CN117556220B - Intelligent auxiliary system and method for rehabilitation nursing - Google Patents
Intelligent auxiliary system and method for rehabilitation nursing Download PDFInfo
- Publication number
- CN117556220B CN117556220B CN202410028392.6A CN202410028392A CN117556220B CN 117556220 B CN117556220 B CN 117556220B CN 202410028392 A CN202410028392 A CN 202410028392A CN 117556220 B CN117556220 B CN 117556220B
- Authority
- CN
- China
- Prior art keywords
- motion state
- sequence
- physiological
- training
- time sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000000474 nursing effect Effects 0.000 title claims abstract description 24
- 230000033001 locomotion Effects 0.000 claims description 303
- 239000013598 vector Substances 0.000 claims description 188
- 230000035790 physiological processes and functions Effects 0.000 claims description 93
- 239000011159 matrix material Substances 0.000 claims description 45
- 238000004458 analytical method Methods 0.000 claims description 35
- 230000004927 fusion Effects 0.000 claims description 13
- 238000012300 Sequence Analysis Methods 0.000 claims description 10
- 238000012098 association analyses Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 8
- 238000010219 correlation analysis Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 14
- 230000036541 health Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 10
- 230000008859 change Effects 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000036772 blood pressure Effects 0.000 description 3
- 230000005802 health problem Effects 0.000 description 3
- 230000036387 respiratory rate Effects 0.000 description 3
- 238000002560 therapeutic procedure Methods 0.000 description 3
- 230000036760 body temperature Effects 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000004064 dysfunction Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000554 physical therapy Methods 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biophysics (AREA)
- Physiology (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Cardiology (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Pulmonology (AREA)
- Physical Education & Sports Medicine (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses an intelligent auxiliary system and method for rehabilitation nursing, and relates to the field of rehabilitation nursing. In this way, the health care personnel can better monitor and manage the rehabilitation process of the patient.
Description
Technical Field
The present application relates to the field of rehabilitation care, and more particularly, to an intelligent assistance system and method for rehabilitation care.
Background
Rehabilitation care is a medical service for helping patients with physical dysfunction caused by diseases, disabilities or aging and the like recover or improve the quality of life through professional assessment, training and guidance. The aim of rehabilitation nursing is to enable patients to achieve better functional states, reduce the occurrence of complications and improve self-care ability and social participation. Rehabilitation care involves the collaboration of multiple disciplines including physical therapy, functional therapy, linguistic therapy, psychological therapy, and the like.
In the process of performing rehabilitation care, the rehabilitation process and progress of each patient may be different, and thus, the rehabilitation care process needs to make a personalized rehabilitation scheme according to individual conditions of the patient, and adjust according to feedback and progress of the patient to meet specific requirements thereof. However, conventional rehabilitation and nursing systems are generally based on generalized rehabilitation schemes, and cannot sufficiently consider individual differences and special needs of patients. In addition, traditional rehabilitation care generally relies on subjective feedback of a patient to evaluate rehabilitation progress, however, subjective feeling of the patient may have errors or subjective bias, accurate information cannot be provided for medical staff to judge and adjust a rehabilitation scheme, so that rehabilitation progress of the patient may be delayed or insufficient, and meanwhile, the problem of patient condition cannot be monitored in real time in the mode, and potential health problems or rehabilitation risks cannot be found in time.
Accordingly, an intelligent assistance system for rehabilitation care is desired.
Disclosure of Invention
In view of this, the present application proposes an intelligent assistance system and method for rehabilitation care, which can ensure the rehabilitation safety and quality of a patient by adjusting rehabilitation exercise data, so that medical staff can monitor and manage the rehabilitation process of the patient better.
According to an aspect of the present application, there is provided an intelligent assistance system for rehabilitation care, comprising:
a data acquisition module for acquiring physiological data and motion data of a monitored patient acquired by the rehabilitation care auxiliary device at a plurality of predetermined time points within a predetermined time period;
the physiological data time sequence arrangement module is used for arranging the physiological data of the plurality of preset time points into a physiological data time sequence matrix according to the time dimension and the physiological sample dimension;
the motion data time sequence coding module is used for respectively coding the motion data of the plurality of preset time points to obtain a sequence of motion data coding vectors;
the motion state feature global association analysis module is used for carrying out context global motion state association feature analysis on the sequence of the motion data coding vectors to obtain a sequence of context motion state feature vectors;
The physiological state embedded motion state time sequence analysis module is used for carrying out embedded association feature analysis on the sequence of the context motion state feature vector and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features; and
and the rehabilitation scheme adjusting module is used for determining whether the rehabilitation scheme of the monitored patient needs to be adjusted or not based on the physiological state embedded motion state time sequence characteristics.
Further, the motion state feature global association analysis module is configured to: the sequence of motion data encoding vectors is passed through a converter-based motion state context encoder to obtain the sequence of contextual motion state feature vectors.
Further, the physiological state embedding motion state time sequence analysis module is used for: and the sequence of the context motion state characteristic vector and the physiological data time sequence matrix are processed through a characteristic embedding module to obtain a physiological state embedded motion state time sequence characteristic vector which is used as the physiological state embedded motion state time sequence characteristic.
Further, the physiological state embedded motion state timing analysis module comprises:
the full convolution feature extraction unit is used for enabling the physiological data time sequence matrix to pass through a feature extractor based on a full convolution network model so as to obtain physiological data time sequence feature vectors;
The primary linear processing unit is used for carrying out linear processing on the physiological data time sequence feature vector so as to obtain a physiological data time sequence feature vector after linear processing;
the secondary linear processing unit is used for carrying out linear processing on the sequence of the context motion state characteristic vectors so as to obtain a sequence of the context motion state characteristic vectors after linear processing;
the linear fusion unit is used for fusing the sequence of the physiological data time sequence characteristic vector after linear processing and the context motion state characteristic vector after linear processing to obtain a physiological data-motion state linear primary fusion vector;
the one-dimensional convolution unit is used for carrying out one-dimensional convolution processing on the sequence of the context motion state feature vector to obtain a sequence of context motion state time sequence neighborhood associated feature vector; and
and the splicing and fusing unit is used for fusing the sequence of the context motion state time sequence neighborhood associated feature vector and the physiological data-motion state linear primary fusion vector based on a splicing mode to obtain the physiological state embedded motion state time sequence feature vector.
