Intelligent medical remote diagnosis system based on multi-source data fusion

Abstract

The invention discloses an intelligent medical remote diagnosis system based on multi-source data fusion, and relates to the field of intelligent medical treatment, and the system comprises a sensing module which is used for configuring an operation cycle, and sensing the physical state information of a user in real time based on the operation cycle; the screening module is used for traversing the user body state information sensed by the sensing module, and screening effective information from all the user body state information; according to the system, various physical state information related to sleep is collected in real time, including electroencephalogram, electrocardiogram, oxygen concentration and the like, so that the sleep condition of a patient is comprehensively monitored, the system adopts specific screening and analysis methods in data processing, key information can be accurately extracted, sleep risks can be deeply analyzed, and the sleep quality of the patient is improved. By means of recording and comparison of historical analysis results, the system can timely diagnose whether the current sleep of the patient has risks or not and can also judge whether the patient has risk tendency or not.

Description

Technical Field

[0001] The present invention relates to the field of smart medical technology, and in particular to a smart medical remote diagnosis system based on multi-source data fusion. Background Art

[0002] Sleep apnea is a common sleep disorder in which the upper airway collapses and becomes blocked during sleep, causing apnea and hypopnea. Symptoms include snoring, respiratory interruptions, and daytime sleepiness. Long-term use can lead to high blood pressure, heart disease, diabetes, and other conditions, and can also affect cognitive function.

[0003] The invention patent application with application number 202111322558.8 discloses a sleep monitoring method, which is applied to a monitoring system for monitoring a patient's vital signs. The method includes: obtaining a monitoring signal of the monitored object, and the monitoring signal includes at least one of the following signals: electrocardiogram signal, respiratory signal, volume wave, blood pressure signal, body temperature signal, EEG signal, and motion signal; based on the monitoring signal, displaying at least one physiological parameter information of the monitored object; extracting one or more sleep-related feature information of one or more monitoring signals; based on each of the feature information and the sleep stage model corresponding to each of the feature information, determining the estimated sleep state corresponding to each of the feature information; determining the sleep state of the monitored object based on the estimated sleep state corresponding to each of the feature information; outputting a display of the sleep state of the monitored object Sleep state, this application aims to solve the problem that "the current sleep staging technology in hospitals is mainly through polysomnography, which monitors a variety of signals and requires a special sleep laboratory. Finally, relevant experts need to manually label them before doctors can make a diagnosis. Polysomnography is not only complex in system and imposes high constraints on patients, but also expensive and cannot meet the monitoring needs of most patients in the hospital; there are many consumer-grade sleep staging technologies on the market, mainly through smart bracelets and forehead-mounted brain electrodes. The physiological signals it collects are limited, so the accuracy of sleep staging is also poor, which cannot meet the needs of medical sleep staging monitoring; because polysomnography and consumer-grade sleep monitoring technology cannot meet the needs of most departments in the hospital, clinicians often use subjective questioning of patients to understand their sleep conditions, but the results of subjective questioning are not accurate and reliable."

[0004] However, to improve the accuracy of the diagnosis results for patients with sleep apnea syndrome, long-term sleep monitoring of the patients is necessary. However, patients with such diseases are often reluctant to undergo long-term sleep monitoring during hospitalization.

[0005] To this end, we proposed a smart medical remote diagnosis system based on multi-source data fusion. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides a smart medical remote diagnosis system based on multi-source data fusion, which can effectively solve the problems of the prior art.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] The present invention discloses a smart medical remote diagnosis system based on multi-source data fusion, comprising:

[0009] The perception module is used to configure the operation cycle and perceive the user's physical condition information in real time based on the operation cycle; the screening module is used to traverse the user's physical condition information perceived by the perception module and filter out valid information from all the user's physical condition information; the analysis module is used to receive the user's physical condition information filtered by the screening module and analyze the user's sleep risk step by step based on the user's physical condition information; the recording module is used to record the historical user sleep risk analysis results of the analysis module and record the analysis results; the diagnosis module is used to receive the user's sleep risk analysis results output in real time by the analysis module and the user's sleep risk history analysis results recorded in the recording module, and diagnose whether the user's sleep is at risk and whether there is a risk tendency.

