CN115937903A - Fall risk assessment method and device, electronic equipment and storage medium - Google Patents
Fall risk assessment method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115937903A CN115937903A CN202211707449.2A CN202211707449A CN115937903A CN 115937903 A CN115937903 A CN 115937903A CN 202211707449 A CN202211707449 A CN 202211707449A CN 115937903 A CN115937903 A CN 115937903A
- Authority
- CN
- China
- Prior art keywords
- risk assessment
- characteristic data
- evaluated
- sample
- data
- 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.)
- Pending
Links
Images
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The embodiment of the application provides a fall risk assessment method, a fall risk assessment device, an electronic device and a storage medium, wherein the method comprises the following steps: acquiring a radar echo signal generated by a radar signal reflected by an object to be evaluated; performing characteristic analysis on the radar echo signal to obtain target characteristic data of an object to be evaluated, wherein the target characteristic data comprises: at least one of pose feature data, velocity feature data, and sign feature data; and inputting the target characteristic data into a pre-trained falling risk assessment model, and acquiring a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed. By applying the embodiment of the application, the misjudgment rate of the fall risk assessment can be reduced, and the accuracy of the fall risk assessment is improved.
Description
Technical Field
The present application relates to the field of medical assessment technologies, and in particular, to a fall risk assessment method and apparatus, an electronic device, and a storage medium.
Background
With the increasing aging of the world population, daily care for the elderly becomes an important social need. Aging and diseases bring about the decline of the motor function of the old, so that the falling risk of the old is increased, and the falling also becomes a main reason for the old to seek medical attention due to injury. The falling risk of the old is evaluated, the old with higher falling risk is nursed in a targeted manner and relevant exercise guidance is given, the falling probability of the old is reduced, and the life quality of the old is improved.
At present, a fall risk assessment method is mainly used for assessing the fall risk of the elderly by related medical staff by using various fall risk assessment scales. However, such an assessment method is relatively dependent on the professional ability of the relevant medical staff, and if the professional ability of the relevant medical staff is insufficient, misjudgment of the fall risk is likely to be caused, so that the accuracy of the fall risk assessment is low.
Disclosure of Invention
An object of the embodiments of the present application is to provide a method and an apparatus for fall risk assessment, an electronic device, and a storage medium, so as to reduce the misjudgment rate of fall risk assessment and improve the accuracy of fall risk assessment. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present application provides a fall risk assessment method, including:
acquiring a radar echo signal generated by a radar signal reflected by an object to be evaluated;
performing feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated, wherein the target feature data comprises: at least one of pose feature data, velocity feature data, and sign feature data;
inputting the target characteristic data into a pre-trained falling risk assessment model, and acquiring a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the falling risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample labels of each group of sample characteristic data represent the preset falling risk of the sample objects with the group of sample characteristic data.
Optionally, in a specific implementation manner, the target feature data includes: the posture characteristic data, the speed characteristic data and the physical sign characteristic data;
the performing feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated includes:
performing fast Fourier transform of a slow time dimension on the radar echo signal to obtain a time-frequency graph, and determining speed characteristic data and sign characteristic data of the object to be detected based on the motion characteristic of the object to be evaluated, which is represented by the time-frequency graph;
and converting the radar echo signal into a point cloud picture, and determining attitude characteristic data of the object to be evaluated based on the human body space position information of the object to be evaluated represented by the point cloud picture.
Optionally, in a specific implementation manner, the target feature data includes: the posture characteristic data, the speed characteristic data and the physical sign characteristic data; the fall risk assessment model comprises a feature extraction layer and a result assessment layer;
the inputting the target feature data into a pre-trained fall risk assessment model and acquiring a fall risk assessment result about the object to be assessed, which is output by the fall risk assessment model, includes:
inputting the posture characteristic data, the speed characteristic data and the sign characteristic data into the falling risk assessment model, and obtaining a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the feature extraction layer is used for respectively performing high-dimensional feature extraction on the attitude feature data, the speed feature data and the physical sign feature data to obtain a plurality of high-dimensional feature information, and performing fusion calculation on the plurality of high-dimensional feature information to obtain a target high-dimensional feature; the result evaluation layer is used for learning the target high-dimensional features and outputting fall risk evaluation results related to the object to be evaluated and corresponding to the target high-dimensional features.
Optionally, in a specific implementation manner, the obtaining a radar echo signal generated by a radar signal reflected by an object to be evaluated includes:
and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal in the process of executing the specified action by the object to be evaluated.
Optionally, in a specific implementation manner, the method further includes:
and determining a target care plan corresponding to the fall risk assessment result as a care plan of the subject to be assessed according to a preset corresponding relation between the care plan and the risk assessment result.
In a second aspect, an embodiment of the present application provides a fall risk assessment apparatus, including:
the signal acquisition module is used for acquiring a radar echo signal generated by the radar signal reflected by the object to be evaluated;
a feature analysis module, configured to perform feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated, where the target feature data includes: at least one of pose feature data, velocity feature data, and sign feature data;
the risk evaluation module is used for inputting the target characteristic data into a pre-trained falling risk evaluation model and acquiring a falling risk evaluation result which is output by the falling risk evaluation model and is about the object to be evaluated; the fall risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample label of each group of sample characteristic data represents the preset fall risk of the sample object with the group of sample characteristic data.