Further, the rehabilitation regimen adjustment module is configured to: and embedding the physiological state into the time sequence feature vector of the motion state through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rehabilitation scheme of the monitored patient needs to be adjusted.
Further, a training module for training the transducer-based motion state context encoder, the feature embedding module, and the classifier is also included.
Further, the training module includes:
the system comprises a training data acquisition unit, a rehabilitation nursing auxiliary device and a rehabilitation data acquisition unit, wherein the training data acquisition unit is used for acquiring training data, the training data comprises training physiological data and training motion data which are acquired by the rehabilitation nursing auxiliary device and used for training a monitored patient at a plurality of preset time points in a preset time period, and whether a rehabilitation scheme of the monitored patient needs to be adjusted or not is judged;
the training physiological data time sequence arrangement unit is used for arranging training physiological data of the plurality of preset time points into a training physiological data time sequence matrix according to the time dimension and the physiological sample dimension;
the training motion data time sequence coding unit is used for respectively coding the training motion data of the plurality of preset time points to obtain a sequence of training motion data coding vectors;
the training motion state feature global association analysis unit is used for carrying out context global motion state association feature analysis on the sequence of training motion data coding vectors to obtain a sequence of training context motion state feature vectors;
The training physiological state embedded motion state time sequence analysis unit is used for carrying out embedded association feature analysis on the sequence of the training context motion state feature vector and the training physiological data time sequence matrix through the feature embedding module so as to obtain a training physiological state embedded motion state time sequence feature vector;
the loss value calculation unit is used for embedding the training physiological state into the time sequence feature vector of the motion state and passing through the classifier to obtain a classification loss function value; and
and the loss training unit is used for training the motion state context encoder, the characteristic embedding module and the classifier based on the classification loss function value and through gradient descent direction propagation, wherein the training physiological state embedding motion state time sequence characteristic vector is subjected to training optimization during each training weight matrix iteration.
Further, the loss training unit is configured to: processing the training physiological state embedded motion state time sequence feature vector by using the classifier according to the following training classification formula to obtain a training classification result; wherein, training classification formula is:
The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>To->Is a weight matrix>To->As a result of the offset vector,embedding a motion state time sequence feature vector for the training physiological state; and calculating a cross entropy value between the training classification result and the true value as the classification loss function value.
According to another aspect of the present application, there is provided an intelligent assistance method for rehabilitation care, comprising:
acquiring physiological data and motion data of a monitored patient acquired by a rehabilitation care auxiliary device at a plurality of predetermined time points within a predetermined time period;
arranging the physiological data of the plurality of preset time points into a physiological data time sequence matrix according to a time dimension and a physiological sample dimension;
encoding the motion data of the plurality of predetermined time points respectively to obtain a sequence of motion data encoding vectors;
performing context global motion state association feature analysis on the sequence of motion data coding vectors to obtain a sequence of context motion state feature vectors;
performing embedded association feature analysis on the sequence of the context motion state feature vectors and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features; and
Based on the physiological state embedded motion state timing characteristics, it is determined whether a rehabilitation regimen of the monitored patient requires adjustment.
Further, performing a contextual global motion state association feature analysis on the sequence of motion data encoding vectors to obtain a sequence of contextual motion state feature vectors, comprising: the sequence of motion data encoding vectors is passed through a converter-based motion state context encoder to obtain the sequence of contextual motion state feature vectors.
Firstly, physiological data of a plurality of preset time points are arranged into a physiological data time sequence matrix according to a time dimension and a physiological sample dimension, then, the motion data of the preset time points are respectively encoded to obtain a sequence of motion data encoding vectors, then, context global motion state association feature analysis is carried out on the sequence of the motion data encoding vectors to obtain a sequence of context motion state feature vectors, then, embedded association feature analysis is carried out on the sequence of the context motion state feature vectors and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features, and finally, whether a rehabilitation scheme of a monitored patient needs to be adjusted is determined based on the physiological state embedded motion state time sequence features. In this way, the health care personnel can better monitor and manage the rehabilitation process of the patient.
Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present application and together with the description, serve to explain the principles of the present application.
Fig. 1 shows a block diagram of an intelligent assistance system for rehabilitation care according to an embodiment of the present application.
Fig. 2 shows a flow chart of an intelligent assistance method for rehabilitation care according to an embodiment of the present application.
Fig. 3 shows an architectural schematic diagram of an intelligent assistance method for rehabilitation care according to an embodiment of the present application.
Fig. 4 shows an application scenario diagram of an intelligent assistance system for rehabilitation care according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, are also within the scope of the present application.
As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present application. It will be understood by those skilled in the art that the present application may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits have not been described in detail as not to unnecessarily obscure the present application.
According to the technical scheme, the intelligent auxiliary system for rehabilitation nursing is provided, physiological data and motion data of a patient can be monitored and collected in real time through the rehabilitation nursing auxiliary device, and a data processing and analyzing algorithm is introduced into the rear end to conduct time sequence collaborative analysis of the physiological data and the motion data of the patient, so that whether the rehabilitation scheme of the monitored patient needs to be adjusted or not is judged. By the method, real-time feedback and guidance, such as automatic adjustment of rehabilitation schemes and parameters, can be provided according to the rehabilitation demands and the progress of patients, and medical staff can be helped to find abnormal conditions in time and take corresponding measures, such as adjusting rehabilitation exercise data to ensure the rehabilitation safety and quality of the patients, so that the medical staff can monitor and manage the rehabilitation process of the patients better.