[0010] Furthermore, the sensing module is integrated with a bioelectric sensor, an oxygen concentration sensor, a blood oxygen saturation sensor, and an acoustic sensor. The user's physical status information sensed by the sensing module includes: electroencephalogram, electromyogram, electrocardiogram, oxygen concentration, blood oxygen saturation, and snoring audio spectrum;

[0011] The bioelectric sensors are provided in at least three groups, which are respectively deployed on the user's head, chest in front of the heart, and limbs. The blood oxygen saturation sensor is integrated into a bracelet worn on the user's hand. The acoustic sensor is deployed within two meters of the user. The oxygen concentration sensor is deployed between the user's two nostrils.

[0012] The sensing module is internally provided with a storage unit, and the storage unit is used to receive the user's physical state information sensed by the sensing module and store the user's physical state information;

[0013] When storing the user's body state information, the storage unit differentiates and stores the user's body state information based on the type of the user's body state information, and sorts and stores the user's body state information based on the perception period of the user's body state information.

[0014] Furthermore, the operating cycle configured by the perception module when perceiving the user's physical status information complies with:

[0015]

[0016] Where: T is the operating cycle of the perception module; T0 is the system preset operating cycle base of the perception module; Customize the average daily sleep time for users; T norr Plan sleeping time for the user;

[0017] in, and t norr Actively input by the user, T0 is customized by the system user, and T0 ≥ 5min / time, and t norr / T≥4, at t norr / If the condition of T≥4 is not met, then T norr The setting is invalid and the system ends synchronously.

[0018] Furthermore, during the operation phase of the screening module, the user physical status information stored in the storage unit is synchronously traversed to identify the number of user physical status information of the same type in each differentiated storage interval. When the number of user physical status information of the same type in each differentiated storage interval is 4, the first and last two user physical status information in each differentiated storage interval are discarded, and the remaining user physical status information is regarded as valid information;

[0019] When the number of the same type of user physical status information in each differentiated storage interval is not less than 5, the first and last two pieces of user physical status information in each differentiated storage interval are discarded, and then the remaining user physical status information is picked up at intervals, so that the piece of user physical status information between the two adjacent picked user physical status information is used as the user physical status information discarded in the second round. This operation is repeated until there are two or three pieces of user physical status information remaining in each differentiated storage interval, that is, the valid information is obtained by screening.

[0020] Furthermore, an early warning module is provided between the screening module and the storage unit, and an emergency contact address book is stored in the early warning module. An early warning trigger interval is provided in the early warning module, and the early warning trigger interval corresponds one-to-one to each type of user physical status information. The early warning module is used to forward the user physical status information stored in the storage unit, and after receiving the user physical status information, compare the latest information of each type of user physical status information with the corresponding early warning trigger interval. When any type of information meets the early warning trigger interval, the early warning module sends a communication request to all emergency contacts in the emergency contact address book at the same time. When all types of information do not meet the early warning trigger interval, the received user physical status information is further forwarded to the screening module;

[0021] Among them, the early warning module is integrated by any remote communication equipment.

[0022] Furthermore, a classification unit is provided at a lower level of the analysis module, and the classification unit is used to obtain the user's physical state information received by the analysis module, and classify the electroencephalogram, electromyogram, electrocardiogram, and blood oxygen saturation in the user's physical state information into Class I, and oxygen concentration and snoring audio spectrum into Class II;

[0023] The logic expression of the analysis module analyzing the user's sleep risk based on the Class I user's physical condition information is as follows:

[0024]

[0025] Where: e REM 、e NREM is the ratio of the duration of the rapid eye movement period to the non-rapid eye movement period in the sleep cycle in the electroencephalogram; m is the normalized value of the muscle electrical activity intensity during sleep; h HRV is the normalized value of heart rate variability during sleep; s is the normalized value of blood oxygen saturation during sleep;

[0026] The larger the R is, the higher the sleep risk of the user is, and vice versa, the healthier the user's sleep is.

[0027]

[0028] Among them, t REM represents the duration of rapid eye movement, t NREM represents the duration of the non-rapid eye movement period, T represents the time span covered by all user physical status information, m avg Represents the average value of muscle electrical activity during sleep, m max Indicates the preset maximum value of muscle electrical activity during sleep; min Indicates the intensity of muscle electrical activity during sleep, h HRV The heart rate variability is calculated by taking the ratio of the standard deviation of the difference between adjacent normal heartbeat cycles to the average heartbeat cycle, and then normalizing it to the interval [0, 1].

