CN114271799B - Cardiovascular detection method based on intelligent terminal - Google Patents
Cardiovascular detection method based on intelligent terminal Download PDFInfo
- Publication number
- CN114271799B CN114271799B CN202111569410.4A CN202111569410A CN114271799B CN 114271799 B CN114271799 B CN 114271799B CN 202111569410 A CN202111569410 A CN 202111569410A CN 114271799 B CN114271799 B CN 114271799B
- Authority
- CN
- China
- Prior art keywords
- cardiovascular disease
- cardiovascular
- concurrent
- physiological signals
- type
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a cardiovascular detection method based on an intelligent terminal, which comprises the following steps: step S1, extracting a plurality of groups of cardiovascular cases from large cardiovascular case data as case samples, and constructing a disease type recognition model based on the case samples; and S2, acquiring concurrency characteristics among cardiovascular disease categories and diagnosis indexes of the cardiovascular disease categories in the cardiovascular disease case big data, and constructing a concurrency guiding thinking diagram based on the concurrency characteristics and the diagnosis indexes, wherein the concurrency guiding thinking diagram is used for guiding the cardiovascular disease categories with concurrency according to the concurrency cardiovascular disease categories. The invention closely pays attention to the cardiovascular disease category with the possibility of concurrence with the cardiovascular disease category, realizes high-frequency capturing of the physiological signals of the cardiovascular disease category with the possibility of concurrence, can identify and early warn at the early stage of the concurrence, and reduces the danger caused by the concurrence of the cardiovascular disease category.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to a cardiovascular detection method based on an intelligent terminal.
Background
Cardiovascular disease has become a very common disease, severely threatening the health of humans. However, under the large background of relatively lack of medical resources and aggravated aging in China, the intelligent and personalized medical diagnosis mode has a huge development prospect. In the future, wearable health monitoring equipment capable of fusing multiple physiological parameters will become the dominant force army of digital ambulatory medical treatment.
Among the numerous physiological signals, electrocardiographic signals are important means for detecting heart diseases, particularly cardiovascular diseases with sudden and random nature. The information of the pulse wave in aspects of form, intensity, speed and the like reflects important physiological and pathological information of the cardiovascular system of the human body. The electrocardio and pulse signals belong to weak signals, have low amplitude and low frequency, so the electrocardio and pulse signals are extremely easy to be interfered by various types in the process of extracting the heart pulse signals.
Patent number CN201811085130.4 discloses a cardiovascular disease early warning system based on multi-physiological signal deep fusion. The system comprises a wearing device, a multi-physiological signal acquisition device, a transmission device, intelligent terminal equipment and a cloud server; the multiple physiological signal acquisition devices and the transmission devices are arranged on the wearing device; the multi-physiological signal acquisition device acquires physiological signals of a detected person and transmits the acquired physiological signal data to the intelligent terminal equipment through the transmission device; the intelligent terminal equipment is internally provided with a program, judges whether a detected person has cardiovascular disease risk or not, and transmits the processed physiological signals to the cloud server; the cloud server is internally provided with a program, performs disease classification and diagnosis on multiple physiological signals of a wearer, and feeds back the multiple physiological signals to the intelligent terminal; the cardiovascular disease early warning system based on the multi-physiological signal depth fusion provides more accurate references for cardiac function assessment and cardiac disease diagnosis.
Although the above prior art provides more accurate references for evaluation of cardiac functions and diagnosis of cardiac diseases, and can realize early warning of diseases, certain drawbacks still exist, such as: when the physiological signals exceed the health threshold, the physiological signals need to be monitored more closely, a group of diagnosis signals are obtained more quickly, the disease category of the target object is identified quickly, and when the physiological signals do not exceed the health threshold, the physiological signals are indicated to be monitored more closely in a healthy state, so that the technology monitors multiple physiological signals according to frequency indiscriminately, the physiological signals do not exceed the health threshold only causes the waste of high-frequency monitoring resources of the monitoring terminal, and when the physiological signals exceed the health threshold, the diagnosis signals are possibly omitted, finally the cardiovascular disease monitoring precision is reduced, and the life and property of the target object are damaged.
Disclosure of Invention
The invention aims to provide a cardiovascular detection method based on an intelligent terminal, which aims to solve the technical problems that in the prior art, multiple physiological signals are monitored according to indiscriminate mode, only high-frequency monitoring resources of a monitoring terminal are wasted when the physiological signals do not exceed a health threshold, diagnostic signals are possibly omitted when the physiological signals exceed the health threshold, and finally cardiovascular disease monitoring accuracy is reduced.
In order to solve the technical problems, the invention specifically provides the following technical scheme:
a cardiovascular detection method based on an intelligent terminal comprises the following steps:
step S1, extracting a plurality of groups of cardiovascular cases from large data of cardiovascular cases as case samples, and constructing a disease type identification model based on the case samples, wherein the disease type identification model is used for identifying the type of the cardiovascular diseases which are caused by the occurrence according to physiological signals, and the physiological signals are characterized as diagnostic indexes for judging the type of the cardiovascular diseases which are caused by the occurrence;
s2, acquiring concurrency characteristics among cardiovascular disease categories and diagnosis indexes of the cardiovascular disease categories in the large cardiovascular disease data, and constructing a concurrency guiding thinking diagram based on the concurrency characteristics and the diagnosis indexes, wherein the concurrency guiding thinking diagram is used for guiding the cardiovascular disease categories with concurrency according to the concurrency cardiovascular disease categories;
step S3, monitoring a plurality of types of physiological signals of a target object in real time at a fixed frequency, synchronously extracting the physiological signals of the super-health threshold, adjusting the monitoring frequency of the physiological signals of the super-health threshold from the fixed frequency to an adaptive frequency, carrying out timing long monitoring on the physiological signals of the super-health threshold by the adaptive frequency to obtain a group of physiological signals for judging cardiovascular disease types as diagnosis signals, obtaining the cardiovascular disease types of the target object through a disease type identification model by the diagnosis signals, synchronously feeding back the cardiovascular disease types of the target object to a concurrent guiding thinking guide graph, and representing the physiological signals of the super-health threshold as physiological signals which are separated from the healthy state and are in abnormal states in the target object;
And S4, outputting the concurrent cardiovascular disease category and the diagnosis index of the concurrent cardiovascular disease category according to the concurrent cardiovascular disease category of the target object by utilizing the concurrent thinking guide graph, synchronously feeding back the concurrent cardiovascular disease category and the diagnosis index of the concurrent cardiovascular disease category to the monitoring terminal to improve the fixed frequency of the physiological signal which is the same as the diagnosis index of the concurrent cardiovascular disease category, and monitoring the physiological signal which is the same as the diagnosis index of the concurrent cardiovascular disease category in real time by the improved fixed frequency so as to improve the attention degree of the concurrent cardiovascular disease category to ensure that the disease occurrence detection of the concurrent cardiovascular disease has high timeliness.