Fig. 1 shows a block diagram schematic of an intelligent assistance system for rehabilitation care according to an embodiment of the present application. As shown in fig. 1, an intelligent assistance system 100 for rehabilitation care according to an embodiment of the present application includes: a data acquisition module 110 for acquiring physiological data and motion data of a monitored patient acquired by the rehabilitation and care assistance device at a plurality of predetermined time points within a predetermined time period; a physiological data timing arrangement module 120, configured to arrange physiological data of the plurality of predetermined time points into a physiological data timing matrix according to a time dimension and a physiological sample dimension; a motion data timing encoding module 130, configured to encode the motion data at the plurality of predetermined time points respectively to obtain a sequence of motion data encoding vectors; the motion state feature global association analysis module 140 is configured to perform a contextual global motion state association feature analysis on the sequence of motion data encoding vectors to obtain a sequence of contextual motion state feature vectors; the physiological state embedded motion state time sequence analysis module 150 is configured to perform embedded association feature analysis on the sequence of the contextual motion state feature vectors and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features; and a rehabilitation regimen adjustment module 160 for determining whether the rehabilitation regimen of the monitored patient needs adjustment based on the physiological state embedded motion state timing characteristics.
It should be appreciated that the data acquisition module 110 is responsible for collecting physiological and exercise data of the patient from the rehabilitation care assistance device. The physiological data timing module 120 sequences the collected physiological data in a time sequence and with different physiological parameters for subsequent analysis and processing. The motion data timing encoding module 130 converts the collected motion data into a form of encoded vectors for subsequent analysis and processing. The motion state feature global correlation analysis module 140 analyzes correlation features between motion data to obtain a sequence of feature vectors describing motion states. The physiological state embedded motion state timing analysis module 150 performs a correlation analysis on the physiological data and the motion state characteristics to obtain a physiological state and motion state embedded timing characteristic. The rehabilitation regimen adjustment module 160 evaluates the rehabilitation progress of the patient using the analyzed physiological state and movement state timing characteristics and proposes adjustment advice for the rehabilitation regimen as needed. These modules together comprise the intelligent auxiliary system 100 for rehabilitation care, which implements the functions of collecting, analyzing, and adjusting rehabilitation regimen for physiological data and motion data of a patient, so as to provide better rehabilitation care services.
Specifically, in the technical solution of the present application, first, physiological data and movement data of a monitored patient at a plurality of predetermined time points within a predetermined period of time acquired by a rehabilitation care auxiliary device are acquired. In particular, in one specific example of the present application, the physiological data may be heart rate, blood pressure, respiratory rate, body temperature, etc. physiological parameters of the patient, which may provide an assessment of the patient's physiological state and progress of rehabilitation and help healthcare workers find potential health problems in time; the exercise data may be the number of steps, range of motion, posture, etc. of the patient during rehabilitation training, which may be used to assess the exercise capacity and function of the patient and adjust the rehabilitation regimen as desired.
Next, considering that physiological data of the patient, including heart rate, blood pressure, respiratory rate and body temperature data, have respective time sequence dynamic change rules in a time dimension, and have time sequence cooperative association relationships between the physiological data, the association relationships can better reflect the physiological state changes of the patient, so as to be used for evaluating the physiological state and rehabilitation progress of the patient, and can also be used for helping medical staff to find potential health problems in time. Therefore, in the technical solution of the present application, in order to facilitate the time-sequence collaborative analysis of each data item in the physiological status data, the physiological data of the plurality of predetermined time points needs to be arranged into a physiological data time sequence matrix according to a time dimension and a physiological sample dimension, so as to integrate the distribution information of each data item in the physiological data of the patient in the sample and the time dimension.
It should be appreciated that the motion data includes a plurality of motion data items of the patient, including a step number, a motion range, a posture, etc., and in order to enable time-series collaborative association analysis of the motion data items, time-series encoding of each motion data item is required. Therefore, in the technical solution of the present application, the motion data at the plurality of predetermined time points are encoded respectively to obtain a sequence of motion data encoding vectors. By encoding the motion data of the plurality of preset time points, motion state time sequence characteristic information such as a motion mode, a change trend, motion capacity and the like of a patient can be captured. This helps the healthcare staff to understand the patient's progress and ability to exercise and to adjust and guide the rehabilitation regimen accordingly to the patient's needs.
Then, since the patient's rehabilitation exercise is a dynamic process, the patient's state of motion may change over time. In addition, each time sequence characteristic of each motion data item in the motion data has an association relationship with each other, so in order to better understand the motion behavior and the change trend of a patient, in the technical scheme of the application, the sequence of the motion data coding vectors is further coded in a motion state context coder based on a converter so as to extract the time sequence coding characteristic information based on the global context between the time sequence coding characteristics of each data item in the motion data, thereby obtaining the sequence of the context motion state characteristic vectors.
Accordingly, the motion state feature global association analysis module 140 is configured to: the sequence of motion data encoding vectors is passed through a converter-based motion state context encoder to obtain the sequence of contextual motion state feature vectors.
It is worth mentioning that the converter-based motion state context encoder is used for converting a sequence of motion data encoding vectors into a sequence of context motion state feature vectors. In particular, a converter refers to a component or model for converting input data from one representation to another, and in a motion state feature global correlation analysis module, the converter is used to convert motion data encoding vectors to obtain contextual motion state feature vectors. The specific implementation of the converter may be various machine learning models, such as neural networks, self-encoders, variational self-encoders, and the like. These models can transform input data into more meaningful or useful representations by learning the inherent structural and characteristic representations of the data. In the sequence of context motion state feature vectors, the role of the converter is to extract and encode context features of the global motion state by converting motion data encoding vectors. These contextual motion state feature vectors may capture associations and patterns between motion data, providing more information for subsequent analysis and decision making. Therefore, the converter plays a key role in the motion state characteristic global association analysis module, and helps to extract and represent the context characteristics of the global motion state by converting motion data coding vectors, so that more accurate and comprehensive information is provided for subsequent analysis and decision.