Optionally, in a specific implementation manner, the target feature data includes: the posture characteristic data, the speed characteristic data and the physical sign characteristic data;
the feature analysis module includes:
the time-frequency diagram analysis submodule is used for performing fast Fourier transform of a slow time dimension on the radar echo signal to obtain a time-frequency diagram, and determining speed characteristic data and sign characteristic data of the object to be detected based on the motion characteristic of the object to be evaluated represented by the time-frequency diagram;
and the point cloud picture analysis submodule is used for converting the radar echo signal into a point cloud picture and determining the attitude characteristic data of the object to be evaluated based on the human body space position information of the object to be evaluated, which is represented by the point cloud picture.
Optionally, in a specific implementation manner, the target feature data includes: the posture characteristic data, the speed characteristic data and the physical sign characteristic data; the fall risk assessment model comprises a feature extraction layer and a result assessment layer;
the risk assessment module is specifically configured to:
inputting the posture characteristic data, the speed characteristic data and the sign characteristic data into the falling risk assessment model, and obtaining a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the feature extraction layer is used for respectively performing high-dimensional feature extraction on the attitude feature data, the speed feature data and the physical sign feature data to obtain a plurality of high-dimensional feature information, and performing fusion calculation on the plurality of high-dimensional feature information to obtain a target high-dimensional feature; the result evaluation layer is used for learning the target high-dimensional features and outputting fall risk evaluation results related to the object to be evaluated and corresponding to the target high-dimensional features.
Optionally, in a specific implementation manner, the signal obtaining module is specifically configured to:
and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal in the process of executing the specified action by the object to be evaluated.
Optionally, in a specific implementation manner, the apparatus further includes:
and the scheme determining module is used for determining a target care scheme corresponding to the fall risk assessment result as the care scheme of the object to be assessed according to the preset corresponding relation between the care scheme and the risk assessment result.
In a third aspect, an embodiment of the present application provides an electronic device, including:
a memory for storing a computer program;
a processor, configured to implement any of the above-mentioned fall risk assessment methods when executing the programs stored in the memory.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements any of the fall risk assessment methods described above.
Embodiments of the present application also provide a computer program product containing instructions, which when run on a computer, cause the computer to perform any of the above-described fall risk assessment methods.
The embodiment of the application has the following beneficial effects:
as can be seen from the above, by applying the scheme provided by the embodiment of the application, when the fall risk of the object to be evaluated is evaluated, the radar echo signal generated by the object to be evaluated reflecting the radar signal can be obtained first; then, performing characteristic analysis on the acquired radar echo signal to obtain target characteristic data of the object to be evaluated; and then, the target characteristic data can be input into a pre-trained falling risk assessment model to obtain a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed.
Based on this, by applying the scheme provided by the embodiment of the present application, the sample feature data used for training the fall risk assessment model is the sample feature data determined based on the real sample object, and further, the fall risk assessment model trained based on the sample feature data can more accurately reflect the correspondence between the feature data and the fall risk. Therefore, the fall risk assessment result about the object to be assessed, which is obtained by inputting the target feature data of the object to be assessed into the fall risk assessment model, has higher accuracy. By applying the method provided by the embodiment of the application to fall risk assessment, the misjudgment rate of the fall risk assessment can be reduced, and the accuracy of the fall risk assessment is improved.
In addition, in the process of carrying out the fall risk assessment, the cost for acquiring the target characteristic data of the object to be assessed based on the radar signal is low, and the target characteristic data of the object to be assessed can be acquired under the condition that the object to be assessed does not wear equipment and does not acquire images of the object to be assessed, so that the fall risk assessment method provided by the embodiment of the application has good privacy. In addition, by applying the scheme provided by the embodiment of the application, the falling risk assessment model can replace related medical personnel to carry out falling risk assessment, so that automation and intellectualization of the falling risk assessment process are realized, and the labor cost is further reduced.
Of course, not all advantages described above need to be achieved at the same time in the practice of any one product or method of the present application.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and it is also obvious for a person skilled in the art to obtain other embodiments according to the drawings.
Fig. 1 is a schematic flow chart of a fall risk assessment method according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a training manner of a fall risk assessment model according to an embodiment of the present application;
fig. 3 is another schematic flow chart of a fall risk assessment method provided in the embodiment of the present application;
fig. 4 is a schematic flowchart of a fall risk assessment method according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a fall risk assessment apparatus according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application are within the scope of protection of the present application.
At present, fall risk assessment methods mainly comprise that an executive evaluator evaluates fall risks of the elderly by using various fall risk assessment scales. However, such an evaluation method is relatively dependent on the professional ability of the performing evaluator, and if the professional ability of the performing evaluator is insufficient, misjudgment of the fall risk is likely to result. Therefore, the current fall risk assessment method is not favorable for popularization of fall risk assessment.
In order to solve the above problem, embodiments of the present application provide a fall risk assessment method.
The method is suitable for various application scenes for evaluating the falling risk, for example, the falling risk of the old is evaluated in real time in a daily life scene of the old, so that when the falling risk of the old is high, early warning is timely sent out to inform relevant persons (such as children and nurses of the old) to take corresponding measures (such as key nursing, medical treatment and the like) in time; for example, when the elderly people enter the nursing home, the risk of falling of the elderly people is evaluated, so that the staff in the nursing home can take care of and care of the elderly people with a high risk of falling.