As a preferable scheme of the invention, a plurality of groups of cardiovascular cases are extracted from the big data of cardiovascular cases as case samples, and a disease type recognition model is constructed based on the case samples, comprising the following steps:
extracting a group of physiological signals for diagnosing the type of the cardiovascular diseases of the cardiovascular cases from the medical records report of each cardiovascular case in sequence as sample data, taking the type of the cardiovascular diseases of the cardiovascular cases as a sample label of the group of physiological signals, and combining the sample data and the sample label to obtain a case sample, wherein the case sample is marked as [ physiological signals, type of the cardiovascular diseases ];
Constructing a mapping model representing nonlinear mapping relation between the physiological signal and the type of the cardiovascular disease by using the BP neural network to the physiological signal and the type of the cardiovascular disease, wherein the mapping model is used as a disease type identification model, and the functional expression of the disease type identification model is as follows:
Z y ＝BP[X y ]；
wherein Z is y Characterization of the type of cardiovascular disease that has developed for the sample of the y-th case, X y Physiological signals characterized as a y-th case sample, and BP characterized as a BP neural network;
and dividing all the case samples into a training set and a test set according to the data volume of 6:4, and carrying out model training on the model by using the model identification model so that the model identification model accurately maps the nonlinear relation between the physiological signals and the type of the developed cardiovascular diseases.
As a preferred embodiment of the present invention, the obtaining, in the cardiovascular disease case big data, the concurrence characteristics between cardiovascular disease categories and the diagnostic index of the cardiovascular disease categories includes:
sequentially counting the total number of cardiovascular cases of each of the categories of the developed cardiovascular diseases in the large data of the cardiovascular disease cases, and counting all the concurrent cardiovascular disease categories of all the cardiovascular cases in each category of the developed cardiovascular diseases;
Counting the total number of cardiovascular cases of each concurrent cardiovascular disease category in each concurrent cardiovascular disease category, and taking the ratio of the total number of cardiovascular cases of each concurrent cardiovascular disease category to the total number of cardiovascular cases of each concurrent cardiovascular disease category as the incidence of each concurrent cardiovascular disease category;
a concurrency characteristic of more than 60% of the concurrency cardiovascular disease categories in each of the concurrency cardiovascular disease categories as corresponding concurrency cardiovascular disease categories, the concurrency characteristic characterized by all concurrency cardiovascular disease categories of the concurrency cardiovascular disease categories;
sequentially counting the diagnosis indexes of each type of the developed cardiovascular diseases in the large data of the cardiovascular disease cases, counting the total number of the cardiovascular cases applied by each diagnosis index in each type of the developed cardiovascular diseases, and taking the ratio of the total number of the cardiovascular cases applied by each diagnosis index to the total number of the cardiovascular cases of each type of the developed cardiovascular diseases as the application rate of each diagnosis index;
and selecting the diagnosis index corresponding to the highest application rate from each developed cardiovascular disease category as the diagnosis index of the corresponding cardiovascular disease category.
As a preferred solution of the present invention, the constructing a concurrent guiding thinking guide graph based on the concurrent feature and the diagnostic index includes:
taking the concurrent guide of the cardiovascular diseases as a core theme, taking the concurrent feature as a first branch theme of the category of the cardiovascular diseases, and taking the diagnosis index as the first branch theme of the category of the cardiovascular diseases;
all the developed cardiovascular disease categories are used as main nodes connected in parallel at a core theme, all the developed cardiovascular disease categories in the concurrence characteristic are used as first branch nodes connected in parallel at a first branch theme, and physiological signal categories corresponding to diagnostic indexes are used as second branch nodes connected at a second branch theme;
the second branch theme and the first branch theme are connected in parallel at each main node so as to realize the concurrent guiding thinking guide graph which constructs the core theme, the first branch theme, the main node, the first branch node and the second branch node into a tree shape.
As a preferred embodiment of the present invention, the synchronously extracting the physiological signals exceeding the health threshold includes:
Setting a signal threshold value for each class of physiological signals as the health threshold value, comparing the physiological signals of a plurality of classes with the health threshold value in turn, wherein,
when the physiological signal is greater than the health threshold, extracting the corresponding physiological signal;
when the physiological signal is smaller than or equal to the health threshold value, the corresponding physiological signal does not need to be extracted.
As a preferred embodiment of the present invention, the adjusting the monitoring frequency of the physiological signal with the super-health threshold from the fixed frequency to the adaptive frequency includes:
taking the fixed frequency as an initial monitoring frequency of the physiological signal exceeding the health threshold in the time sequence interval of the timing length monitoring, and sequentially calculating the overflow degree of the physiological signal and the health threshold at each time sequence in the time sequence interval of the timing length monitoring;
the calculation formula of the overflow degree is as follows:
wherein p is i,t Ith characterized as physiological signal as super-health thresholdDegree of overflow of physiological signal at t-th time sequence in time sequence interval and health threshold of physiological signal-like signal, S i,t A physiological signal at a t-th time sequence in a time sequence interval, characterized by an i-th type physiological signal as a physiological signal of a super-health threshold, S i,o The health threshold value of the ith class of physiological signals, which is characterized as physiological signals with super health threshold values, at the t time sequence in the time sequence interval, i, t and o are metering constants, and have no substantial meaning;
adaptively adjusting the monitoring frequency at each time sequence in the time sequence interval based on the overflow degree to obtain an adaptive frequency on the basis of the fixed frequency, wherein,
when p is i,t Not less than 0, the monitoring frequency at each time sequence is adjusted to be
When p is i,t < 0, the monitoring frequency at each timing is adjusted to f i,t ＝f i,o ；
Wherein f i,t The adaptive monitoring frequency, f, at the t-th time instant in a time instant interval of an i-th class of physiological signal characterized as a physiological signal that is a super-health threshold i,o A fixed frequency of class i physiological signals characterized as physiological signals above a health threshold.