Further, it is contemplated that both physiological data and movement data contain information important to the patient's recovery. In particular, physiological data may provide an assessment of a patient's physiological state, such as heart rate, blood pressure, respiratory rate, etc. While the sequence of contextual movement state feature vectors may reflect the patient's motor ability and rehabilitation progress. By fusing the two, the overall state of the patient can be comprehensively evaluated, so that the rehabilitation condition of the patient can be more accurately known, and the rehabilitation scheme can be adjusted according to the requirement. Based on the above, in the technical solution of the present application, the sequence of the context motion state feature vector and the physiological data time sequence matrix are passed through a feature embedding module to obtain a physiological state embedded motion state time sequence feature vector. The physiological state time sequence collaborative full-connection associated feature information can be embedded into time sequence context associated features of the motion data through the processing of the feature embedding module, and in such a way, the time sequence feature information of the physiological state time sequence collaborative full-connection associated feature information and the time sequence feature information of the physiological state collaborative full-connection associated feature information can be mutually supplemented and combined, so that more comprehensive and comprehensive feature representation is provided, the medical staff can better understand the rehabilitation process and state change of a patient, and the physiological state collaborative full-connection associated feature information is adjusted and guided according to the specific situation and the requirement of the patient.
Accordingly, the physiological state embedding motion state timing analysis module 150 is configured to: and the sequence of the context motion state characteristic vector and the physiological data time sequence matrix pass through a characteristic embedding module to obtain a physiological state embedded motion state time sequence characteristic vector serving as the physiological state embedded motion state time sequence characteristic.
It is worth mentioning that the feature embedding module is used for feature embedding the sequence of the context motion state feature vector and the physiological data time sequence matrix to obtain the physiological state embedded motion state time sequence feature vector as the physiological state embedded motion state time sequence feature. The feature embedding module refers to a component or model that maps input data to a new feature space, which converts the original data into feature vectors with more expressive and discriminant capabilities by learning a feature representation of the data. In the physiological state embedding motion state time sequence analysis module, a feature embedding module is used for feature embedding the context motion state feature vector and the physiological data time sequence matrix. This means that it will combine the motion state features and the physiological data to generate a new set of feature vectors that better represent the time series features of the physiological state embedded in the motion state by learning the correlations and patterns between the data.
Specifically, the physiological state embedded motion state timing analysis module 150 includes: the full convolution feature extraction unit is used for enabling the physiological data time sequence matrix to pass through a feature extractor based on a full convolution network model so as to obtain physiological data time sequence feature vectors; the primary linear processing unit is used for carrying out linear processing on the physiological data time sequence feature vector so as to obtain a physiological data time sequence feature vector after linear processing; the secondary linear processing unit is used for carrying out linear processing on the sequence of the context motion state characteristic vectors so as to obtain a sequence of the context motion state characteristic vectors after linear processing; the linear fusion unit is used for fusing the sequence of the physiological data time sequence characteristic vector after linear processing and the context motion state characteristic vector after linear processing to obtain a physiological data-motion state linear primary fusion vector; the one-dimensional convolution unit is used for carrying out one-dimensional convolution processing on the sequence of the context motion state feature vector to obtain a sequence of context motion state time sequence neighborhood associated feature vector; and a splicing and fusing unit, configured to fuse the sequence of the context motion state timing sequence neighborhood associated feature vector and the physiological data-motion state linear primary fusion vector based on a splicing manner to obtain the physiological state embedded motion state timing feature vector.
And then, embedding the physiological state into the time sequence feature vector of the motion state through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rehabilitation scheme of the monitored patient needs to be adjusted. That is, the physiological state features of the patient are embedded into the fusion feature information of the motion state features to perform classification processing, so as to determine whether the rehabilitation scheme of the monitored patient needs to be adjusted, in this way, real-time feedback and guidance, such as automatic adjustment of the rehabilitation scheme and parameters, can be provided according to the rehabilitation requirement and progress of the patient, and help medical staff to find abnormal conditions in time and take corresponding measures, such as adjusting rehabilitation motion data to ensure the rehabilitation safety and quality of the patient, so that the medical staff can monitor and manage the rehabilitation process of the patient better.
Accordingly, the rehabilitation regimen adjustment module 160 is configured to: and embedding the physiological state into the time sequence feature vector of the motion state through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rehabilitation scheme of the monitored patient needs to be adjusted.
Specifically, the physiological state is embedded into a time sequence feature vector of the motion state to obtain a classification result through a classifier, wherein the classification result is used for indicating whether a rehabilitation scheme of a monitored patient needs to be adjusted or not, and the method comprises the following steps of: performing full-connection coding on the physiological state embedded motion state time sequence feature vector by using a full-connection layer of the classifier to obtain a coding classification feature vector; and inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
It should be appreciated that the role of the classifier is to learn the classification rules and classifier using a given class, known training data, and then classify (or predict) the unknown data. Logistic regression (logistics), SVM, etc. are commonly used to solve the classification problem, and for multi-classification problems (multi-class classification), logistic regression or SVM can be used as well, but multiple bi-classifications are required to compose multiple classifications, but this is error-prone and inefficient, and the commonly used multi-classification method is the Softmax classification function.