[0029] Furthermore, the analysis module analyzes the user's sleep risk based on the Class II user's physical condition information in the following logical expression:

[0030]

[0031] Where: C O 、L O 、D S 、f S The cumulative deviation of oxygen concentration per unit time, the ratio of the longest time when oxygen concentration is continuously lower than the threshold to the total monitoring time, the discrete degree of energy distribution of the snoring audio spectrum band, and the normalized value of the frequency of snoring per unit time;

[0032] in, Where u is the total amount of oxygen concentration information around the user's nostrils sensed in a single cycle; ΔO v is the difference between the ith oxygen concentration information and the preset normal oxygen concentration threshold, C O 、L O 、D S Before being applied to the above formula, all values ​​are constrained to be in the range of [0, 1] through normalization.

[0033] Furthermore, when calculating R, the calculation is based on the information of the same period in each partitioned storage interval, and then R(I)1, R(I)2, R(I)3, ... are calculated based on each piece of information in each partitioned storage interval. Similarly, F is calculated as F(II)1, F(II)2, F(II)3, ...;

[0034] The step-by-step analysis results of the user's sleep risk in the analysis module are as follows:

[0035] Q=MAX[(R(I)1·F(II)1), (R(I)2·F(II)2), (R(I)3·F(II)3),...];

[0036] Wherein, MAX[·] means taking the maximum value in the brackets, and R(I)·F(II) means the product operation of R(I) and F(II).

[0037] Furthermore, a risk determination threshold is set in the diagnosis module, and the diagnosis module compares the risk determination threshold with the user's sleep risk obtained by the latest analysis of the analysis module to diagnose whether the user is at risk;

[0038] The diagnosis module arranges the received user sleep risk history analysis results based on time sequence, and diagnoses the user as having a sleep risk tendency when any three consecutive analysis results are identified to be showing a continuous upward trend; otherwise, the user is diagnosed as not having a sleep risk tendency;

[0039] Among them, when the diagnosis module diagnoses that the user's sleep is at risk or has a risk tendency, it triggers the early warning module to run.

[0040] Furthermore, the perception module is connected to the screening module through the network, the perception module is internally connected to a storage unit through the network, the storage unit is connected to the early warning module through the network, the screening module is connected to the analysis module through the network, the analysis module is lower-level connected to the classification unit through the network, and the analysis module is connected to the recording module and the diagnosis module through the network.

[0041] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0042] The present invention provides an intelligent medical remote diagnosis system based on multi-source data fusion. During operation, the system realizes comprehensive monitoring of the patient's sleep condition by real-time collection of various sleep-related physical status information, including electroencephalogram, electrocardiogram, oxygen concentration, etc. In data processing, the system adopts specific screening and analysis methods, which can accurately extract key information and deeply analyze sleep risks. With the help of recording and comparing historical analysis results, the system can not only timely diagnose whether the patient's current sleep is risky, but also determine whether there is a risk tendency. The whole process does not require the patient to be hospitalized for long-term monitoring, but can be completed in the patient's daily life scene, which greatly reduces the patient's resistance and improves the accuracy of the diagnosis results, providing patients with sleep apnea syndrome with a more convenient, efficient and accurate diagnosis service. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0044] Figure 1 This is a structural diagram of the smart medical remote diagnosis system based on multi-source data fusion. DETAILED DESCRIPTION

[0045] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] The present invention will be further described below with reference to the embodiments.

[0047] Example:

[0048] The smart medical remote diagnosis system based on multi-source data fusion in this embodiment is as follows: Figure 1 As shown, the system is used for users with sleep apnea syndrome, including:

[0049] A perception module is used to configure an operation cycle and perceive the user's physical status information in real time based on the operation cycle;

[0050] The perception module is integrated with bioelectric sensors, oxygen concentration sensors, blood oxygen saturation sensors, and acoustic sensors. The user's physical status information sensed by the perception module includes: electroencephalogram, electromyogram, electrocardiogram, oxygen concentration, blood oxygen saturation, and snoring audio spectrum;

[0051] At least three groups of bioelectric sensors are deployed on the user's head, chest in front of the heart, and limbs. The blood oxygen saturation sensor is integrated into a bracelet worn on the user's hand. The acoustic sensor is deployed within two meters of the user. The oxygen concentration sensor is deployed between the user's two nostrils.