As a preferred embodiment of the present invention, the step S3 further includes:
after the timing long monitoring of the physiological signals of the super-health threshold is completed, the monitoring frequency obtained by carrying out the averaging processing on all the monitoring frequencies in the timing interval of the timing long monitoring is used as the fixed frequency of the physiological signals of the super-health threshold, so that the follow-up attention to the physiological signals of the super-health threshold is improved.
As a preferred embodiment of the present invention, the step of synchronously feeding back the same physiological signal to the monitoring terminal as the diagnostic index of the concurrent cardiovascular disease category increases the fixed frequency, including:
Extracting the morbidity of physiological signals identical to the diagnostic index of the concurrent cardiovascular disease category, and constructing a fixed frequency improvement formula according to the morbidity, wherein the fixed frequency improvement formula is as follows:
wherein f j,q A fixed frequency, f, after an increase of a physiological signal of class j, characterized by the same physiological signal as a diagnostic index of the concurrent cardiovascular disease class j,o A fixed frequency before improvement of a physiological signal of class j characterized as the same physiological signal as a diagnostic index of a concurrent cardiovascular disease class, u j The incidence of physiological signals of class j, characterized as the same physiological signals as the diagnostic index of the concurrent cardiovascular disease class.
As a preferred embodiment of the present invention, the real-time monitoring of the physiological signals with the same diagnostic index of the concurrent cardiovascular disease category at an improved fixed frequency includes:
the monitoring terminal monitors a plurality of physiological signals of the target object, which are the same as the diagnosis indexes of the concurrent cardiovascular disease class, in real time at the improved fixed frequency, and synchronously extracts physiological signals which are the same as the diagnosis indexes of the concurrent cardiovascular disease class and exceed the health threshold;
Adjusting the monitoring frequency of the physiological signals with the same super-health threshold value and the diagnostic index of the concurrent cardiovascular disease category from the fixed frequency to an adaptive frequency, and carrying out timing long monitoring on the physiological signals with the same super-health threshold value and the diagnostic index of the concurrent cardiovascular disease category by the adaptive frequency to obtain a group of diagnostic signals;
passing the diagnostic signal through a disease-seed identification model, wherein,
if the disease type identification model outputs the type of the cardiovascular disease of the target object, the type of the cardiovascular disease of the target object is synchronously fed back to the concurrent guidance thinking guide graph, and the steps S4 and S3 are executed to obtain all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease, physiological signals with the same diagnosis indexes of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease, and the monitoring frequency of the physiological signals with the same diagnosis indexes of all the type of the cardiovascular disease of the target object;
if the disease type identification model cannot output the type of the cardiovascular disease of the target object, the type of the cardiovascular disease of the target object is not required to be fed back to the concurrent guidance thought map synchronously, and the step S3 is executed to monitor the physiological signals with the same diagnostic indexes of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease in real time, so that the attention degree of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease is improved.
As a preferred embodiment of the present invention, the method further comprises:
and monitoring and early warning are carried out while the disease type identification model outputs the type of the cardiovascular disease, so as to remind a target object of timely medical intervention.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the physiological signals with the super-health threshold are subjected to disease type identification by constructing the disease type identification model to obtain the type of the cardiovascular diseases which are sent out by the target user, and then the concurrent guiding thinking guide graph for guiding the type of the cardiovascular diseases which are concurrent according to the type of the cardiovascular diseases which are sent out is constructed, so that the type of the cardiovascular diseases which are concurrent with the type of the cardiovascular diseases which are sent out is closely concerned, the physiological signals of the type of the cardiovascular diseases which are concurrent are subjected to high-frequency capture, early warning can be identified at the early stage of the disease, and the danger caused by the disease of the type of the cardiovascular diseases which are concurrent is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It will be apparent to those of ordinary skill in the art that the drawings in the following description are exemplary only and that other implementations can be obtained from the extensions of the drawings provided without inventive effort.
Fig. 1 is a flowchart of a cardiovascular detection method according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, the detection and early warning for cardiovascular diseases are generally established by using an intelligent terminal (such as a wearable terminal device, a smart watch, a smart sphygmomanometer and the like) to perform constant-frequency real-time monitoring on a target object (such as heart beat, pulse, blood pressure and the like), judging whether the physiological signal exceeds a health threshold value when monitoring the physiological signal, taking the physiological signal exceeding the health threshold value as an index for judging that the target object has cardiovascular diseases, and directly performing safety early warning to remind the target object to perform medical care, but adopting constant-frequency monitoring in the process, setting the frequency to be too high can cause excessive data volume of the physiological signal obtained by monitoring, the data processing volume is increased greatly, but the physiological signal without risk on the target object is in a relatively stable and normal state for a long time, the invention provides a cardiovascular detection method based on an intelligent terminal, which is used for identifying the type of the cardiovascular diseases of the target object and determining the type of the cardiovascular diseases accompanied with the type of the cardiovascular diseases of the target object by a concurrent guiding thinking diagram, thereby determining the physiological signal with higher risk for the target object, and the monitoring frequency of the higher physiological signals is improved to ensure that the physiological signals without risk to the target object are subjected to low-frequency acquisition, the data redundancy of the physiological signals is removed while the monitoring precision is not influenced, the physiological signals with risk to the target object are subjected to high-frequency acquisition, the data volume of the physiological signals is improved to judge the cardiovascular disease category represented by the physiological signals more accurately, the monitoring resources of the intelligent terminal are reasonably distributed, and the detection precision of cardiovascular detection is improved.