Further, in the technical solution of the present application, the intelligent auxiliary system for rehabilitation and nursing further includes a training module for training the converter-based motion state context encoder, the feature embedding module and the classifier. It should be appreciated that the training module plays a vital role in intelligent assistance systems for rehabilitation care, which are used to train transducer-based motion state context encoders, feature embedding modules, and classifiers. The main purpose of the training module is to train the individual modules in the system by using the existing marker data set so that they can learn and adapt to specific rehabilitation care tasks. By training, the modules can automatically learn key features and modes in the rehabilitation and nursing process from the data, so that the performance and accuracy of the system are improved. Specifically, the training module functions as follows: 1. training of converter-based motion state context encoders: the training module uses the labeled motion data to train a converter model that is capable of converting motion data encoding vectors into a sequence of contextual motion state feature vectors. Through training, the converter can learn the inherent structure and association of the motion data, thereby extracting useful context information. 2. Training of a feature embedding module: the training module trains the feature embedding module using the labeled contextual motion state feature vectors and the physiological data timing matrix. Through training, the feature embedding module can learn how to effectively perform feature fusion and embedding on the contextual movement state features and the physiological data so as to generate a physiological state embedded movement state time sequence feature vector with more expressive ability. 3. Training of a classifier: the training module uses the labeled rehabilitation and care task related data to train the classifier model. The classifier embeds the time sequence feature vector of the motion state according to the input physiological state to classify or predict the rehabilitation nursing task. Through training, the classifier can learn distinguishing features and modes among different rehabilitation tasks, so that accurate classification and prediction are realized. In summary, the training module trains each module in the system by using the marked data, so that the training module can learn the relevant characteristics and modes of the rehabilitation nursing task, and the performance and effect of the system are improved. The training module is a key component of the intelligent assistance system, which ensures that the system is able to accommodate specific rehabilitation care needs and provides accurate assistance and decision support.
In one example, the training module includes: the system comprises a training data acquisition unit, a rehabilitation nursing auxiliary device and a rehabilitation data acquisition unit, wherein the training data acquisition unit is used for acquiring training data, the training data comprises training physiological data and training motion data which are acquired by the rehabilitation nursing auxiliary device and used for training a monitored patient at a plurality of preset time points in a preset time period, and whether a rehabilitation scheme of the monitored patient needs to be adjusted or not is judged; the training physiological data time sequence arrangement unit is used for arranging training physiological data of the plurality of preset time points into a training physiological data time sequence matrix according to the time dimension and the physiological sample dimension; the training motion data time sequence coding unit is used for respectively coding the training motion data of the plurality of preset time points to obtain a sequence of training motion data coding vectors; the training motion state feature global association analysis unit is used for carrying out context global motion state association feature analysis on the sequence of training motion data coding vectors to obtain a sequence of training context motion state feature vectors; the training physiological state embedded motion state time sequence analysis unit is used for carrying out embedded association feature analysis on the sequence of the training context motion state feature vector and the training physiological data time sequence matrix through the feature embedding module so as to obtain a training physiological state embedded motion state time sequence feature vector; the loss value calculation unit is used for embedding the training physiological state into the time sequence feature vector of the motion state and passing through the classifier to obtain a classification loss function value; and a loss training unit for training the converter-based motion state context encoder, the feature embedding module and the classifier based on the classification loss function value and propagating through a gradient descent direction, wherein the training physiological state embedded motion state time sequence feature vector is subjected to training optimization at each weight matrix iteration of the training.
Wherein, the loss training unit is used for: processing the training physiological state embedded motion state time sequence feature vector by using the classifier according to the following training classification formula to obtain a training classification result; wherein, training classification formula is:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>To->Is a weight matrix>To->As a result of the offset vector,embedding a motion state time sequence feature vector for the training physiological state; and calculating a cross entropy value between the training classification result and the true value as the classification loss function value.
In particular, in the above technical solution, the sequence expression of the training context motion state feature vector encodes the motion state feature of the training motion data based on the time sequence context association, so that after the sequence of the training context motion state feature vector and the training physiological data time sequence matrix pass through the feature embedding module, the feature expression of the sequence of the training context motion state feature vector is constrained based on the sample-time sequence cross dimension local association feature expression of the training physiological data time sequence matrix, and thus, considering the heterogeneity of the training physiological data and the training motion data and the time sequence associated dimension difference of feature extraction, the overall feature distribution of the training physiological state embedded motion state time sequence feature vector has more remarkable inconsistency and instability, thereby affecting the stability of the classification training by the classifier.
Based on the above, when the training physiological state embedded motion state time sequence feature vector is classified and trained through the classifier, the applicant of the application performs training optimization on the training physiological state embedded motion state time sequence feature vector at each iteration.
Accordingly, in one example, training optimization of the training physiological state embedded motion state timing feature vector at each weight matrix iteration of the training includes: training and optimizing the training physiological state embedded motion state time sequence feature vector by using the following optimization formula to obtain an optimized training physiological state embedded motion state time sequence feature vector; wherein, the optimization formula is:
；
wherein,is the training physiological state embedded motion state time sequence feature vector,>is the training physiological state embedded motion state time sequence feature vector +.>Characteristic value of>And->Respectively, when the training physiological state is embedded into the motion stateSequence feature vector->1-norm and 2-norm of +.>Is the time sequence characteristic vector of the training physiological state embedded motion stateLength of (2), and->Is in combination with->Related weight superparameter +.>An exponential operation representing a value of a natural exponential function value raised to a power by the value, + >Is the time sequence characteristic vector of the training physiological state embedded motion state after optimization.