[0052] The sensing module is internally provided with a storage unit, which is used to receive the user's physical state information sensed by the sensing module and store the user's physical state information;

[0053] When storing the user's physical state information, the storage unit differentiates and stores the user's physical state information based on its type, and sorts and stores the user's physical state information based on its perception period.

[0054] The operating cycle configured by the perception module when perceiving the user's physical status information follows:

[0055]

[0056] Where: T is the operating cycle of the perception module; T0 is the system preset operating cycle base of the perception module; Customize the average daily sleep time for users; T norr Plan sleeping time for the user;

[0057] in, and t norr Actively input by the user, T0 is customized by the system user, and T0 ≥ 5min / time, and t norr / T≥4, at t norr / If the condition of T≥4 is not met, then T norr The setting is invalid and the system ends running synchronously;

[0058] By calculating the above logical formula, the operating cycle of the perception module is designed to ensure that the perception module operates based on the operating cycle and can obtain more effective and valuable information for further diagnosis of system operation;

[0059] A screening module is used to traverse the user's physical status information sensed by the perception module and filter out valid information from all the user's physical status information;

[0060] During the operation phase of the screening module, the user physical status information stored in the storage unit is traversed synchronously to identify the number of user physical status information of the same type in each storage interval. When the number of user physical status information of the same type in each storage interval is 4, the first and last two user physical status information in each storage interval are discarded, and the remaining user physical status information is regarded as valid information;

[0061] When the number of user physical status information of the same type in each differentiated storage interval is not less than 5, the first and last two pieces of user physical status information in each differentiated storage interval are discarded, and the remaining user physical status information is then picked up at intervals, with the piece of user physical status information between the two adjacent picked up pieces of user physical status information being used as the user physical status information to be discarded in the second round. This operation is repeated until there are two or three pieces of user physical status information remaining in each differentiated storage interval, indicating that valid information has been screened out.

[0062] An early warning module is provided between the screening module and the storage unit. An emergency contact address book is stored in the early warning module. An early warning trigger interval is provided in the early warning module. The early warning trigger interval corresponds one-to-one to each type of user physical status information. The early warning module is used to forward the user physical status information stored in the storage unit. After receiving the user physical status information, the latest information of each type of user physical status information is compared with the corresponding early warning trigger interval. When any type of information meets the early warning trigger interval, the early warning module sends a communication request to all emergency contacts in the emergency contact address book at the same time. When all types of information do not meet the early warning trigger interval, the received user physical status information is further forwarded to the screening module;

[0063] Among them, the early warning module is integrated by any remote communication device;

[0064] An analysis module is configured to receive the user's physical condition information filtered by the screening module and analyze the user's sleep risk step by step based on the user's physical condition information;

[0065] The analysis module is provided with a classification unit at the lower level. The classification unit is used to obtain the user's physical status information received by the analysis module, and classify the user's physical status information such as electroencephalogram, electromyogram, electrocardiogram, and blood oxygen saturation into Class I, and oxygen concentration and snoring audio spectrum into Class II;

[0066] The analysis module analyzes the user's sleep risk based on the Class I user's physical condition information as follows:

[0067]

[0068] Where: e REM 、e NREMis the ratio of the duration of the rapid eye movement period to the non-rapid eye movement period in the sleep cycle in the electroencephalogram; m is the normalized value of the muscle electrical activity intensity during sleep; h HRV is the normalized value of heart rate variability during sleep; s is the normalized value of blood oxygen saturation during sleep;

[0069] The larger the R is, the higher the user's sleep risk is, and vice versa, the healthier the user's sleep is.

[0070]

[0071] Among them, t REM represents the duration of rapid eye movement, t NREM represents the duration of the non-rapid eye movement period, T represents the time span covered by all user physical status information, m avg Represents the average value of muscle electrical activity during sleep, m max Indicates the preset maximum value of muscle electrical activity during sleep; min Indicates the intensity of muscle electrical activity during sleep, h HRV The heart rate variability is obtained by taking the ratio of the standard deviation of the difference between adjacent normal heartbeat cycles and the average heartbeat cycle, and then normalizing it to the interval [0, 1].