In addition, the method can be applied to various electronic devices which can acquire and process data, such as servers, notebook computers, desktop computers and the like, and is hereinafter referred to as fall risk assessment equipment for short. Wherein the fall risk assessment device may be a stand-alone electronic device, such as a desktop computer communicatively connected to the radar, a radar with data processing capabilities; it may also be a plurality of electronic devices, for example, a device cluster consisting of a data processing device and a radar. Based on this, the application scenario and the execution subject of the method are not limited in the embodiments of the present application.
The fall risk assessment method provided by the embodiment of the application can include the following steps:
acquiring a radar echo signal generated by a radar signal reflected by an object to be evaluated;
performing feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated, wherein the target feature data comprises: at least one of pose feature data, velocity feature data, and sign feature data;
inputting the target characteristic data into a pre-trained falling risk assessment model, and acquiring a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the falling risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample labels of each group of sample characteristic data represent the preset falling risk of the sample objects with the group of sample characteristic data.
As can be seen from the above, by applying the scheme provided by the embodiment of the application, when the fall risk of the object to be evaluated is evaluated, the radar echo signal generated by the object to be evaluated reflecting the radar signal can be obtained first; then, performing characteristic analysis on the acquired radar echo signal to obtain target characteristic data of the object to be evaluated; and then, the target characteristic data can be input into a pre-trained falling risk assessment model to obtain a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed.
Based on this, by applying the scheme provided by the embodiment of the present application, the sample feature data used for training the fall risk assessment model is the sample feature data determined based on the real sample object, and further, the fall risk assessment model trained based on the sample feature data can more accurately reflect the corresponding relationship between the feature data and the fall risk. Therefore, the fall risk assessment result about the subject to be assessed, which is obtained by inputting the target feature data of the subject to be assessed into the fall risk assessment model, has higher accuracy. By applying the method provided by the embodiment of the application to fall risk assessment, the misjudgment rate of fall risk assessment can be reduced, and the accuracy of fall risk assessment is improved.
In addition, in the process of carrying out the fall risk assessment, the cost for acquiring the target characteristic data of the object to be assessed based on the radar signal is low, and the target characteristic data of the object to be assessed can be acquired under the condition that the object to be assessed does not wear equipment and does not acquire images of the object to be assessed, so that the fall risk assessment method provided by the embodiment of the application has good privacy. In addition, by applying the scheme provided by the embodiment of the application, the falling risk assessment model can replace related medical personnel to carry out falling risk assessment, so that automation and intellectualization of the falling risk assessment process are realized, and the labor cost is further reduced.
Next, a fall risk assessment method provided in an embodiment of the present application is specifically described with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a fall risk assessment method provided in an embodiment of the present application, and as shown in fig. 1, the method may include the following steps S101 to S103.
S101: and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal.
S102: and performing characteristic analysis on the radar echo signal to obtain target characteristic data of the object to be evaluated.
Wherein the target feature data includes: at least one of pose feature data, velocity feature data, and sign feature data.
Generally, a radar signal transmitted by a radar is reflected when contacting an obstacle, and a part of the reflected radar signal returns to the radar, and such a returned radar signal is called a radar return signal. When the shape of the obstacle, the position and speed of the obstacle relative to the radar, and other characteristics change, the characteristics of the radar return signal reflected by the obstacle also change, for example, the frequency of the radar return signal, the time interval between the radar transmitting the radar signal and the radar receiving the radar return signal, and the like. Therefore, when the object to be evaluated is subjected to fall risk evaluation, the fall risk evaluation device may first acquire a radar echo signal generated by the object to be evaluated reflecting a radar signal, and then perform feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated.
After the radar echo signal of the object to be evaluated is obtained, attitude characteristic analysis, speed characteristic analysis and sign characteristic analysis can be performed on the radar echo signal in a Fourier transform mode and the like, so that attitude characteristic data, speed characteristic data and sign characteristic data of the object to be evaluated are obtained.
Optionally, in a specific implementation manner, the target feature data may include: attitude characteristic data, speed characteristic data and physical sign characteristic data; further, the step S102: the characteristic analysis of the radar echo signal to obtain target characteristic data of the object to be evaluated may include the following steps 11-12.
Step 11: and performing fast Fourier transform of a slow time dimension on the radar echo signal to obtain a time-frequency graph, and determining speed characteristic data and sign characteristic data of the object to be detected based on the motion characteristics of the object to be evaluated represented by the time-frequency graph.
After the radar echo signal of the object to be evaluated is obtained, the radar echo signal can be subjected to fast Fourier transform of a slow time dimension to obtain a time-frequency diagram, and then the speed characteristic data and the sign characteristic data of the object to be evaluated are determined according to the motion characteristic of the object to be evaluated represented by the time-frequency diagram.
Step 12: and converting the radar echo signals into a point cloud picture, and determining attitude characteristic data of the object to be evaluated based on the human body space position information of the object to be evaluated represented by the point cloud picture.
After the radar echo signal of the object to be evaluated is obtained, the radar echo signal can be converted into a point cloud picture, and then attitude characteristic data of the object to be evaluated is determined based on human body space position information of the object to be evaluated, which is represented by the point cloud picture.
Optionally, the posture characteristic data may include at least one of gait characteristic data, body posture characteristic data, and swing arm posture characteristic data during walking of the object to be evaluated; the speed characteristic data can comprise at least one of average speed and instantaneous speed of the object to be evaluated in a certain time period; the vital sign feature data may include at least one of a respiratory frequency and a heartbeat frequency of the subject to be evaluated.