Here, the motion state time sequence feature vector is embedded through the training physiological stateStructural consistency and stability representation of the global feature distribution of (a) under rigid and non-rigid structures of absolute and spatial distances, respectively, such that the training physiological state is embedded in the motion state temporal feature vector +.>Has a certain repeatability for local mode change to embed motion state time sequence feature vector in the training physiological state>When classifying by the classifier, the global feature distribution is classified by the classifierThe scale and rotation change of the weight matrix of the training device has robustness, and the stability of classification training is improved. Thus, real-time feedback and guidance can be provided according to the rehabilitation demands and the progress of the patient, and medical staff is helped to find abnormal conditions in time and take corresponding measures, so that the medical staff can monitor and manage the rehabilitation process of the patient better, and the safety and the quality of rehabilitation nursing of the patient are ensured.
In summary, an intelligent assistance system 100 for rehabilitation care is illustrated that may enable a healthcare worker to better monitor and manage a patient's rehabilitation process, in accordance with embodiments of the present application.
As described above, the intelligent assistance system 100 for rehabilitation nursing according to the embodiment of the present application may be implemented in various terminal devices, for example, a server or the like having an intelligent assistance algorithm for rehabilitation nursing. In one example, the intelligent assistance system 100 for rehabilitation care may be integrated into the terminal device as one software module and/or hardware module. For example, the intelligent assistance system 100 for rehabilitation care may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the intelligent assistance system 100 for rehabilitation care may also be one of a number of hardware modules of the terminal device.
Alternatively, in another example, the intelligent assistance system for rehabilitation care 100 and the terminal device may be separate devices, and the intelligent assistance system for rehabilitation care 100 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a contracted data format.
Fig. 2 shows a flow chart of an intelligent assistance method for rehabilitation care according to an embodiment of the present application. Fig. 3 shows a schematic diagram of a system architecture of an intelligent assistance method for rehabilitation care according to an embodiment of the present application. As shown in fig. 2 and 3, an intelligent assistance method for rehabilitation care according to an embodiment of the present application includes: s110, acquiring physiological data and motion data of a monitored patient at a plurality of preset time points in a preset time period, wherein the physiological data and the motion data are acquired by a rehabilitation nursing auxiliary device; s120, arranging the physiological data of the plurality of preset time points into a physiological data time sequence matrix according to a time dimension and a physiological sample dimension; s130, respectively encoding the motion data of the plurality of preset time points to obtain a sequence of motion data encoding vectors; s140, performing context global motion state association feature analysis on the sequence of motion data coding vectors to obtain a sequence of context motion state feature vectors; s150, carrying out embedded association feature analysis on the sequence of the context motion state feature vector and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features; and S160, determining whether the rehabilitation regimen of the monitored patient needs to be adjusted based on the physiological state embedded motion state time sequence feature.
In one possible implementation, performing a contextual global motion state association feature analysis on the sequence of motion data encoding vectors to obtain a sequence of contextual motion state feature vectors, includes: the sequence of motion data encoding vectors is passed through a converter-based motion state context encoder to obtain the sequence of contextual motion state feature vectors.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described intelligent assistance method for rehabilitation nursing have been described in detail in the above description of the intelligent assistance system for rehabilitation nursing with reference to fig. 1, and thus, repetitive descriptions thereof will be omitted.
Fig. 4 shows an application scenario diagram of an intelligent assistance system for rehabilitation care according to an embodiment of the present application. As shown in fig. 4, in this application scenario, first, physiological data (e.g., D1 illustrated in fig. 4) and motion data (e.g., D2 illustrated in fig. 4) of a plurality of predetermined time points of a monitored patient acquired by a rehabilitation care assistance device within a predetermined period of time are acquired, and then the physiological data and the motion data of the plurality of predetermined time points are input into a server (e.g., S illustrated in fig. 4) where an intelligent assistance algorithm for rehabilitation care is deployed, wherein the server can process the physiological data and the motion data of the plurality of predetermined time points using the intelligent assistance algorithm for rehabilitation care to obtain a classification result for indicating whether the rehabilitation regimen of the monitored patient needs to be adjusted.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The embodiments of the present application have been described above, the foregoing description is exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. An intelligent assistance system for rehabilitation and care comprising:
a data acquisition module for acquiring physiological data and motion data of a monitored patient acquired by the rehabilitation care auxiliary device at a plurality of predetermined time points within a predetermined time period;
the physiological data time sequence arrangement module is used for arranging the physiological data of the plurality of preset time points into a physiological data time sequence matrix according to the time dimension and the physiological sample dimension;
the motion data time sequence coding module is used for respectively coding the motion data of the plurality of preset time points to obtain a sequence of motion data coding vectors;
the motion state feature global association analysis module is used for carrying out context global motion state association feature analysis on the sequence of the motion data coding vectors to obtain a sequence of context motion state feature vectors;
the physiological state embedded motion state time sequence analysis module is used for carrying out embedded association feature analysis on the sequence of the context motion state feature vector and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features;
the rehabilitation scheme adjusting module is used for determining whether the rehabilitation scheme of the monitored patient needs to be adjusted or not based on the physiological state embedded motion state time sequence characteristics;
Wherein, the physiological state is embedded into a motion state time sequence analysis module for: the sequence of the context motion state characteristic vectors and the physiological data time sequence matrix are processed through a characteristic embedding module to obtain physiological state embedded motion state time sequence characteristic vectors which are used as the physiological state embedded motion state time sequence characteristics;
wherein, the physiological state embedding motion state time sequence analysis module includes:
the full convolution feature extraction unit is used for enabling the physiological data time sequence matrix to pass through a feature extractor based on a full convolution network model so as to obtain physiological data time sequence feature vectors;
the primary linear processing unit is used for carrying out linear processing on the physiological data time sequence feature vector so as to obtain a physiological data time sequence feature vector after linear processing;
the secondary linear processing unit is used for carrying out linear processing on the sequence of the context motion state characteristic vectors so as to obtain a sequence of the context motion state characteristic vectors after linear processing;
the linear fusion unit is used for fusing the sequence of the physiological data time sequence characteristic vector after linear processing and the context motion state characteristic vector after linear processing to obtain a physiological data-motion state linear primary fusion vector;
The one-dimensional convolution unit is used for carrying out one-dimensional convolution processing on the sequence of the context motion state feature vector to obtain a sequence of context motion state time sequence neighborhood associated feature vector;
and the splicing and fusing unit is used for fusing the sequence of the context motion state time sequence neighborhood associated feature vector and the physiological data-motion state linear primary fusion vector based on a splicing mode to obtain the physiological state embedded motion state time sequence feature vector.