[0072] The analysis module analyzes the user's sleep risk based on the physical status information of Class II users as follows:

[0073]

[0074] Where: C O 、L O 、D S 、f S The cumulative deviation of oxygen concentration per unit time, the ratio of the longest time when oxygen concentration is continuously lower than the threshold to the total monitoring time, the discrete degree of energy distribution of the snoring audio spectrum band, and the normalized value of the frequency of snoring per unit time;

[0075] in, Where u is the total amount of oxygen concentration information around the user's nostrils sensed in a single cycle; ΔO v is the difference between the ith oxygen concentration information and the preset normal oxygen concentration threshold, C O 、L O 、D S Before being applied to the above formula, all values ​​are constrained to be within the range of [0, 1] through normalization processing;

[0076] Through the above logical formula, the user's sleep analysis is analyzed step by step based on the two types of user physical status information, providing support for further comprehensive analysis of the user's sleep risks;

[0077] When R is calculated, it is calculated based on the information of the same period in each partition storage interval. Then, based on each piece of information in each partition storage interval, R(I)1, R(I)2, R(I)3, ... are calculated. Similarly, F is calculated as F(II)1, F(II)2, F(II)3, ...;

[0078] The step-by-step analysis results of user sleep risk in the analysis module are as follows:

[0079] Q=MAX[(R(I)1·F(II)1), (R(I)2·F(II)2), (R(I)3·F(II)3),...];

[0080] Wherein, MAX[·] means taking the maximum value in the brackets, and R(I)·F(II) means the product operation of R(I) and F(II);

[0081] Through the above logical formula calculation, the user's sleep risk is finally analyzed;

[0082] The recording module is used to record the historical user sleep risk analysis results of the analysis module and record the analysis results;

[0083] The diagnosis module is used to receive the user sleep risk analysis results output in real time by the analysis module and the user sleep risk history analysis results recorded by the recording module, and diagnose whether the user has sleep risks and whether there are risk tendencies;

[0084] The diagnosis module sets a risk determination threshold. The diagnosis module compares the risk determination threshold with the user's sleep risk obtained from the latest analysis of the analysis module to diagnose whether the user is at risk.

[0085] The diagnosis module arranges the received user sleep risk history analysis results based on time sequence. When any three consecutive analysis results are identified to show a continuous upward trend, the user is diagnosed with a sleep risk tendency. Otherwise, the user is diagnosed with no sleep risk tendency.

[0086] Among them, when the diagnosis module diagnoses that the user has a risk or a risk tendency in sleeping, it triggers the early warning module to run;

[0087] The perception module is connected to the screening module through the network. The perception module is internally connected to a storage unit through the network. The storage unit is connected to the early warning module through the network. The screening module is connected to the analysis module through the network. The analysis module is connected to the classification unit through the network. The analysis module is connected to the recording module and the diagnosis module through the network.

[0088] In this embodiment, the perception module runs a configuration operation cycle, and perceives the user's physical state information in real time based on the operation cycle. The storage unit synchronously receives the user's physical state information perceived by the perception module and stores the user's physical state information. The early warning module forwards the user's physical state information stored in the storage unit in real time. After receiving the user's physical state information, the latest information of each type of user's physical state information is compared with the corresponding early warning trigger interval. When any type of information meets the early warning trigger interval, the early warning module sends a communication request to all emergency contacts in the emergency contact address book at the same time. When all types of information do not meet the early warning trigger interval, the received user's physical state information is further forwarded to the screening module. The screening module is post-operated to traverse the information perceived by the perception module. User physical status information, valid information is filtered out from all user physical status information, and then the analysis module receives the user physical status information filtered by the screening module, and analyzes the user sleep risk step by step based on the user physical status information. The classification unit synchronously obtains the user physical status information received by the analysis module, and classifies the electroencephalogram, electromyogram, electrocardiogram, and blood oxygen saturation in the user physical status information into Class I, and the oxygen concentration and snoring audio spectrum into Class II. The recording module then records the historical user sleep risk analysis results of the analysis module and records the analysis results. Finally, the diagnosis module receives the user sleep risk analysis results output in real time by the analysis module and the user sleep risk historical analysis results recorded in the recording module to diagnose whether there is a risk in the user's sleep and whether there is a risk tendency.