Of course, the posture characteristic data, the speed characteristic data and the physical sign characteristic data may also include other characteristic data that may be used to reflect the fall state of the subject to be evaluated, and the embodiment of the present invention is not particularly limited thereto.
In addition, the radar signal may be a signal transmitted by various radars which can be used for obstacle detection, such as a millimeter wave radar and an ultra-wideband biological radar.
For example, in the case that the characteristic data of the physical sign includes a respiratory frequency and a heartbeat frequency of the object to be evaluated, the respiration and the heartbeat may cause periodic displacement with a small amplitude in the thoracic cavity of the human body, and therefore, the radar signal may be a millimeter wave radar signal, an ultra-wideband biological radar signal, or other radar signals sensitive to a small disturbance, so as to improve the accuracy of the target characteristic data obtained based on the radar echo signal analysis.
S103: and inputting the target characteristic data into a pre-trained falling risk assessment model, and acquiring a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed.
The fall risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample labels of each group of sample characteristic data represent preset fall risks of the sample objects with the group of sample characteristic data.
In the process of training the preset initial model by using the plurality of groups of sample characteristic data, the initial model can learn the corresponding relation between each group of sample characteristic data and the sample label of the group of sample characteristic data, and the corresponding relation between the sample characteristic data and the sample label is gradually established through learning the corresponding relation between a large amount of sample characteristic data and the sample label.
Thus, after the target feature data of the object to be evaluated is obtained, the target feature data can be input into a pre-trained fall risk evaluation model, and the fall risk evaluation model can determine the label corresponding to the input target feature data based on the determined corresponding relation and output the label as a fall risk evaluation result of the object to be evaluated.
Optionally, in a specific implementation manner, the training manner of the fall risk assessment model may include the following steps 21 to 23.
Step 21: multiple sets of sample characteristic data with sample labels are obtained.
In the process of training to obtain the fall risk assessment model, a plurality of sets of sample feature data with sample labels may be obtained first.
The sample characteristic data may be acquired in a manner of: firstly, obtaining a radar echo signal generated by a sample object reflecting a radar signal; then, performing characteristic analysis on the radar echo signal to obtain sample characteristic data of the sample object: and finally, evaluating the preset falling risk of the sample object, thereby adding a sample label to each group of sample characteristic data.
Regarding the selection of the sample object, for example, a plurality of nursing homes (or hospitals) may be randomly selected, and the elderly (or patients) in the plurality of nursing homes (or hospitals) may be used as the sample object; for another example, a plurality of persons whose presence falls to history and a plurality of persons whose absence falls to history may be taken as sample objects, and the like. Regarding the selection of the sample object, the determination and adjustment can be performed by those skilled in the relevant field, and the embodiment of the present application is not particularly limited.
Optionally, the preset fall risk may be determined in the following manner: and (3) evaluating the falling risks of the sample object by professional persons in a plurality of falling risk evaluation fields, and taking the average value of the evaluation results as the preset falling risk of the sample object.
Optionally, the preset fall risk may be determined in the following manner: the preset fall risk of the sample object is determined according to the fall history of the sample object, for example, the preset fall risk of the sample object falling back to the history is determined as 1, and the preset fall risk of the sample object not falling back to the history is determined as 0.
Optionally, in a specific implementation manner, after multiple sets of sample feature data with sample labels are obtained, data preprocessing operations such as outlier cleaning and data standardization can be performed on the multiple sets of sample feature data, so as to improve the quality of the sample feature data used for training the initial model, and further improve the subsequent model training effect.
Step 22: and training a preset initial model based on the multiple groups of sample characteristic data.
Step 23: and when the initial model meets the preset conditions, stopping training to obtain a falling risk evaluation model.
After obtaining a plurality of groups of sample characteristic data with sample labels, the plurality of groups of sample characteristic data can be used as training set and test set data of the initial model to train the preset initial model, and when the initial model meets preset conditions, the training is stopped to obtain a final fall risk assessment model.
The electronic device for model training and the fall risk assessment device for performing the fall risk assessment method provided by the embodiment of the present application may be the same electronic device or different electronic devices.
The initial model may be a fall risk assessment model constructed based on one or more of Neural networks such as CNN (Convolutional Neural Network), GRU (Gated Recurrent Unit), LSTM (Long Short-Term Memory), and the like, and the construction of the initial model is not specifically limited in the embodiment of the present application.
Optionally, the preset condition may be that the number of iterations of each set of sample feature data reaches a preset number.
Optionally, the sample feature data may be divided into training set sample feature data and test set sample feature data, and further, the preset condition may be that an error between a true value and a predicted value of a sample label of each group of test set sample feature data is smaller than a preset error, where the true value of the sample label of the test set sample feature data may be a value of the sample label of the test set sample feature data, and the predicted value of the sample label of the test set sample feature data may be a fall risk assessment result obtained by inputting the test set sample feature data into a fall risk assessment model.
Based on this, by applying the scheme provided by the embodiment of the present application, the sample feature data used for training the fall risk assessment model is the sample feature data determined based on the real sample object, and further, the fall risk assessment model trained based on the sample feature data can more accurately reflect the corresponding relationship between the feature data and the fall risk. Therefore, the fall risk assessment result about the subject to be assessed, which is obtained by inputting the target feature data of the subject to be assessed into the fall risk assessment model, has higher accuracy. By applying the method provided by the embodiment of the application to fall risk assessment, the misjudgment rate of the fall risk assessment can be reduced, and the accuracy of the fall risk assessment is improved.
