectors through a limb volume temporal node semantic propagation reasoner based on a node energy decay mechanism to obtain a limb volume temporal semantic propagation reasoning feature vector; After semantically encoding the treatment data of the monitored lymphedema limb patient to obtain a physical therapy data semantic encoding feature vector and a drug therapy data semantic encoding feature vector, the physical therapy data semantic encoding feature vector and the drug therapy data semantic encoding feature vector are fused to obtain a treatment multimodal semantic fusion feature vector; The treatment multimodal semantic fusion feature vector and the limb volume temporal semantic propagation reasoning feature vector are passed through a dynamic feature selection interactive fusion module based on a bidirectional attention network to obtain a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature vector as a treatment multimodal semantic-limb volume temporal reasoning semantic dynamic selection interactive fusion feature; Based on the treatment multimodal semantics-limb volume temporal reasoning semantics dynamic selection interactive fusion features, an abnormal detection result is obtained, and based on the abnormal detection result, a corresponding early warning prompt signal is generated to the medical staff; The method of arranging the time series of the lymphedema limb volume data into a lymphedema limb volume time series input vector and then inputting the vector into a lymphedema limb volume time series pattern feature extractor to obtain a sequence of limb volume local time series pattern feature vectors includes: Arranging the time series of the lymphedema limb volume data according to the time dimension into the lymphedema limb volume time series input vector; The lymphedema limb volume temporal input vector is input into a lymphedema limb volume temporal pattern feature extractor based on a 1D convolutional neural network model to obtain a sequence of the limb volume local temporal pattern feature vectors.