[0089] Through the operation of the system in the above embodiment, remote diagnosis services are provided for patients with sleep apnea syndrome, which facilitates the patient's diagnosis process and provides maintenance effects for the patient's sleep health.

[0090] In summary, during operation, the system in the above embodiment collects a variety of sleep-related physical status information in real time, including electroencephalogram, electrocardiogram, oxygen concentration, etc., to achieve comprehensive monitoring of the patient's sleep condition. In data processing, the system adopts specific screening and analysis methods, which can accurately extract key information and deeply analyze sleep risks. With the help of recording and comparing historical analysis results, the system can not only timely diagnose whether the patient's current sleep is risky, but also determine whether there is a risk tendency. The entire process does not require the patient to be hospitalized for long-term monitoring, but can be completed in the patient's daily life scenes, which greatly reduces the patient's resistance and improves the accuracy of the diagnosis results, providing patients with sleep apnea syndrome with a more convenient, efficient and accurate diagnostic service.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. The intelligent medical remote diagnosis system based on multi-source data fusion is characterized by: The system is used for users with sleep apnea syndrome, including: A perception module is used to configure an operation cycle and perceive the user's physical status information in real time based on the operation cycle; A screening module is used to traverse the user's physical status information sensed by the perception module and filter out valid information from all the user's physical status information; An analysis module is configured to receive the user's physical condition information filtered by the screening module and analyze the user's sleep risk step by step based on the user's physical condition information; The recording module is used to record the historical user sleep risk analysis results of the analysis module and record the analysis results; The diagnosis module is used to receive the user sleep risk analysis results output in real time by the analysis module and the user sleep risk history analysis results recorded in the recording module, and diagnose whether the user has sleep risks and whether there are risk tendencies.

2. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: The sensing module is integrated with a bioelectric sensor, an oxygen concentration sensor, a blood oxygen saturation sensor, and an acoustic sensor. The user's physical status information sensed by the sensing module includes: electroencephalogram, electromyogram, electrocardiogram, oxygen concentration, blood oxygen saturation, and snoring audio spectrum; The bioelectric sensors are provided in at least three groups, which are respectively deployed on the user's head, chest in front of the heart, and limbs. The blood oxygen saturation sensor is integrated into a bracelet worn on the user's hand. The acoustic sensor is deployed within two meters of the user. The oxygen concentration sensor is deployed between the user's two nostrils. The sensing module is internally provided with a storage unit, and the storage unit is used to receive the user's physical state information sensed by the sensing module and store the user's physical state information; When storing the user's body state information, the storage unit differentiates and stores the user's body state information based on the type of the user's body state information, and sorts and stores the user's body state information based on the perception period of the user's body state information.

3. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: The operating cycle configured by the perception module when perceiving the user's physical status information complies with: Where: T is the operating cycle of the perception module; T0 is the system preset operating cycle base of the perception module; Customize the average daily sleep time for users; T norr Plan sleeping time for the user; in, and t norr Actively input by the user, T0 is customized by the system user, and T0 ≥ 5min / time, and t norr / T≥4, at t norr / If the condition of T≥4 is not met, then T norr The setting is invalid and the system ends synchronously.

4. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: During the operation phase of the screening module, the user physical status information stored in the storage unit is synchronously traversed to identify the number of user physical status information of the same type in each differentiated storage interval. When the number of user physical status information of the same type in each differentiated storage interval is 4, the first and last two user physical status information in each differentiated storage interval are discarded, and the remaining user physical status information is regarded as valid information; When the number of the same type of user physical status information in each differentiated storage interval is not less than 5, the first and last two pieces of user physical status information in each differentiated storage interval are discarded, and then the remaining user physical status information is picked up at intervals, so that the piece of user physical status information between the two adjacent picked user physical status information is used as the user physical status information discarded in the second round. This operation is repeated until there are two or three pieces of user physical status information remaining in each differentiated storage interval, that is, the valid information is obtained by screening.

5. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: An early warning module is provided between the screening module and the storage unit. An emergency contact address book is stored in the early warning module. An early warning trigger interval is provided in the early warning module. The early warning trigger interval corresponds one-to-one to each type of user physical status information. The early warning module is used to forward the user physical status information stored in the storage unit. After receiving the user physical status information, the latest information of each type of user physical status information is compared with the corresponding early warning trigger interval. When any type of information meets the early warning trigger interval, the early warning module sends a communication request to all emergency contacts in the emergency contact address book at the same time. When all types of information do not meet the early warning trigger interval, the received user physical status information is further forwarded to the screening module; Among them, the early warning module is integrated by any remote communication equipment.

6. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: The analysis module is provided with a classification unit at a lower level, which is used to obtain the user's physical status information received by the analysis module, and classify the user's physical status information such as electroencephalogram, electromyogram, electrocardiogram, and blood oxygen saturation into Class I, and oxygen concentration and snoring audio spectrum into Class II; The logic expression of the analysis module analyzing the user's sleep risk based on the Class I user's physical condition information is as follows: Where: e REM 、e NREM is the ratio of the duration of the rapid eye movement period to the non-rapid eye movement period in the sleep cycle in the electroencephalogram; m is the normalized value of the muscle electrical activity intensity during sleep; h HRV is the normalized value of heart rate variability during sleep; s is the normalized value of blood oxygen saturation during sleep; The larger the R is, the higher the sleep risk of the user is, and vice versa, the healthier the user's sleep is. e REM =t REM / T;e NREM =t NREM / T;m=m avg / m max ; s=(100-s min ) / 100;; Among them, t REM represents the duration of rapid eye movement, t NREM represents the duration of the non-rapid eye movement period, T represents the time span covered by all user physical status information, m avg Represents the average value of muscle electrical activity during sleep, m max Indicates the preset maximum value of muscle electrical activity during sleep; min Indicates the intensity of muscle electrical activity during sleep, h HRV The heart rate variability is calculated by taking the ratio of the standard deviation of the difference between adjacent normal heartbeat cycles to the average heartbeat cycle, and then normalizing it to the interval [0, 1].

7. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 6 is characterized in that: The logic expression of the analysis module analyzing the user's sleep risk based on the Class II user's physical status information is as follows: Where: C O , L O 、D S 、f S The cumulative deviation of oxygen concentration per unit time, the ratio of the longest time when oxygen concentration is continuously lower than the threshold to the total monitoring time, the discrete degree of energy distribution of the snoring audio spectrum band, and the normalized value of the frequency of snoring per unit time; in, Where u is the total amount of oxygen concentration information around the user's nostrils sensed in a single cycle; ΔO v is the difference between the ith oxygen concentration information and the preset normal oxygen concentration threshold, C O , L O 、D S Before being applied to the above formula, all values ​​are constrained to be in the range of [0, 1] through normalization.

8. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 7 is characterized in that: When calculating R, the calculation is based on the information of the same period in each partitioned storage interval, and then R(I)1, R(I)2, R(I)3, ... are calculated based on each piece of information in each partitioned storage interval. Similarly, F is calculated as F(II)1, F(II)2, F(II)3, ...; The step-by-step analysis results of the user's sleep risk in the analysis module are as follows: Q=MAX[(R(I)1·F(II)1), (R(I)2·F(II)2), (R(I)3·F(II)3),...]; Wherein, MAX[·] means taking the maximum value in the brackets, and R(I)·F(II) means the product operation of R(I) and F(II).

9. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: The diagnosis module is set with a risk determination threshold, and the diagnosis module compares the risk determination threshold with the user's sleep risk obtained by the latest analysis of the analysis module to diagnose whether the user is at risk; The diagnosis module arranges the received user sleep risk history analysis results based on time sequence, and diagnoses the user as having a sleep risk tendency when any three consecutive analysis results are identified to be showing a continuous upward trend; otherwise, the user is diagnosed as not having a sleep risk tendency; Among them, when the diagnosis module diagnoses that the user's sleep is at risk or has a risk tendency, it triggers the early warning module to run.

10. The intelligent medical remote diagnosis system based on multi-source data fusion according to claim 1 is characterized in that: The perception module is connected to the screening module via the network, the perception module is internally connected to a storage unit via the network, the storage unit is connected to the early warning module via the network, the screening module is connected to the analysis module via the network, the analysis module is lower-level connected to the classification unit via the network, and the analysis module is connected to the recording module and the diagnosis module via the network.

Images