In addition, in the process of carrying out the fall risk assessment, the cost of acquiring the target feature data of the object to be assessed based on the radar signal is low, and the target feature data of the object to be assessed can be acquired under the condition that the object to be assessed does not wear equipment and does not acquire images of the object to be assessed, so that the fall risk assessment method provided by the embodiment of the application has good privacy. In addition, by applying the scheme provided by the embodiment of the application, the falling risk assessment model can replace related medical personnel to carry out falling risk assessment, so that automation and intellectualization of the falling risk assessment process are realized, and the labor cost is further reduced.
Optionally, in a specific implementation manner, the number of the objects to be evaluated may be multiple, and then the fall risk evaluation device may simultaneously obtain radar echo signals generated by the multiple objects to be evaluated reflecting radar signals; performing characteristic analysis on the radar echo signals to obtain target characteristic data of each object to be evaluated; and then, inputting the target characteristic data of each object to be evaluated into a pre-trained falling risk evaluation model, and acquiring a falling risk evaluation result which is output by the falling risk evaluation model and is about each object to be evaluated.
Based on this, the fall risk assessment device can simultaneously perform fall risk assessment on a plurality of objects to be assessed, thereby improving the efficiency of the fall risk assessment.
Optionally, in a specific implementation manner, the target feature data includes: attitude characteristic data, speed characteristic data and physical sign characteristic data; the falling risk assessment model comprises a feature extraction layer and a result assessment layer; further, as shown in fig. 2, the step S103: inputting the target feature data into a pre-trained fall risk assessment model, and obtaining a fall risk assessment result about the object to be assessed output by the fall risk assessment model, may include the following step S201.
Step S201: inputting the posture characteristic data, the speed characteristic data and the physical sign characteristic data into a falling risk assessment model, and obtaining a falling risk assessment result which is output by the falling risk assessment model and is about an object to be assessed.
The feature extraction layer is used for respectively carrying out high-dimensional feature extraction on the attitude feature data, the speed feature data and the physical sign feature data to obtain a plurality of high-dimensional feature information, and carrying out fusion calculation on the plurality of high-dimensional feature information to obtain a target high-dimensional feature; and the result evaluation layer is used for learning the target high-dimensional features and outputting fall risk evaluation results which correspond to the target high-dimensional features and are about the object to be evaluated.
The falling risk assessment model can comprise a feature extraction layer and a result assessment layer, and the feature extraction layer can respectively perform high-dimensional feature extraction on the attitude feature data, the speed feature data and the sign feature data to obtain a plurality of high-dimensional feature information, and perform fusion calculation on the plurality of high-dimensional feature information to obtain target high-dimensional features; the result evaluation layer can learn the target high-dimensional features and output the fall risk evaluation result of the target to be evaluated corresponding to the target high-dimensional features, so that after the posture feature data, the speed feature data and the physical sign feature data of the target to be evaluated are obtained, the posture feature data, the speed feature data and the physical sign feature data can be input into the fall risk evaluation model, and the fall risk evaluation result of the target to be evaluated output by the fall risk evaluation model is obtained.
Based on this, by applying the scheme provided by the embodiment of the application, the fall risk assessment model can fuse the posture characteristic data, the speed characteristic data and the physical sign characteristic data of the object to be assessed, and perform fall risk assessment according to the fused characteristics. Therefore, the fall risk assessment result about the object to be assessed output by the fall risk assessment model is determined based on the feature data about multiple dimensions of the object to be assessed, and the accuracy is high. By applying the scheme provided by the embodiment of the application to fall risk assessment, the accuracy of the fall risk assessment can be improved.
Optionally, in a specific implementation manner, as shown in fig. 3, the step S101: acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal may include the following step S301.
Step S301: and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal in the process of executing the specified action by the object to be evaluated.
The specified actions may be actions specified by existing medical staff when the medical staff detects the fall risk of the elderly, or a series of specified actions specified by professionals in related fields according to the embodiments of the present application. In this regard, the present embodiment is not particularly limited.
For example, the above-mentioned specified action may be walking according to a specified route and crossing over a specified obstacle, and further, a radar echo signal generated by the object to be evaluated reflecting a radar signal during the process that the object to be evaluated walks according to the specified route and crosses over the specified obstacle may be obtained.
In the process of carrying out fall risk assessment on an object to be assessed, the object to be assessed can be allowed to execute an appointed action, further, in the process of executing the appointed action by the object to be assessed, a fall risk assessment device can acquire a radar echo signal generated by the object to be assessed reflecting a radar signal, then target characteristic data of the object to be assessed in the process of executing the appointed action is obtained through analysis according to the radar echo signal, finally the target characteristic data of the object to be assessed in the process of executing the appointed action is input into a pre-trained fall risk assessment model, and a fall risk assessment result which is output by the fall risk assessment model and is related to the object to be assessed is obtained.
Accordingly, in this specific implementation manner, each set of sample feature data used for training the fall risk assessment model includes sample feature data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal in the process of executing a specified action, and a sample label of each set of sample feature data represents a preset fall risk of the sample object having the set of sample feature data.
As can be seen from the above, the target feature data of the object to be evaluated and the sample feature data of the sample object are both acquired during the execution of the specified action. Based on this, the accuracy of the fall risk assessment result can be further improved.
Optionally, in a specific implementation manner, as shown in fig. 4, the fall risk assessment method provided in the embodiment of the present application may further include the following step S401.