A cardiovascular detection method based on an intelligent terminal comprises the following steps:
step S1, extracting a plurality of groups of cardiovascular cases from large data of cardiovascular disease cases as case samples, and constructing a disease type identification model based on the case samples, wherein the disease type identification model is used for identifying the type of the cardiovascular disease which is caused by the occurrence according to physiological signals, and the physiological signals are characterized as diagnostic indexes for judging the type of the cardiovascular disease which is caused by the occurrence;
the developed cardiovascular disease category is characterized as a cardiovascular disease category that the target subject has developed, and the large cardiovascular disease category is characterized as a cardiovascular disease category that there is a common possibility of developing along with the cardiovascular disease category that the target subject has developed.
Extracting a plurality of groups of cardiovascular cases from the cardiovascular disease case big data as case samples, and constructing a disease type identification model based on the case samples, wherein the method comprises the following steps:
extracting a group of physiological signals for diagnosing the type of the cardiovascular diseases of the cardiovascular cases from the medical records report of each cardiovascular case in sequence as sample data, taking the type of the cardiovascular diseases of the cardiovascular cases as a sample label of the group of physiological signals, and combining the sample data and the sample label to obtain a case sample, wherein the case sample is marked as [ physiological signals, type of the cardiovascular diseases ];
Constructing a mapping model representing nonlinear mapping relation between physiological signals and the type of the cardiovascular diseases by using BP neural network to the physiological signals and the type of the cardiovascular diseases, wherein the mapping model is used as a disease type identification model, and the functional expression of the disease type identification model is as follows:
Z y ＝BP[X y ]；
wherein Z is y Characterization of the type of cardiovascular disease that has developed for the sample of the y-th case, X y Physiological signals characterized as a y-th case sample, and BP characterized as a BP neural network;
and dividing all the case samples into a training set and a test set according to the data volume of 6:4, and carrying out model training on the model by using the model identification model so that the model identification model accurately maps the nonlinear relation between the physiological signals and the type of the developed cardiovascular diseases.
S2, acquiring concurrency characteristics among cardiovascular disease categories and diagnosis indexes of the cardiovascular disease categories in the large cardiovascular disease data, and constructing a concurrency guiding thinking chart based on the concurrency characteristics and the diagnosis indexes, wherein the concurrency guiding thinking chart is used for guiding the cardiovascular disease categories with concurrency according to the concurrency cardiovascular disease categories;
obtaining the concurrency characteristics among cardiovascular disease categories and the diagnosis indexes of the cardiovascular disease categories in the big data of the cardiovascular disease cases comprises the following steps:
Sequentially counting the total number of cardiovascular cases of each of the categories of the developed cardiovascular diseases in the large cardiovascular disease data, and counting all the concurrent cardiovascular disease categories of all the cardiovascular cases in each of the categories of the developed cardiovascular diseases;
counting the total number of cardiovascular cases of each concurrent cardiovascular disease category in each concurrent cardiovascular disease category, and taking the ratio of the total number of cardiovascular cases of each concurrent cardiovascular disease category to the total number of cardiovascular cases of each concurrent cardiovascular disease category as the incidence of each concurrent cardiovascular disease category;
a concurrency class of cardiovascular diseases having a incidence of more than 60% in each of the categories of cardiovascular diseases being a concurrency characteristic of the corresponding category of cardiovascular diseases, the concurrency characteristic being characterized by all of the concurrency classes of cardiovascular diseases of the category of cardiovascular diseases;
sequentially counting the diagnosis indexes of each of the categories of the cardiovascular diseases in the large cardiovascular disease data, counting the total number of the cardiovascular cases applied by each of the diagnosis indexes in each of the categories of the cardiovascular diseases, and taking the ratio of the total number of the cardiovascular cases applied by each of the diagnosis indexes to the total number of the cardiovascular cases of each of the categories of the cardiovascular diseases as the application rate of each of the diagnosis indexes;
And selecting the diagnosis index corresponding to the highest application rate from each developed cardiovascular disease category as the diagnosis index of the corresponding cardiovascular disease category.
Constructing a concurrent guideline mind map based on the concurrent features and the diagnostic indicators, comprising:
the method comprises the steps of taking a cardiovascular disease concurrency guide as a core theme, taking concurrency characteristics as a first branch theme of an established cardiovascular disease category, and taking a diagnosis index as the first branch theme of the established cardiovascular disease category;
all the categories of the developed cardiovascular diseases are used as main nodes connected in parallel at the core theme, all the categories of the developed cardiovascular diseases in the concurrent feature are used as first branch nodes connected in parallel at the first branch theme, and the physiological signal category corresponding to the diagnosis index is used as second branch nodes connected at the second branch theme;
the second branch theme and the first branch theme are connected in parallel at each main node so as to realize the concurrent guiding thinking guide graph which constructs the core theme, the first branch theme, the main node, the first branch node and the second branch node into a tree shape.
The embodiment provides an example of construction of a concurrent guidance chart, in practice, for example, atherosclerosis in cardiovascular system diseases, there may be concurrent hyperlipidemia, and the complications of hyperlipidemia may also include diabetes, hypothyroidism and sick syndromes, so when the concurrent cardiovascular disease category includes a, B, G, W, R, M, and the concurrent cardiovascular disease category of the concurrent cardiovascular disease category a includes B, G, W, and the concurrent cardiovascular disease category of the concurrent cardiovascular disease category B includes R, M, where the diagnostic indexes of a, B are a, B, G, W, R, M, respectively, the concurrent guidance chart is constructed:
Cardiovascular disease concurrency guidelines
When in specific use, the concurrent guiding thinking guide graph is converted into a tree structure and stored in a computer system to obtain
When the cardiovascular disease is identified as A, the computer identifies the cardiovascular disease with concurrency as B, G and W in the tree structure of the concurrency guide thinking guide graph, and identifies the cardiovascular disease with concurrency as B, G and W, and the diagnosis indexes of B, G and W indicate that the diagnosis indexes of B, G and W represent that the risk of the target object in an abnormal state is higher, so that the detection frequency of the physiological signals corresponding to the diagnosis indexes of B, G and W is improved, the high-frequency monitoring of the diagnosis indexes of B, G and W is realized, the close attention to the physiological signals with higher risk is realized, and early warning is carried out at the detectable position of the concurrency early stage of the cardiovascular disease of B, G and W, so that the complication risk of the target object is reduced.
Step S3, monitoring a plurality of types of physiological signals of a target object in real time at a fixed frequency, synchronously extracting physiological signals with an ultra-health threshold, adjusting the monitoring frequency of the physiological signals with the ultra-health threshold from the fixed frequency to an adaptive frequency, carrying out timing long monitoring on the physiological signals with the ultra-health threshold by the adaptive frequency to obtain a group of physiological signals for judging cardiovascular disease types as diagnosis signals, then obtaining the cardiovascular disease types of the target object through a disease type identification model by the diagnosis signals, synchronously feeding the cardiovascular disease types of the target object back to a concurrent guiding thinking figure, and characterizing the physiological signals with the ultra-health threshold as physiological signals which are separated from the health state and are in an abnormal state in the target object;
Synchronously extracting physiological signals exceeding a health threshold, comprising:
a signal threshold is set for each class of physiological signals as a health threshold, physiological signals of multiple classes are compared with the health threshold in sequence, wherein,
when the physiological signal is greater than the health threshold, extracting the corresponding physiological signal;
when the physiological signal is smaller than or equal to the health threshold value, the corresponding physiological signal does not need to be extracted.
Adjusting the monitoring frequency of the physiological signal exceeding the health threshold from a fixed frequency to an adaptive frequency, comprising:
taking the fixed frequency as the initial monitoring frequency of the physiological signal exceeding the health threshold in a time sequence interval of the timing length monitoring, and sequentially calculating the overflow degree of the physiological signal and the health threshold at each time sequence in the time sequence interval of the timing length monitoring;
the calculation formula of the overflow degree is as follows:
wherein p is i,t Degree of overflow of physiological signal at t-th time sequence in time sequence interval and health threshold of ith class of physiological signal characterized as physiological signal exceeding health threshold, S i,t A physiological signal at a t-th time sequence in a time sequence interval, characterized by an i-th type physiological signal as a physiological signal of a super-health threshold, S i,o The health threshold value of the ith class of physiological signals, which is characterized as physiological signals with super health threshold values, at the t time sequence in the time sequence interval, i, t and o are metering constants, and have no substantial meaning;
the monitoring frequency at each time sequence in the time sequence interval is adaptively adjusted on the basis of the fixed frequency based on the overflow degree to obtain an adaptive frequency, wherein,
when p is i,t Not less than 0, the monitoring frequency at each time sequence is adjusted to be
When p is i,t < 0, the monitoring frequency at each timing is adjusted to f i,t ＝f i,o ；
Wherein f i,t An adaptive monitoring frequency, f, at a t-th time instant in a time instant interval, of a physiological signal of class i characterized as a physiological signal of super-health threshold i,o A fixed frequency of class i physiological signals characterized as physiological signals above a health threshold.
The higher the physiological signal level exceeding the health threshold, the higher the risk level of occurrence of an abnormal state, and therefore the more data details of the physiological signal need to be captured, and therefore the more data volume is needed, and for the purpose of realizing the more data volume, the higher the monitoring frequency is given to the physiological signal with the higher overflow level, namely, the monitoring frequency is set to the adaptive frequency f i,t Can be realized.
Step S3 further includes:
After the timing long monitoring of the physiological signals of the super-health threshold is completed, the monitoring frequency obtained by carrying out the averaging processing on all the monitoring frequencies in the timing interval of the timing long monitoring is used as the fixed frequency of the physiological signals of the super-health threshold, so that the follow-up attention to the physiological signals of the super-health threshold is improved.
Synchronous feedback to the monitoring terminal increases the fixed frequency of the physiological signal identical to the diagnostic index of the concurrent cardiovascular disease category, comprising:
extracting the morbidity of physiological signals identical to the diagnostic index of the concurrent cardiovascular disease category, and constructing a fixed frequency improvement formula according to the morbidity, wherein the fixed frequency improvement formula is as follows:
wherein f j,q A fixed frequency, f, after an increase of a physiological signal of class j, characterized by the same physiological signal as a diagnostic index of the concurrent cardiovascular disease class j,o A fixed frequency before improvement of a physiological signal of class j characterized as the same physiological signal as a diagnostic index of a concurrent cardiovascular disease class, u j The incidence of physiological signals of class j, characterized as the same physiological signals as the diagnostic index of the concurrent cardiovascular disease class.
The higher the incidence, the higher the degree of risk of abnormal state of physiological signal same as diagnostic index of concurrent cardiovascular disease category, therefore, the more data detail of the physiological signal needs to be captured, and more data volume is needed, for the purpose of realizing more data volume, the higher the monitoring frequency is given to the physiological signal with higher incidence, namely, the monitoring frequency is increased to f j,q Can realize
And S4, outputting the concurrent cardiovascular disease category and the diagnosis index of the concurrent cardiovascular disease category according to the concurrent cardiovascular disease category of the target object by utilizing the concurrent thinking guide graph, synchronously feeding back the concurrent cardiovascular disease category and the diagnosis index of the concurrent cardiovascular disease category to the monitoring terminal to improve the fixed frequency of the physiological signal which is the same as the diagnosis index of the concurrent cardiovascular disease category, and monitoring the physiological signal which is the same as the diagnosis index of the concurrent cardiovascular disease category in real time by the improved fixed frequency so as to improve the attention degree of the concurrent cardiovascular disease category and ensure that the concurrent cardiovascular disease is detected with high timeliness.
Real-time monitoring of physiological signals of the same diagnostic index of the category of concurrent cardiovascular disease at an increased fixed frequency, comprising:
the monitoring terminal monitors a plurality of physiological signals of the target object, which are the same as the diagnosis indexes of the concurrent cardiovascular disease class, in real time at the improved fixed frequency, and synchronously extracts physiological signals which are the same as the diagnosis indexes of the concurrent cardiovascular disease class and exceed the health threshold;
the monitoring frequency of the physiological signals with the super-health threshold value and the same diagnostic index of the concurrent cardiovascular disease class is adjusted from the fixed frequency to the self-adaptive frequency, and the physiological signals with the super-health threshold value and the same diagnostic index of the concurrent cardiovascular disease class are monitored at regular time by the self-adaptive frequency to obtain a group of diagnostic signals;
Passing the diagnostic signal through a disease species identification model, wherein,
if the disease type identification model outputs the type of the cardiovascular disease of the target object, the type of the cardiovascular disease of the target object is synchronously fed back to the concurrent guidance thinking guide graph, and the steps S4 and S3 are executed to obtain all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease, physiological signals with the same diagnosis indexes of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease, and the monitoring frequency of the physiological signals with the same diagnosis indexes of all the type of the cardiovascular disease of the target object;
if the disease type identification model cannot output the type of the cardiovascular disease of the target object, the type of the cardiovascular disease of the target object is not required to be fed back to the concurrent guidance thought map synchronously, and the step S3 is executed to monitor the physiological signals with the same diagnostic indexes of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease in real time, so that the attention degree of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease is improved.
Further comprises: and monitoring and early warning are carried out while the disease type identification model outputs the type of the cardiovascular disease, so as to remind a target object of timely medical intervention.
According to the invention, the physiological signals with the super-health threshold are subjected to disease type identification by constructing the disease type identification model to obtain the type of the cardiovascular diseases which are sent out by the target user, and then the concurrent guiding thinking guide graph for guiding the type of the cardiovascular diseases which are concurrent according to the type of the cardiovascular diseases which are sent out is constructed, so that the type of the cardiovascular diseases which are concurrent with the type of the cardiovascular diseases which are sent out is closely concerned, the physiological signals of the type of the cardiovascular diseases which are concurrent are subjected to high-frequency capture, early warning can be identified at the early stage of the disease, and the danger caused by the disease of the type of the cardiovascular diseases which are concurrent is reduced.
The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application, the scope of which is defined by the claims. Various modifications and equivalent arrangements may be made to the present application by those skilled in the art, which modifications and equivalents are also considered to be within the scope of the present application.
Claims (9)
1. The cardiovascular detection method based on the intelligent terminal is characterized by comprising the following steps of:
step S1, extracting a plurality of groups of cardiovascular cases from large data of cardiovascular cases as case samples, and constructing a disease type identification model based on the case samples, wherein the disease type identification model is used for identifying the type of the cardiovascular diseases which are caused by the occurrence according to physiological signals, and the physiological signals are characterized as diagnostic indexes for judging the type of the cardiovascular diseases which are caused by the occurrence;
s2, acquiring concurrency characteristics among cardiovascular disease categories and diagnosis indexes of the cardiovascular disease categories in the large cardiovascular disease data, and constructing a concurrency guiding thinking diagram based on the concurrency characteristics and the diagnosis indexes, wherein the concurrency guiding thinking diagram is used for guiding the cardiovascular disease categories with concurrency according to the concurrency cardiovascular disease categories;
step S3, monitoring a plurality of types of physiological signals of a target object in real time at a fixed frequency, synchronously extracting the physiological signals of the super-health threshold, adjusting the monitoring frequency of the physiological signals of the super-health threshold from the fixed frequency to an adaptive frequency, carrying out timing long monitoring on the physiological signals of the super-health threshold by the adaptive frequency to obtain a group of physiological signals for judging cardiovascular disease types as diagnosis signals, obtaining the cardiovascular disease types of the target object through a disease type identification model by the diagnosis signals, synchronously feeding back the cardiovascular disease types of the target object to a concurrent guiding thinking guide graph, and representing the physiological signals of the super-health threshold as physiological signals which are separated from the healthy state and are in abnormal states in the target object;
Step S4, outputting a concurrent cardiovascular disease category and diagnostic indexes of the concurrent cardiovascular disease category according to the concurrent cardiovascular disease category of the target object by utilizing a concurrent thinking guide graph, and synchronously feeding back the concurrent cardiovascular disease category and the diagnostic indexes of the concurrent cardiovascular disease category to a monitoring terminal to improve the fixed frequency of physiological signals which are the same as the diagnostic indexes of the concurrent cardiovascular disease category, so that the improved fixed frequency monitors the physiological signals which are the same as the diagnostic indexes of the concurrent cardiovascular disease category in real time, and the degree of attention of the concurrent cardiovascular disease category is improved to ensure that the concurrent cardiovascular disease is detected with high timeliness;
the adjusting the monitoring frequency of the physiological signal exceeding the health threshold from the fixed frequency to an adaptive frequency comprises:
taking the fixed frequency as an initial monitoring frequency of the physiological signal exceeding the health threshold in the time sequence interval of the timing length monitoring, and sequentially calculating the overflow degree of the physiological signal and the health threshold at each time sequence in the time sequence interval of the timing length monitoring;
the calculation formula of the overflow degree is as follows:
wherein p is i,t Degree of overflow of physiological signal at t-th time sequence in time sequence interval and health threshold of ith class of physiological signal characterized as physiological signal exceeding health threshold, S i,t A physiological signal at a t-th time sequence in a time sequence interval, characterized by an i-th type physiological signal as a physiological signal of a super-health threshold, S i,o The health threshold value of the ith class of physiological signals, which is characterized as physiological signals with super health threshold values, at the t time sequence in the time sequence interval, i, t and o are metering constants, and have no substantial meaning;
adaptively adjusting the monitoring frequency at each time sequence in the time sequence interval based on the overflow degree to obtain an adaptive frequency on the basis of the fixed frequency, wherein,
when p is i,t Not less than 0, the monitoring frequency at each time sequence is adjusted to be
When p is i,t <0, the monitoring frequency at each time sequence is adjusted to f i,t ＝ i,o ；
Wherein f i,t The adaptive frequency, f, at the t-th time instant in a time instant interval, of a physiological signal of class i characterized as a physiological signal of super-health threshold i,o A fixed frequency of class i physiological signals characterized as physiological signals above a health threshold.
2. The smart terminal-based cardiovascular detection method as claimed in claim 1, wherein: extracting a plurality of groups of cardiovascular cases from the cardiovascular disease case big data as case samples, and constructing a disease type identification model based on the case samples, wherein the method comprises the following steps:
Extracting a group of physiological signals for diagnosing the type of the cardiovascular diseases of the cardiovascular cases from the medical records report of each cardiovascular case in sequence as sample data, taking the type of the cardiovascular diseases of the cardiovascular cases as a sample label of the group of physiological signals, and combining the sample data and the sample label to obtain a case sample, wherein the case sample is marked as [ physiological signals, type of the cardiovascular diseases ];
constructing a mapping model representing nonlinear mapping relation between the physiological signal and the type of the cardiovascular disease by using the BP neural network to the physiological signal and the type of the cardiovascular disease, wherein the mapping model is used as a disease type identification model, and the functional expression of the disease type identification model is as follows:
Z y ＝P[X y ]；
wherein Z is y Characterization of the idiopathic nature of the y-th case sampleCardiovascular disease category, X y Physiological signals characterized as a y-th case sample, and BP characterized as a BP neural network;
and dividing all the case samples into a training set and a test set according to the data volume of 6:4, and carrying out model training on the model by using the model identification model so that the model identification model accurately maps the nonlinear relation between the physiological signals and the type of the developed cardiovascular diseases.
3. The smart terminal-based cardiovascular detection method as claimed in claim 2, wherein: the method for acquiring the concurrence characteristics among cardiovascular disease categories and the diagnosis indexes of the cardiovascular disease categories in the big data of the cardiovascular disease cases comprises the following steps:
sequentially counting the total number of cardiovascular cases of each of the categories of the developed cardiovascular diseases in the large data of the cardiovascular disease cases, and counting all the concurrent cardiovascular disease categories of all the cardiovascular cases in each category of the developed cardiovascular diseases;
counting the total number of cardiovascular cases of each concurrent cardiovascular disease category in each concurrent cardiovascular disease category, and taking the ratio of the total number of cardiovascular cases of each concurrent cardiovascular disease category to the total number of cardiovascular cases of each concurrent cardiovascular disease category as the incidence of each concurrent cardiovascular disease category;
a concurrency characteristic of more than 60% of the concurrency cardiovascular disease categories in each of the concurrency cardiovascular disease categories as corresponding concurrency cardiovascular disease categories, the concurrency characteristic characterized by all concurrency cardiovascular disease categories of the concurrency cardiovascular disease categories;
Sequentially counting the diagnosis indexes of each type of the developed cardiovascular diseases in the large data of the cardiovascular disease cases, counting the total number of the cardiovascular cases applied by each diagnosis index in each type of the developed cardiovascular diseases, and taking the ratio of the total number of the cardiovascular cases applied by each diagnosis index to the total number of the cardiovascular cases of each type of the developed cardiovascular diseases as the application rate of each diagnosis index;
and selecting the diagnosis index corresponding to the highest application rate from each developed cardiovascular disease category as the diagnosis index of the corresponding cardiovascular disease category.
4. A cardiovascular detection method based on an intelligent terminal according to claim 3, wherein: the constructing a concurrent guiding thinking guide graph based on the concurrent features and the diagnosis indexes comprises the following steps:
taking a cardiovascular disease concurrency guide as a core theme, taking the concurrency characteristic as a first branch theme of an established cardiovascular disease category, and taking a diagnosis index as a second branch theme of the established cardiovascular disease category;
all the developed cardiovascular disease categories are used as main nodes connected in parallel at a core theme, all the developed cardiovascular disease categories in the concurrence characteristic are used as first branch nodes connected in parallel at a first branch theme, and physiological signal categories corresponding to diagnostic indexes are used as second branch nodes connected at a second branch theme;
The second branch theme and the first branch theme are connected in parallel at each main node so as to realize the concurrent guiding thinking guide graph which constructs the core theme, the first branch theme, the second branch theme, the main node, the first branch node and the second branch node into a tree shape.
5. The smart terminal-based cardiovascular detection method as claimed in claim 4, wherein: the synchronously extracting the physiological signals exceeding the health threshold comprises the following steps:
setting a signal threshold value for each class of physiological signals as the health threshold value, comparing the physiological signals of a plurality of classes with the health threshold value in turn, wherein,
when the physiological signal is greater than the health threshold, extracting the corresponding physiological signal;
when the physiological signal is smaller than or equal to the health threshold value, the corresponding physiological signal does not need to be extracted.
6. The smart terminal-based cardiovascular testing method as recited in claim 5, wherein the step S3 further comprises:
after the timing long monitoring of the physiological signals of the super-health threshold is completed, the monitoring frequency obtained by carrying out the averaging processing on all the monitoring frequencies in the timing interval of the timing long monitoring is used as the fixed frequency of the physiological signals of the super-health threshold, so that the follow-up attention to the physiological signals of the super-health threshold is improved.
7. The smart terminal-based cardiovascular testing method of claim 6, wherein the synchronous feedback to the monitoring terminal increases the fixed frequency of physiological signals that are identical to diagnostic indicators of the concurrent cardiovascular disease category, comprising:
extracting the morbidity of physiological signals identical to the diagnostic index of the concurrent cardiovascular disease category, and constructing a fixed frequency improvement formula according to the morbidity, wherein the fixed frequency improvement formula is as follows:
wherein f j,q A fixed frequency, f, after an increase of a physiological signal of class j, characterized by the same physiological signal as a diagnostic index of the concurrent cardiovascular disease class j,o A fixed frequency before improvement of a physiological signal of class j characterized as the same physiological signal as a diagnostic index of a concurrent cardiovascular disease class, u j The incidence of physiological signals of class j, characterized as the same physiological signals as the diagnostic index of the concurrent cardiovascular disease class.
8. The smart terminal-based cardiovascular testing method of claim 7, wherein the real-time monitoring of the physiological signals with the same diagnostic index of the concurrent cardiovascular disease category at an improved fixed frequency comprises:
The monitoring terminal monitors a plurality of physiological signals of the target object, which are the same as the diagnosis indexes of the concurrent cardiovascular disease class, in real time at the improved fixed frequency, and synchronously extracts physiological signals which are the same as the diagnosis indexes of the concurrent cardiovascular disease class and exceed the health threshold;
adjusting the monitoring frequency of the physiological signals with the same super-health threshold value and the diagnostic index of the concurrent cardiovascular disease category from the fixed frequency to an adaptive frequency, and carrying out timing long monitoring on the physiological signals with the same super-health threshold value and the diagnostic index of the concurrent cardiovascular disease category by the adaptive frequency to obtain a group of diagnostic signals;
passing the diagnostic signal through a disease-seed identification model, wherein,
if the disease type identification model outputs the type of the cardiovascular disease of the target object, the type of the cardiovascular disease of the target object is synchronously fed back to the concurrent guidance thinking guide graph, and the steps S4 and S3 are executed to obtain all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease, physiological signals with the same diagnosis indexes of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease, and the monitoring frequency of the physiological signals with the same diagnosis indexes of all the type of the cardiovascular disease of the target object;
If the disease type identification model cannot output the type of the cardiovascular disease of the target object, the type of the cardiovascular disease of the target object is not required to be fed back to the concurrent guidance thought map synchronously, and the step S3 is executed to monitor the physiological signals with the same diagnostic indexes of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease in real time, so that the attention degree of all the type of the cardiovascular disease of the target object and the concurrent type of the cardiovascular disease is improved.
9. The smart terminal-based cardiovascular detection method of claim 8, further comprising:
and monitoring and early warning are carried out while the disease type identification model outputs the type of the cardiovascular disease, so as to remind a target object of timely medical intervention.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111569410.4A CN114271799B (en) | 2021-12-21 | 2021-12-21 | Cardiovascular detection method based on intelligent terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111569410.4A CN114271799B (en) | 2021-12-21 | 2021-12-21 | Cardiovascular detection method based on intelligent terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114271799A CN114271799A (en) | 2022-04-05 |
| CN114271799B true CN114271799B (en) | 2023-07-04 |
Family
ID=80873422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111569410.4A Active CN114271799B (en) | 2021-12-21 | 2021-12-21 | Cardiovascular detection method based on intelligent terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114271799B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012070969A1 (en) * | 2010-11-22 | 2012-05-31 | Farber Boris Slavinovich | Diagnostic method for predicting the development of cardiovascular diseases and monitoring treatment efficacy |
| RU2656560C1 (en) * | 2017-06-23 | 2018-06-05 | Юрий Ильич Гурфинкель | Method for assessment of risk of complications of cardiovascular diseases with associated pathology |
| WO2020072128A1 (en) * | 2018-10-06 | 2020-04-09 | Kinaptic, LLC | Methods and apparatus for a wearable electronic digital therapeutic device |
| CN110200619A (en) * | 2019-06-06 | 2019-09-06 | 中山大学孙逸仙纪念医院 | It is a kind of to establish major cardiovascular disease risks Early-warning Model method and device |
| CA3165746A1 (en) * | 2019-12-23 | 2021-07-01 | Analytics For Life Inc. | Method and system for signal quality assessment and rejection using heart cycle variability |
-
2021
- 2021-12-21 CN CN202111569410.4A patent/CN114271799B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114271799A (en) | 2022-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3803709B1 (en) | Detecting abnormalities in ecg signals | |
| US11617528B2 (en) | Systems and methods for reduced lead electrocardiogram diagnosis using deep neural networks and rule-based systems | |
| EP3801238A1 (en) | Detecting abnormalities in ecg signals | |
| CN104173043A (en) | Electrocardiogram (ECG) data analysis method suitable for mobile platform | |
| CN103970975B (en) | Electrocardiogram (ECG) data processing method and system | |
| CN107530015B (en) | Vital sign analysis method and system | |
| CN108836310A (en) | A kind of method and system for judging automatically user's electrocardio state based on artificial intelligence | |
| CN107692988A (en) | The method of monitoring of blood pressure method and graphic software platform based on intelligent platform | |
| CN113995419A (en) | Atrial fibrillation risk prediction system based on heartbeat rhythm signal and application thereof | |
| CN105611872A (en) | An apparatus and method for evaluating multichannel ECG signals | |
| Zaman et al. | Estimating reliability of signal quality of physiological data from data statistics itself for real-time wearables | |
| CN106096544B (en) | Contactless blink and heart rate combined detection system and method based on second-order blind identification | |
| Lei et al. | AFC-ECG: An adaptive fuzzy ECG classifier | |
| CN114271799B (en) | Cardiovascular detection method based on intelligent terminal | |
| CN110464369A (en) | A kind of state of mind based on human body physical sign numerical value judges algorithm | |
| CN109589137A (en) | Method, apparatus, terminal and the computer readable storage medium of movement of the foetus identification | |
| CN106361327B (en) | Waking state detection method and system in sleep state analysis | |
| CN112957018A (en) | Heart state detection method and device based on artificial intelligence | |
| CN113647959A (en) | Method, device and equipment for identifying waveform of electrocardiographic waveform signal | |
| TW202247816A (en) | Non-contact heart rhythm category monitoring system and method | |
| US20220359073A1 (en) | Edge-intelligent Iot-based Wearable Device For Detection of Cravings in Individuals | |
| Moukadem et al. | Automatic heart sound analysis module based on Stockwell transform | |
| CN116115201B (en) | Physical health state assessment system | |
| TWI765420B (en) | Assembly of heart failure prediction | |
| CN117457229B (en) | Anesthesia depth monitoring system and method based on artificial intelligence |
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 |