2. The intelligent assistance system for rehabilitation care according to claim 1, wherein the motion state feature global correlation analysis module is configured to: the sequence of motion data encoding vectors is passed through a converter-based motion state context encoder to obtain the sequence of contextual motion state feature vectors.
3. The intelligent assistance system for rehabilitation care of claim 2, wherein the rehabilitation regimen adjustment module is configured to: and embedding the physiological state into the time sequence feature vector of the motion state through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rehabilitation scheme of the monitored patient needs to be adjusted.
4. The intelligent assistance system for rehabilitation care according to claim 3, further comprising a training module for training the transducer-based motion state context encoder, the feature embedding module, and the classifier.
5. The intelligent assistance system for rehabilitation and caretaking as defined in claim 4, wherein the training module comprises:
the system comprises a training data acquisition unit, a rehabilitation nursing auxiliary device and a rehabilitation data acquisition unit, wherein the training data acquisition unit is used for acquiring training data, the training data comprises training physiological data and training motion data which are acquired by the rehabilitation nursing auxiliary device and used for training a monitored patient at a plurality of preset time points in a preset time period, and whether a rehabilitation scheme of the monitored patient needs to be adjusted or not is judged;
the training physiological data time sequence arrangement unit is used for arranging training physiological data of the plurality of preset time points into a training physiological data time sequence matrix according to the time dimension and the physiological sample dimension;
the training motion data time sequence coding unit is used for respectively coding the training motion data of the plurality of preset time points to obtain a sequence of training motion data coding vectors;
the training motion state feature global association analysis unit is used for carrying out context global motion state association feature analysis on the sequence of training motion data coding vectors to obtain a sequence of training context motion state feature vectors;
The training physiological state embedded motion state time sequence analysis unit is used for carrying out embedded association feature analysis on the sequence of the training context motion state feature vector and the training physiological data time sequence matrix through the feature embedding module so as to obtain a training physiological state embedded motion state time sequence feature vector;
the loss value calculation unit is used for embedding the training physiological state into the time sequence feature vector of the motion state and passing through the classifier to obtain a classification loss function value;
and the loss training unit is used for training the motion state context encoder, the characteristic embedding module and the classifier based on the classification loss function value and through gradient descent direction propagation, wherein the training physiological state embedding motion state time sequence characteristic vector is subjected to training optimization during each training weight matrix iteration.
6. The intelligent assistance system for rehabilitation care according to claim 5, wherein the loss training unit is configured to: processing the training physiological state embedded motion state time sequence feature vector by using the classifier according to the following training classification formula to obtain a training classification result; wherein, training classification formula is:
The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>To->Is a weight matrix>To->For the bias vector +.>Embedding a motion state time sequence feature vector for the training physiological state; and calculating a cross entropy value between the training classification result and the true value as the classification loss function value.
7. An intelligent assistance method for rehabilitation care, comprising:
acquiring physiological data and motion data of a monitored patient acquired by a rehabilitation care auxiliary device at a plurality of predetermined time points within a predetermined time period;
arranging the physiological data of the plurality of preset time points into a physiological data time sequence matrix according to a time dimension and a physiological sample dimension;
encoding the motion data of the plurality of predetermined time points respectively to obtain a sequence of motion data encoding vectors;
performing context global motion state association feature analysis on the sequence of motion data coding vectors to obtain a sequence of context motion state feature vectors;
performing embedded association feature analysis on the sequence of the context motion state feature vectors and the physiological data time sequence matrix to obtain physiological state embedded motion state time sequence features;
determining whether a rehabilitation regimen of the monitored patient needs to be adjusted based on the physiological state embedded motion state timing characteristics;
The method for performing embedded association feature analysis on the sequence of the context motion state feature vector and the physiological data time sequence matrix to obtain the physiological state embedded motion state time sequence feature comprises the following steps: the sequence of the context motion state characteristic vectors and the physiological data time sequence matrix are processed through a characteristic embedding module to obtain physiological state embedded motion state time sequence characteristic vectors which are used as the physiological state embedded motion state time sequence characteristics;
the method for performing embedded association feature analysis on the sequence of the context motion state feature vector and the physiological data time sequence matrix to obtain the physiological state embedded motion state time sequence feature comprises the following steps:
the physiological data time sequence matrix passes through a feature extractor based on a full convolution network model to obtain physiological data time sequence feature vectors;
performing linear processing on the physiological data time sequence feature vector to obtain a physiological data time sequence feature vector after linear processing;
performing linear processing on the sequence of the context motion state feature vectors to obtain a sequence of the context motion state feature vectors after linear processing;
fusing the sequence of the physiological data time sequence characteristic vector after linear processing and the context motion state characteristic vector after linear processing to obtain a physiological data-motion state linear primary fusion vector;
Carrying out one-dimensional convolution processing on the sequence of the context motion state feature vectors to obtain a sequence of context motion state time sequence neighborhood associated feature vectors;
and fusing the sequence of the context motion state time sequence neighborhood associated feature vectors and the physiological data-motion state linear primary fusion vector based on a splicing mode to obtain the physiological state embedded motion state time sequence feature vector.