Step S401: and determining a target care plan corresponding to the fall risk assessment result as a care plan of the subject to be assessed according to a preset corresponding relation between the care plan and the risk assessment result.
After the fall risk assessment result of the object to be assessed is obtained, the fall risk assessment device can also determine the care scheme of the object to be assessed according to the fall risk assessment result of the object to be assessed.
Optionally, the fall risk assessment apparatus stores or can obtain the care plan corresponding to each fall risk assessment result in a networked manner, and then after obtaining the fall risk assessment result of the object to be assessed, the fall risk assessment apparatus can determine the care plan corresponding to the fall risk assessment result as the care plan of the object to be assessed.
Illustratively, the care plan may include notes, dietary recommendations, exercise recommendations, nursing recommendations, hospitalization recommendations, etc. regarding the subject to be evaluated.
Based on the method, the subject to be evaluated and the nursing staff thereof can not only know the falling risk of the subject to be evaluated, but also obtain the nursing scheme of the subject to be evaluated, and the subject to be evaluated and the nursing staff thereof can take care of the subject to be evaluated according to the nursing scheme, so that the falling possibility of the subject to be evaluated is reduced, and the falling prevention effect is achieved.
Optionally, in a specific implementation manner, the fall risk assessment method provided in the embodiment of the present application may further include the following step 1.
Step 1: and when the falling risk evaluation result of the object to be evaluated meets a preset condition, sending a preset signal to the specified equipment.
Illustratively, when the fall risk assessment result of the object to be assessed reaches a high risk, the fall risk assessment device may send a high risk early warning to the mobile communication device of the caregiver of the object to be assessed, so that the caregiver of the object to be assessed can take corresponding measures in time to reduce the fall probability of the object to be assessed.
Optionally, when the fall risk assessment result of the object to be assessed meets the preset condition, the target feature data and the fall risk assessment result of the object to be assessed can also be uploaded to the specified storage space.
Illustratively, when the fall risk assessment result of the object to be assessed reaches a high risk, the fall risk assessment device can also upload the target characteristic data of the object to be assessed and the fall risk assessment result to the cloud case of the object to be assessed, and then when the object to be assessed is hospitalizing, medical staff can know the physical condition of the object to be assessed more comprehensively by looking over the cloud case of the object to be assessed.
Corresponding to the method for evaluating a fall risk provided by the embodiment of the present application, the embodiment of the present application further provides a device for evaluating a fall risk.
Fig. 5 is a schematic structural diagram of a fall risk assessment apparatus provided in an embodiment of the present application, and as shown in fig. 5, the fall risk assessment apparatus may include the following modules:
the signal acquiring module 501 is configured to acquire a radar echo signal generated by a radar signal reflected by an object to be evaluated;
a feature analysis module 502, configured to perform feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated, where the target feature data includes: at least one of pose feature data, velocity feature data, and sign feature data;
a risk assessment module 503, configured to input the target feature data into a pre-trained fall risk assessment model, and obtain a fall risk assessment result, which is output by the fall risk assessment model and is related to the object to be assessed; the falling risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample labels of each group of sample characteristic data represent the preset falling risk of the sample objects with the group of sample characteristic data.
Based on this, by applying the scheme provided by the embodiment of the present application, the sample feature data used for training the fall risk assessment model is the sample feature data determined based on the real sample object, and further, the fall risk assessment model trained based on the sample feature data can more accurately reflect the corresponding relationship between the feature data and the fall risk. Therefore, the fall risk assessment result about the subject to be assessed, which is obtained by inputting the target feature data of the subject to be assessed into the fall risk assessment model, has higher accuracy. By applying the method provided by the embodiment of the application to fall risk assessment, the misjudgment rate of fall risk assessment can be reduced, and the accuracy of fall risk assessment is improved.
In addition, in the process of carrying out the fall risk assessment, the cost for acquiring the target characteristic data of the object to be assessed based on the radar signal is low, and the target characteristic data of the object to be assessed can be acquired under the condition that the object to be assessed does not wear equipment and does not acquire images of the object to be assessed, so that the fall risk assessment method provided by the embodiment of the application has good privacy. In addition, by applying the scheme provided by the embodiment of the application, the falling risk assessment model can replace relevant medical personnel to carry out falling risk assessment, so that automation and intellectualization of a falling risk assessment process are realized, and the labor cost is reduced.
Optionally, in a specific implementation manner, the target feature data includes: the attitude characteristic data, the speed characteristic data and the physical sign characteristic data;
the feature analysis module includes:
the time-frequency image analysis sub-module is used for performing fast Fourier transform of a slow time dimension on the radar echo signal to obtain a time-frequency image, and determining speed characteristic data and sign characteristic data of the object to be detected based on the motion characteristics of the object to be evaluated represented by the time-frequency image;
and the point cloud picture analysis submodule is used for converting the radar echo signal into a point cloud picture and determining the attitude characteristic data of the object to be evaluated based on the human body space position information of the object to be evaluated, which is represented by the point cloud picture.
Optionally, in a specific implementation manner, the target feature data includes: the posture characteristic data, the speed characteristic data and the physical sign characteristic data; the fall risk assessment model comprises a feature extraction layer and a result assessment layer;
the risk assessment module is specifically configured to:
inputting the posture characteristic data, the speed characteristic data and the sign characteristic data into the falling risk assessment model, and obtaining a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the feature extraction layer is used for respectively performing high-dimensional feature extraction on the attitude feature data, the speed feature data and the physical sign feature data to obtain a plurality of high-dimensional feature information, and performing fusion calculation on the plurality of high-dimensional feature information to obtain a target high-dimensional feature; the result evaluation layer is used for learning the target high-dimensional features and outputting fall risk evaluation results related to the object to be evaluated and corresponding to the target high-dimensional features.
Optionally, in a specific implementation manner, the signal obtaining module is specifically configured to:
and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal in the process of executing the specified action by the object to be evaluated.
Optionally, in a specific implementation manner, the apparatus further includes:
and the scheme determining module is used for determining a target care scheme corresponding to the fall risk assessment result as the care scheme of the object to be assessed according to the preset corresponding relation between the care scheme and the risk assessment result.
An embodiment of the present application further provides an electronic device, as shown in fig. 6, including:
a memory 601 for storing a computer program;
a processor 602, configured to execute the programs stored in the memory 601, and implement the steps of any fall risk assessment method provided in the embodiments of the present application.
The electronic device may further include a communication bus and/or a communication interface, and the processor 602, the communication interface, and the memory 601 complete communication with each other through the communication bus.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
In a further embodiment provided by the present application, there is also provided a computer readable storage medium having stored therein a computer program which, when executed by a processor, carries out the steps of any of the fall risk assessment methods described above.
In a further embodiment provided by the present application, there is also provided a computer program product comprising instructions which, when run on a computer, cause the computer to perform any of the fall risk assessment methods of the embodiments described above.
In the above embodiments, all or part of the implementation may be realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, apparatus embodiments, electronic device embodiments, computer-readable storage medium embodiments, and computer program product embodiments are described with relative simplicity as they are substantially similar to method embodiments, where relevant only as described in portions of the method embodiments.
The above description is only for the preferred embodiment of the present application and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application are included in the scope of protection of the present application.
Claims (12)
1. A fall risk assessment method, characterized in that the method comprises:
acquiring a radar echo signal generated by a radar signal reflected by an object to be evaluated;
performing feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated, wherein the target feature data comprises: at least one of pose feature data, velocity feature data, and sign feature data;
inputting the target characteristic data into a pre-trained falling risk assessment model, and acquiring a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the fall risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample label of each group of sample characteristic data represents the preset fall risk of the sample object with the group of sample characteristic data.
2. The method of claim 1, wherein the target feature data comprises: the posture characteristic data, the speed characteristic data and the physical sign characteristic data;
the performing feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated includes:
performing fast Fourier transform of a slow time dimension on the radar echo signal to obtain a time-frequency graph, and determining speed characteristic data and sign characteristic data of the object to be detected based on the motion characteristic of the object to be evaluated, which is represented by the time-frequency graph;
and converting the radar echo signal into a point cloud picture, and determining attitude characteristic data of the object to be evaluated based on the human body space position information of the object to be evaluated represented by the point cloud picture.
3. The method of claim 1, wherein the target feature data comprises: the posture characteristic data, the speed characteristic data and the physical sign characteristic data; the fall risk assessment model comprises a feature extraction layer and a result assessment layer;
the inputting the target feature data into a pre-trained fall risk assessment model and acquiring a fall risk assessment result about the object to be assessed, which is output by the fall risk assessment model, includes:
inputting the posture characteristic data, the speed characteristic data and the sign characteristic data into the falling risk assessment model, and obtaining a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the feature extraction layer is used for respectively performing high-dimensional feature extraction on the attitude feature data, the speed feature data and the physical sign feature data to obtain a plurality of high-dimensional feature information, and performing fusion calculation on the plurality of high-dimensional feature information to obtain a target high-dimensional feature; the result evaluation layer is used for learning the target high-dimensional features and outputting fall risk evaluation results related to the object to be evaluated and corresponding to the target high-dimensional features.
4. The method according to claim 1, wherein the obtaining a radar return signal generated by a radar signal reflected by an object to be evaluated comprises:
and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal in the process of executing the specified action by the object to be evaluated.
5. The method according to any one of claims 1-4, further comprising:
and determining a target care plan corresponding to the fall risk assessment result as a care plan of the subject to be assessed according to a preset corresponding relation between the care plan and the risk assessment result.
6. A fall risk assessment apparatus, characterized in that the apparatus comprises:
the signal acquisition module is used for acquiring radar echo signals generated by the radar signals reflected by the object to be evaluated;
a feature analysis module, configured to perform feature analysis on the radar echo signal to obtain target feature data of the object to be evaluated, where the target feature data includes: at least one of pose feature data, velocity feature data, and sign feature data;
the risk evaluation module is used for inputting the target characteristic data into a pre-trained falling risk evaluation model and acquiring a falling risk evaluation result which is output by the falling risk evaluation model and is about the object to be evaluated; the falling risk assessment model is obtained by training a preset initial model by using a plurality of groups of sample characteristic data with sample labels, each group of sample characteristic data comprises sample characteristic data of a sample object determined based on a radar echo signal generated by the sample object reflecting a radar signal, and the sample labels of each group of sample characteristic data represent the preset falling risk of the sample objects with the group of sample characteristic data.
7. The apparatus of claim 6, wherein the target feature data comprises: the posture characteristic data, the speed characteristic data and the physical sign characteristic data;
the feature analysis module includes:
the time-frequency diagram analysis submodule is used for performing fast Fourier transform of a slow time dimension on the radar echo signal to obtain a time-frequency diagram, and determining speed characteristic data and sign characteristic data of the object to be detected based on the motion characteristic of the object to be evaluated represented by the time-frequency diagram;
and the point cloud picture analysis submodule is used for converting the radar echo signal into a point cloud picture and determining the attitude characteristic data of the object to be evaluated based on the human body space position information of the object to be evaluated, which is represented by the point cloud picture.
8. The apparatus of claim 6, wherein the target feature data comprises: the posture characteristic data, the speed characteristic data and the physical sign characteristic data; the fall risk assessment model comprises a feature extraction layer and a result assessment layer;
the risk assessment module is specifically configured to:
inputting the posture characteristic data, the speed characteristic data and the physical sign characteristic data into the falling risk assessment model, and obtaining a falling risk assessment result which is output by the falling risk assessment model and is about the object to be assessed;
the feature extraction layer is used for respectively performing high-dimensional feature extraction on the attitude feature data, the speed feature data and the physical sign feature data to obtain a plurality of high-dimensional feature information, and performing fusion calculation on the plurality of high-dimensional feature information to obtain a target high-dimensional feature; the result evaluation layer is used for learning the target high-dimensional features and outputting fall risk evaluation results related to the object to be evaluated and corresponding to the target high-dimensional features.
9. The apparatus of claim 6, wherein the signal acquisition module is specifically configured to:
and acquiring a radar echo signal generated by the object to be evaluated reflecting the radar signal in the process of executing the specified action by the object to be evaluated.
10. The apparatus according to any one of claims 6-9, further comprising:
and the scheme determining module is used for determining a target care scheme corresponding to the fall risk assessment result as the care scheme of the object to be assessed according to the preset corresponding relation between the care scheme and the risk assessment result.
11. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the method of any one of claims 1 to 5 when executing a program stored in the memory.
12. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211707449.2A CN115937903A (en) | 2022-12-29 | 2022-12-29 | Fall risk assessment method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211707449.2A CN115937903A (en) | 2022-12-29 | 2022-12-29 | Fall risk assessment method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115937903A true CN115937903A (en) | 2023-04-07 |
Family
ID=86555809
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211707449.2A Pending CN115937903A (en) | 2022-12-29 | 2022-12-29 | Fall risk assessment method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115937903A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116338684A (en) * | 2023-05-29 | 2023-06-27 | 长沙莫之比智能科技有限公司 | Human body falling detection method and system based on millimeter wave radar and deep learning |
-
2022
- 2022-12-29 CN CN202211707449.2A patent/CN115937903A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116338684A (en) * | 2023-05-29 | 2023-06-27 | 长沙莫之比智能科技有限公司 | Human body falling detection method and system based on millimeter wave radar and deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Seibert et al. | Application scenarios for artificial intelligence in nursing care: rapid review | |
| Shabbir et al. | Exploratory data analysis, classification, comparative analysis, case severity detection, and internet of things in COVID-19 telemonitoring for smart hospitals | |
| CN114787832A (en) | Method and server for federal machine learning | |
| CA3137079A1 (en) | Computer-implemented machine learning for detection and statistical analysis of errors by healthcare providers | |
| Broniatowski et al. | Assessing causal claims about complex engineered systems with quantitative data: internal, external, and construct validity | |
| US20190139643A1 (en) | Facilitating medical diagnostics with a prediction model | |
| Zaitseva et al. | Application of the Structure Function in the Evaluation of the Human Factor in Healthcare | |
| Howedi et al. | Exploring entropy measurements to identify multi-occupancy in activities of daily living | |
| Kanokoda et al. | Gesture prediction using wearable sensing systems with neural networks for temporal data analysis | |
| Alqaysi et al. | Hybrid Diagnosis Models for Autism Patients Based on Medical and Sociodemographic Features Using Machine Learning and Multicriteria Decision‐Making (MCDM) Techniques: An Evaluation and Benchmarking Framework | |
| CN115713256A (en) | Medical training assessment and evaluation method and device, electronic equipment and storage medium | |
| CN114864105A (en) | Animal disease early warning method and system based on social graph network | |
| CN115937903A (en) | Fall risk assessment method and device, electronic equipment and storage medium | |
| Kwon et al. | An explainable spatial-temporal graphical convolutional network to score freezing of gait in parkinsonian patients | |
| CN115512817A (en) | Doctor recommendation method and device based on child telemedicine | |
| CN118280579B (en) | Sepsis patient condition assessment method and system based on multi-mode data fusion | |
| CN117524413B (en) | Motion protection method, device, equipment and medium based on millimeter wave radar | |
| Robben et al. | How is grandma doing? Predicting functional health status from binary ambient sensor data | |
| Dhanalakshmi et al. | Early detection of sepsis using LSTM and reinforcement learning | |
| Sparks et al. | Tele-health monitoring of patient wellness | |
| US11308615B1 (en) | Systems and processes for improving medical diagnoses | |
| Arciniega-Montiel et al. | Compensation of measurement uncertainty in a remote fetal monitor | |
| Thai-Nghe et al. | Intelligent Systems and Data Science: First International Conference, ISDS 2023, Can Tho, Vietnam, November 11–12, 2023, Proceedings, Part I | |
| US10490307B1 (en) | Cognitive-based tools for care and charge of incontinent individuals or animals | |
| Kaliappan et al. | A Data-Driven Approach to Improving Neonatal Care through Artificial Intelligence and Clinical Data Integration |
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 |