8. The intelligent assistance method for rehabilitation care according to claim 7, wherein performing a contextual global motion state association feature analysis on the sequence of motion data encoding vectors to obtain a sequence of contextual motion state feature vectors, comprises: the sequence of motion data encoding vectors is passed through a converter-based motion state context encoder to obtain the sequence of contextual motion state feature vectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410028392.6A CN117556220B (en) | 2024-01-09 | 2024-01-09 | Intelligent auxiliary system and method for rehabilitation nursing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410028392.6A CN117556220B (en) | 2024-01-09 | 2024-01-09 | Intelligent auxiliary system and method for rehabilitation nursing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117556220A CN117556220A (en) | 2024-02-13 |
| CN117556220B true CN117556220B (en) | 2024-03-22 |
Family
ID=89818851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410028392.6A Active CN117556220B (en) | 2024-01-09 | 2024-01-09 | Intelligent auxiliary system and method for rehabilitation nursing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117556220B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114359768A (en) * | 2021-09-30 | 2022-04-15 | 中远海运科技股份有限公司 | Video dense event description method based on multi-mode heterogeneous feature fusion |
| CN115510756A (en) * | 2022-10-11 | 2022-12-23 | 河南省肿瘤医院 | Postoperative auxiliary rehabilitation device and testing method thereof |
| CN116458852A (en) * | 2023-06-16 | 2023-07-21 | 山东协和学院 | Rehabilitation training system and method based on cloud platform and lower limb rehabilitation robot |
| CN117065235A (en) * | 2023-09-27 | 2023-11-17 | 郑州大学第五附属医院 | Auxiliary positioning system for radiotherapy of tumor patient |
| CN117253599A (en) * | 2023-10-02 | 2023-12-19 | 吉林大学 | Remote nursing management system and method based on data feature mining |
| CN117270611A (en) * | 2023-11-22 | 2023-12-22 | 浙江威星电子***软件股份有限公司 | Intelligent environmental control system and method for intelligent sports stadium |
| CN117274450A (en) * | 2023-11-21 | 2023-12-22 | 长春职业技术学院 | Animation image generation system and method based on artificial intelligence |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8388530B2 (en) * | 2000-05-30 | 2013-03-05 | Vladimir Shusterman | Personalized monitoring and healthcare information management using physiological basis functions |
-
2024
- 2024-01-09 CN CN202410028392.6A patent/CN117556220B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114359768A (en) * | 2021-09-30 | 2022-04-15 | 中远海运科技股份有限公司 | Video dense event description method based on multi-mode heterogeneous feature fusion |
| CN115510756A (en) * | 2022-10-11 | 2022-12-23 | 河南省肿瘤医院 | Postoperative auxiliary rehabilitation device and testing method thereof |
| CN116458852A (en) * | 2023-06-16 | 2023-07-21 | 山东协和学院 | Rehabilitation training system and method based on cloud platform and lower limb rehabilitation robot |
| CN117065235A (en) * | 2023-09-27 | 2023-11-17 | 郑州大学第五附属医院 | Auxiliary positioning system for radiotherapy of tumor patient |
| CN117253599A (en) * | 2023-10-02 | 2023-12-19 | 吉林大学 | Remote nursing management system and method based on data feature mining |
| CN117274450A (en) * | 2023-11-21 | 2023-12-22 | 长春职业技术学院 | Animation image generation system and method based on artificial intelligence |
| CN117270611A (en) * | 2023-11-22 | 2023-12-22 | 浙江威星电子***软件股份有限公司 | Intelligent environmental control system and method for intelligent sports stadium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117556220A (en) | 2024-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kidziński et al. | Deep neural networks enable quantitative movement analysis using single-camera videos | |
| CN108596902A (en) | The full reference image quality appraisement method of multitask based on gating convolutional neural networks | |
| CN117253599A (en) | Remote nursing management system and method based on data feature mining | |
| CN116072265B (en) | Sleep stage analysis system and method based on convolution of time self-attention and dynamic diagram | |
| CN113012811B (en) | Traditional Chinese medicine syndrome diagnosis and health evaluation method combining deep convolutional network and graph neural network | |
| CN117598700B (en) | Intelligent blood oxygen saturation detection system and method | |
| CN117598674B (en) | Multi-parameter heart function monitoring system and method | |
| CN117476214A (en) | Data management method and system based on hospital information | |
| CN117556220B (en) | Intelligent auxiliary system and method for rehabilitation nursing | |
| CN117198468A (en) | Intervention scheme intelligent management system based on behavior recognition and data analysis | |
| CN117137488B (en) | Auxiliary identification method for depression symptoms based on electroencephalogram data and facial expression images | |
| CN117034123B (en) | Fault monitoring system and method for fitness equipment | |
| CN116523711A (en) | Education supervision system and method based on artificial intelligence | |
| Nissimagoudar et al. | AlertNet: Deep convolutional-recurrent neural network model for driving alertness detection | |
| CN116543455A (en) | Method, equipment and medium for establishing parkinsonism gait damage assessment model and using same | |
| CN109545350A (en) | A kind of hospital guide's method and device | |
| KR102350593B1 (en) | Apparatus and method for classifying gait pattern based on multi modal sensor using deep learning ensemble | |
| CN111466877B (en) | LSTM network-based oxygen reduction state prediction method | |
| Yin et al. | Predicting Age-Related Macular Degeneration Progression with Contrastive Attention and Time-Aware LSTM | |
| CN117158904B (en) | Old people cognitive disorder detection system and method based on behavior analysis | |
| Ziaratnia et al. | Multimodal Deep Learning for Remote Stress Estimation Using CCT-LSTM | |
| CN113591886B (en) | Method, apparatus, device and computer readable storage medium for information classification | |
| CN118053584A (en) | Hepatic encephalopathy prediction model and method based on bidirectional multi-modal attention network | |
| CN117122286A (en) | Intelligent rehabilitation equipment | |
| Homlong et al. | Automated structuring of gait data for analysis purposes-A deep learning pilot example |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |