Failure patient tracking and analyzing system based on intelligent chip

Abstract

The invention discloses a disabled patient tracking and analysis system based on an intelligent chip, and relates to the technical field of brain diseases, comprising the steps of establishing a brain disease portrait according to brain disease information and body function information of a disabled patient, and establishing an interception time window through the brain disease portrait. The disease change influence condition of the patient is described through the disease portrait, an interception time window is determined, and some symptoms of some patients before disease attack can be accurately captured through the window, so that early warning of disease attack is helped. The state relative matching degree of the real-time state feature set and the standard morbidity state feature set is calculated, the current state condition of the patient is analyzed through the state relative matching degree, comprehensive early warning is conducted on the patient in combination with the state relative matching degree and the state transition probability, the tracking effect and the early warning success rate of the disabled patient are improved, and the early warning efficiency is improved. And the relative safety of the disabled patient is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of brain diseases, and in particular to a tracking and analyzing system for disabled patients based on a smart chip. Background Art

[0002] The tracking and analysis of disabled patients with smart chips mainly relies on the development of the Internet of Things, artificial intelligence and big data processing technology. As a core component, smart chips can monitor the patient's physiological information, brain wave information, location information, etc. in real time, and transmit these data to the cloud server or designated receiving device through the wireless communication module. The cloud server uses big data processing technology and artificial intelligence algorithms to conduct in-depth analysis of these data, identify abnormal physiological indicators or behavior patterns, and issue warnings or take intervention measures in a timely manner. The program aims to improve the quality of life of disabled patients, ensure their safety and health, and reduce the workload of medical staff. Through real-time tracking and analysis of smart chips, medical staff can more accurately grasp the patient's condition and provide patients with personalized rehabilitation programs and service plans.

[0003] In the prior art, analysis is often performed based only on the patient's physiological information, which makes the analysis results inaccurate. The physiological information and brain information are not combined for analysis, resulting in poor tracking effects on disabled patients and low early warning success rates.

[0004] Therefore, how to improve the tracking effect and early warning success rate of disabled patients is a technical problem that needs to be solved. Summary of the invention

[0005] The purpose of the present invention is to solve the problems of poor tracking effect and low early warning success rate of disabled patients in the prior art, and to propose a disabled patient tracking and analysis system based on a smart chip, which includes: A portrait module is used to collect brain disease information and body function information of disabled patients, establish a brain disease portrait based on the brain disease information and body function information of disabled patients, and establish a capture time window through the brain disease portrait; A generation module is used to collect the previous disease state information of disabled patients, construct the disease state information timeline through the previous disease state information, and intercept the disease state information timeline by intercepting the time window to obtain the disease state sample information, and extract the disease state features based on the disease state sample information to construct a standard disease state feature set; A calculation module is used to monitor the real-time status information of disabled patients, extract status features from the real-time status information to construct a real-time status feature set, calculate the relative matching degree between the real-time status feature set and the standard disease status feature set, and determine the warning information according to the relative matching degree; The early warning module is used to send early warning information to multiple monitoring parties, so as to facilitate the monitoring parties to take further preventive measures; Among them, the status information of disabled patients includes brain information and physiological information.

[0006] In some embodiments of the present application, the portrait module is used to establish a brain disease portrait based on the brain disease information and body function information of the disabled patient, which is specifically: The brain disease information and body function information are preprocessed, and the brain disease characteristics and body function characteristics are extracted. The brain disease model and body model are constructed respectively by fusion of the brain disease characteristics and body function characteristics. The impact of lesion evolution on the body model is simulated according to the brain disease model and the body model. The abnormal signs of brain information and physiological information generated in this impact process are analyzed to determine the advance time under different lesion stages, so as to construct a matching relationship between lesion stage and advance time. The current lesion stage, the matching relationship between lesion stage and advance time, and the simulated impact process of lesion evolution on the body model and the abnormal signs are used as the content of the brain disease portrait.

[0007] In some embodiments of the present application, the portrait module is used to establish a capture time window through the brain disease portrait, which is specifically: The interception time window includes two parts: the pre-onset part and the onset part. The advance time node of the pre-onset part is determined by the matching relationship between the current lesion stage and the lesion stage-advance time in the brain disease portrait. The time of the pre-onset part is from the advance time node to the onset start time node, and the time of the onset part is from the onset start time node to the onset end time node.

[0008] In some embodiments of the present application, the generation module is used to construct a timeline of the onset state information through the previous onset state information, and intercept the onset state information timeline by means of the interception time window to obtain the onset state sample information, which is specifically: Arrange the previous disease status information in chronological order and display it in the form of a timeline to construct a disease status information timeline. By intercepting the time nodes corresponding to the pre-onset part and the onset part under the time window, the disease status information timeline is intercepted to obtain the disease status sample information.

[0009] In some embodiments of the present application, the generation module is further used to extract the disease state features based on the disease state sample information to construct a standard disease state feature set, which is specifically: Determine the state characteristics of disabled patients in a non-ill state through brain disease portraits, and use the state characteristics of disabled patients in a non-ill state as the baseline state characteristics; All abnormal state features are extracted from the disease state sample information, and the baseline state features are compared with the abnormal state features to screen out the disease state features and construct a standard disease state feature set. The standard disease state feature set includes a brain disease feature set and a physiological disease feature set.

[0010] In some embodiments of the present application, the computing module includes a monitoring unit, a matching unit and a transfer unit. The monitoring unit is used to extract state features from the real-time state information to construct a real-time state feature set, which is specifically: The real-time state information is real-time brain information and physiological information, and real-time brain features and real-time physiological features are respectively extracted from the real-time brain information and physiological information, and a real-time brain feature set and a real-time physiological feature set are respectively constructed through the real-time brain features and the real-time physiological features. The real-time state feature set includes the real-time brain feature set and the real-time physiological feature set.

[0011] In some embodiments of the present application, the matching unit is used to calculate the relative matching degree between the real-time state feature set and the standard disease state feature set, which is specifically: Calculate the characteristic variation coefficient of the real-time brain characteristic set and the real-time physiological characteristic set, calculate the matching degree between each corresponding characteristic according to the real-time brain characteristic set and the real-time physiological characteristic set as well as the brain disease characteristic set and the physiological disease characteristic set, and calculate the relative matching degree of the state by combining the characteristic variation coefficient of the real-time brain characteristic set and the real-time physiological characteristic set and the matching degree between the corresponding characteristics; ;in, is the relative matching degree of the state, , are the matching conversion coefficients corresponding to the characteristics of brain information and physiological information, , are the number of feature types of the brain disease feature set and the physiological disease feature set, respectively. The first The combined weights of brain features, The first The degree of matching of brain features, , are the overall coefficient of variation of the real-time brain feature set and the real-time physiological feature set, respectively. , are preset constants, The first The combined weight of the physiological characteristics, The first Matching features of physiological features.

[0012] In some embodiments of the present application, the transfer unit is used to determine the warning information according to the relative matching degree of the states, which is specifically: If the state relative matching degree is greater than the state relative matching degree threshold, the warning information is determined according to the state relative matching degree and the specific matching features; Otherwise, the state transition probability from the real-time state feature set to the standard disease state feature set is calculated, and the warning information is determined according to the state transition probability.

[0013] In some embodiments of the present application, the transfer unit is specifically used to calculate the state transition probability from the real-time state feature set to the standard disease state feature set, and determine the warning information according to the state transition probability, which is specifically: The initial state transition probability from the real-time state feature set to the standard disease state feature set is calculated through the Markov chain, and the target state transition probability is determined according to the initial state transition probability and the relative state matching degree; ; in, is the target state transition probability, is the initial state transition probability, is the state transition probability correction, It represents the state transition probability correction amount obtained by mapping the relative matching degree of the state; The warning information is determined by the target state transition probability.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. Establish a brain disease portrait based on the brain disease information and body function information of disabled patients, and establish a time window through the brain disease portrait. The disease portrait describes the impact of the patient's disease changes and determines a time window. Through this window, some signs of some patients before the onset of the disease can be accurately captured, thereby helping to warn of the onset of the disease.

[0015] 2. Calculate the relative matching degree between the real-time state feature set and the standard disease state feature set, analyze the patient's current state through the relative matching degree, and conduct comprehensive early warning for the patient based on the relative matching degree and state transition probability, thereby improving the tracking effect and early warning success rate of disabled patients and ensuring the relative safety of disabled patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the structure of a disabled patient tracking and analysis system based on a smart chip proposed by the present invention; Figure 2 This is a schematic diagram of the structure of a computing module in a disabled patient tracking and analysis system based on a smart chip proposed by the present invention. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0018] Reference Figure 1 , the disabled patient tracking and analysis system based on smart chip includes the following modules or units, The portrait module is used to collect brain disease information and body function information of disabled patients, establish a brain disease portrait based on the brain disease information and body function information of disabled patients, and establish a capture time window through the brain disease portrait.

[0019] In this embodiment, the disabled patients are patients with brain diseases such as stroke and epilepsy. Such diseases may exist in two states. For example, epilepsy patients have obvious states of onset and non-onset. When the disease occurs, the patient will experience epileptic seizures, such as convulsions, loss of consciousness, etc.; when the disease does not occur, the patient may be completely normal and have no symptoms, which is closer to certain chronic or intermittent brain diseases. Brain disease information includes relevant data integrated from multiple data sources such as hospital information systems, medical imaging systems, and electronic medical record systems (brain imaging data (such as MRI, CT scans), neurological examination reports, medical history records, genetic information, living habits data, etc.). Physical function information includes basic information such as height, weight, and blood pressure.

[0020] In this embodiment, the status information of the disabled patient includes brain information (electroencephalogram (EEG) data) and physiological information (such as heart rate, respiratory rate, blood pressure, etc.). The smart chip first acts as a data collector, which can collect the patient's physiological data (such as heart rate, blood pressure, EEG, etc.) and location information in real time. This is the basis for the system to obtain the patient's real-time status. It has certain data processing and preliminary analysis capabilities. For example, the chip can perform simple filtering, signal amplification, or preliminary feature extraction tasks to reduce the amount of data that needs to be uploaded to the cloud and improve the accuracy and effectiveness of the data. The collected data needs to be transmitted to the central processing system (which may be a cloud server) in real time for further analysis and processing. The smart chip establishes communication with the cloud server through the Internet of Things technology to ensure the timely transmission of data.

[0021] In some embodiments of the present application, the portrait module is used to establish a brain disease portrait based on the brain disease information and body function information of the disabled patient, which is specifically: The brain disease information and body function information are preprocessed, and the brain disease characteristics and body function characteristics are extracted. The brain disease model and body model are constructed respectively by fusion of the brain disease characteristics and body function characteristics. The impact of lesion evolution on the body model is simulated according to the brain disease model and the body model. The abnormal signs of brain information and physiological information generated in this impact process are analyzed to determine the advance time under different lesion stages, so as to construct a matching relationship between lesion stage and advance time. The current lesion stage, the matching relationship between lesion stage and advance time, and the simulated impact process of lesion evolution on the body model and the abnormal signs are used as the content of the brain disease portrait.

[0022] In this embodiment, brain disease characteristics and body function characteristics, such as lesion location, size, morphology, pathological type, etc. Such as the mean, standard deviation, and change trend of physiological parameters. The brain disease characteristics and body function characteristics are integrated to form a feature set containing comprehensive information. This may involve methods such as feature splicing, feature weighting, and feature selection to optimize model performance. The brain disease model is used to simulate the evolution of the lesion, including the enlargement, reduction, and morphological changes of the lesion. The simulated lesion evolution process is input into the body model to analyze the impact of the lesion evolution on various body systems (such as the nervous system, circulatory system, respiratory system, etc.). During the simulation process, abnormal signs that may appear in the brain information and physiological information before the onset of the disease are identified and extracted, such as abnormal EEG, fluctuations in physiological parameters, etc. The time interval between abnormal signs and onset is integrated and analyzed to determine the lead time, and the lead time may be different at different lesion stages.

[0023] In some embodiments of the present application, the portrait module is used to establish a capture time window through the brain disease portrait, which is specifically: The interception time window includes two parts: the pre-onset part and the onset part. The advance time node of the pre-onset part is determined by the matching relationship between the current lesion stage and the lesion stage-advance time in the brain disease portrait. The time of the pre-onset part is from the advance time node to the onset start time node, and the time of the onset part is from the onset start time node to the onset end time node.

[0024] In this embodiment, the intercepted time window includes two parts: the pre-onset part and the onset part. The pre-onset part is the period before the onset of the disease (in order to accurately capture the signs before the onset of the disease and avoid the data being too large or too small), and the onset part is from the beginning of the disease to the end of the disease.

[0025] A generation module is used to collect previous disease status information of disabled patients, construct a disease status information timeline through previous disease status information, and intercept information on the disease status information timeline by intercepting the time window to obtain disease status sample information, and extract disease status features based on the disease status sample information to construct a standard disease status feature set.

[0026] In some embodiments of the present application, the generation module is used to construct a timeline of the onset state information through the previous onset state information, and intercept the onset state information timeline by means of the interception time window to obtain the onset state sample information, which is specifically: Arrange the previous disease status information in chronological order and display it in the form of a timeline to construct a disease status information timeline. By intercepting the time nodes corresponding to the pre-onset part and the onset part under the time window, the disease status information timeline is intercepted to obtain the disease status sample information.

[0027] In this embodiment, the disease state information here also includes brain information and physiological information, and the collected previous disease state information is arranged in chronological order to construct a disease state information timeline. The timeline can intuitively display the patient's disease development process and provide a basis for subsequent information interception and feature extraction.

[0028] In some embodiments of the present application, the generation module is further used to extract the disease state features based on the disease state sample information to construct a standard disease state feature set, which is specifically: Determine the state characteristics of disabled patients in a non-ill state through brain disease portraits, and use the state characteristics of disabled patients in a non-ill state as the baseline state characteristics; All abnormal state features are extracted from the disease state sample information, and the baseline state features are compared with the abnormal state features to screen out the disease state features and construct a standard disease state feature set. The standard disease state feature set includes a brain disease feature set and a physiological disease feature set.

[0029] In this embodiment, because brain patients may have some abnormal characteristics due to the influence of long-term illness, such as certain physiological indicators are too high. However, these abnormal characteristics are not abnormal characteristics before or during the onset of the disease, so they need to be excluded to improve the accuracy of subsequent feature matching. Collect brain images and other relevant physiological data of disabled patients in the non-illness state, such as electroencephalogram (EEG), magnetic resonance imaging (MRI), etc., as well as possible physiological indicators (such as heart rate, blood pressure, etc.). The state characteristics of disabled patients in the non-illness state (the state characteristics here are abnormal state characteristics, not normal), the baseline state characteristics are compared with the abnormal state characteristics, and the onset state characteristics are screened out.

[0030] The calculation module is used to monitor the real-time status information of disabled patients, extract status features from the real-time status information to construct a real-time status feature set, calculate the relative matching degree between the real-time status feature set and the standard disease status feature set, and determine the warning information according to the relative matching degree.

[0031] In this embodiment, a period can be set according to the actual situation (frequency of onset, etc.), and the real-time state information of disabled patients within a period can be used to construct a real-time state feature set, so as to facilitate subsequent calculations. The relative state matching degree describes the similarity between the real-time state feature set and the standard onset state feature set, so as to determine whether there is danger.

[0032] In some embodiments of the present application, Figure 2 As shown, the computing module includes a monitoring unit, a matching unit and a transfer unit. The monitoring unit is used to extract state features from the real-time state information to construct a real-time state feature set, which is specifically: The real-time state information is real-time brain information and physiological information, and real-time brain features and real-time physiological features are respectively extracted from the real-time brain information and physiological information, and a real-time brain feature set and a real-time physiological feature set are respectively constructed through the real-time brain features and the real-time physiological features. The real-time state feature set includes the real-time brain feature set and the real-time physiological feature set.

[0033] In some embodiments of the present application, the matching unit is used to calculate the relative matching degree between the real-time state feature set and the standard disease state feature set, which is specifically: Calculate the characteristic variation coefficient of the real-time brain characteristic set and the real-time physiological characteristic set, calculate the matching degree between each corresponding characteristic according to the real-time brain characteristic set and the real-time physiological characteristic set as well as the brain disease characteristic set and the physiological disease characteristic set, and calculate the relative matching degree of the state by combining the characteristic variation coefficient of the real-time brain characteristic set and the real-time physiological characteristic set and the matching degree between the corresponding characteristics; ; in, is the relative matching degree of the state, , are the matching conversion coefficients corresponding to the characteristics of brain information and physiological information, , are the number of feature types of the brain disease feature set and the physiological disease feature set, respectively. The first The combined weights of brain features, The first The degree of matching of brain features, , are the overall coefficient of variation of the real-time brain feature set and the real-time physiological feature set, respectively. , are preset constants, The first The combined weight of the physiological features is the matching feature of the th physiological feature in the physiological disease feature set.

[0034] In this embodiment, the coefficient of variation of the feature value of the real-time state feature set is calculated. The coefficient of variation (CV) is a statistic that measures the degree of data dispersion and is defined as the ratio of the standard deviation to the mean value. It is used to evaluate the fluctuation degree and stability of the feature. The fluctuation will affect the matching result. The matching degree between each corresponding feature is the matching degree of the same feature between the real-time brain feature set and the brain disease feature set, and the matching degree of the same feature between the real-time physiological feature set and the physiological disease feature set. The matching degree is calculated by methods such as Euclidean distance, cosine similarity, Manhattan distance, etc. It represents the correction of the mean value of the coefficient of variation to the sum of the matching degrees of brain features. Similarly, I will not elaborate on it.

[0035] In some embodiments of the present application, the transfer unit is used to determine the warning information according to the relative matching degree of the states, which is specifically: If the state relative matching degree is greater than the state relative matching degree threshold, the warning information is determined according to the state relative matching degree and the specific matching features; Otherwise, the state transition probability from the real-time state feature set to the standard disease state feature set is calculated, and the warning information is determined according to the state transition probability.

[0036] In this embodiment, if the relative state matching degree is greater than the relative state matching degree threshold, it means that the current patient's state is similar to the diseased state, and a warning is issued by combining the relative state matching degree (the larger the value, the higher the warning level) and the specific matching features (those with high matching degrees). If the relative state matching degree is not greater than the relative state matching degree threshold, it means that the current patient's state is not similar to the diseased state, and the transition probability from the current patient's state (real-time state feature set) to the diseased state (standard diseased state feature set) is calculated.

[0037] In some embodiments of the present application, the transfer unit is specifically used to calculate the state transition probability from the real-time state feature set to the standard disease state feature set, and determine the warning information according to the state transition probability, which is specifically: The initial state transition probability from the real-time state feature set to the standard disease state feature set is calculated through the Markov chain, and the target state transition probability is determined according to the initial state transition probability and the relative state matching degree; ; in, is the target state transition probability, is the initial state transition probability, is the state transition probability correction, It represents the state transition probability correction amount obtained by mapping the relative matching degree of the state; The early warning information is determined by the target state transition probability.

[0038] In this embodiment, the difference between the real-time state feature set and the standard disease state feature set reflects the degree of deviation between the patient's current state and the typical disease state (state relative matching degree). This difference may be caused by a variety of factors, such as disease progression, individual differences, environmental factors, etc. The calculated transition probability can be corrected by the state relative matching degree. For example, if the difference between the real-time state and the disease state is large, then the predicted value of the transition probability may need to be reduced; conversely, if the difference is small, then the predicted value of the transition probability may need to be increased.

[0039] In this embodiment, the initial state transition probability from the real-time state feature set to the standard disease state feature set is calculated by using a Markov chain, which is a method of using a probability model to predict and analyze system state changes. Markov chain is a random process, which is characterized by the fact that the next state of the system is only related to the current state, but not to the previous state, that is, there is no aftereffect. In this application, we can regard the real-time state feature set as the current state of the Markov chain, and the standard disease state feature set represents the target state that the system may reach. The initial state transition probability describes the probability that the system transfers from the real-time state feature set to the standard disease state feature set. This probability describes the possibility of the system transferring from the real-time state feature set to the standard disease state feature set at the initial moment.

[0040] The early warning module is used to send early warning information to multiple monitoring parties, so as to facilitate the monitoring parties to take further preventive measures.

[0041] In this embodiment, the warning information may include the warning level, the warning reason, the recommended measures, etc. The warning information is sent to multiple monitoring parties, such as the patient's family, medical staff, telemedicine center, etc. The warning information can be sent in the form of text messages, emails, APP push, etc., to ensure that the information can be conveyed to relevant personnel in a timely and accurate manner. The monitoring party takes further preventive measures based on the warning information, such as adjusting the treatment plan, strengthening monitoring, preparing emergency equipment, etc., to prevent or alleviate the possible deterioration of the patient's condition.

[0042] Compared with the prior art, the present invention has the following beneficial effects: 1. Establish a brain disease portrait based on the brain disease information and body function information of disabled patients, and establish a time window through the brain disease portrait. The disease portrait describes the impact of the patient's disease changes and determines a time window. Through this window, some signs of some patients before the onset of the disease can be accurately captured, thereby helping to warn of the onset of the disease.

[0043] 2. Calculate the relative matching degree between the real-time state feature set and the standard disease state feature set, analyze the patient's current state through the relative matching degree, and conduct comprehensive early warning for the patient based on the relative matching degree and state transition probability, thereby improving the tracking effect and early warning success rate of disabled patients and ensuring the relative safety of disabled patients.

[0044] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0045] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present invention.

[0046] Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more systems different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0047] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A tracking and analysis system for disabled patients based on smart chips, characterized in that: include, A portrait module is used to collect brain disease information and body function information of disabled patients, establish a brain disease portrait based on the brain disease information and body function information of disabled patients, and establish a capture time window through the brain disease portrait; A generation module is used to collect the previous disease state information of disabled patients, construct the disease state information timeline through the previous disease state information, and intercept the disease state information timeline by intercepting the time window to obtain the disease state sample information, and extract the disease state features based on the disease state sample information to construct a standard disease state feature set; A calculation module is used to monitor the real-time status information of disabled patients, extract status features from the real-time status information to construct a real-time status feature set, calculate the relative matching degree between the real-time status feature set and the standard disease status feature set, and determine the warning information according to the relative matching degree; The early warning module is used to send early warning information to multiple monitoring parties, so as to facilitate the monitoring parties to take further preventive measures; Among them, the status information of disabled patients includes brain information and physiological information.

2. The intelligent chip-based disabled patient tracking and analysis system according to claim 1 is characterized in that: The portrait module is used to establish a brain disease portrait based on the brain disease information and body function information of the disabled patient, which is specifically: The brain disease information and body function information are preprocessed, and the brain disease characteristics and body function characteristics are extracted. The brain disease model and body model are constructed respectively by fusion of the brain disease characteristics and body function characteristics. The impact of lesion evolution on the body model is simulated according to the brain disease model and the body model. The abnormal signs of brain information and physiological information generated in this impact process are analyzed to determine the advance time under different lesion stages, so as to construct a matching relationship between lesion stage and advance time. The current lesion stage, the matching relationship between lesion stage and advance time, and the simulated impact process of lesion evolution on the body model and the abnormal signs are used as the content of the brain disease portrait.

3. The intelligent chip-based disabled patient tracking and analysis system according to claim 2 is characterized in that: The image module is used to establish a capture time window through the brain disease image, which is specifically: The intercepted time window includes two parts: the pre-onset part and the onset part. The advance time node of the pre-onset part is determined by matching the current lesion stage in the brain disease portrait with the lesion stage-advance time. The time of the pre-onset part is from the advance time node to the onset start time node, and the time of the onset part is from the onset start time node to the onset end time node.

4. The intelligent chip-based disabled patient tracking and analysis system according to claim 3 is characterized in that: The generation module is used to construct a disease state information time axis through the previous disease state information, and intercept the disease state information time axis by intercepting the time window to obtain disease state sample information, which is specifically: Arrange the previous disease status information in chronological order and display it in the form of a timeline to construct a disease status information timeline. By intercepting the time nodes corresponding to the pre-onset part and the onset part under the time window, the disease status information timeline is intercepted to obtain the disease status sample information.

5. The intelligent chip-based disabled patient tracking and analysis system according to claim 2 is characterized in that: The generation module is also used to extract the disease state features based on the disease state sample information to construct a standard disease state feature set, which is specifically: Determine the state characteristics of disabled patients in a non-ill state through brain disease portraits, and use the state characteristics of disabled patients in a non-ill state as the baseline state characteristics; All abnormal state features are extracted from the disease state sample information, and the baseline state features are compared with the abnormal state features to screen out the disease state features and construct a standard disease state feature set. The standard disease state feature set includes a brain disease feature set and a physiological disease feature set.

6. The intelligent chip-based disabled patient tracking and analysis system according to claim 5, characterized in that: The computing module includes a monitoring unit, a matching unit and a transfer unit. The monitoring unit is used to extract state features from the real-time state information to construct a real-time state feature set, which is specifically: The real-time state information is real-time brain information and physiological information, and real-time brain features and real-time physiological features are respectively extracted from the real-time brain information and physiological information, and a real-time brain feature set and a real-time physiological feature set are respectively constructed through the real-time brain features and the real-time physiological features. The real-time state feature set includes the real-time brain feature set and the real-time physiological feature set.

7. The intelligent chip-based disabled patient tracking and analysis system according to claim 6, characterized in that: The matching unit is used to calculate the relative matching degree between the real-time state feature set and the standard disease state feature set, which is specifically: Calculate the characteristic variation coefficient of the real-time brain characteristic set and the real-time physiological characteristic set, calculate the matching degree between each corresponding characteristic according to the real-time brain characteristic set and the real-time physiological characteristic set as well as the brain disease characteristic set and the physiological disease characteristic set, and calculate the relative matching degree of the state by combining the characteristic variation coefficient of the real-time brain characteristic set and the real-time physiological characteristic set and the matching degree between the corresponding characteristics; ;in, is the relative matching degree of the state, , are the matching conversion coefficients corresponding to the characteristics of brain information and physiological information, , are the number of feature types of the brain disease feature set and the physiological disease feature set, respectively. The first The combined weights of brain features, The first The degree of matching of brain features, , are the overall coefficient of variation of the real-time brain feature set and the real-time physiological feature set, respectively. , are preset constants, The first The combined weight of the physiological characteristics, is the matching feature of the th physiological feature in the physiological pathogenesis feature set.

8. The intelligent chip-based disabled patient tracking and analysis system according to claim 7, characterized in that: The transfer unit is used to determine the warning information according to the relative matching degree of the states, which is specifically: If the state relative matching degree is greater than the state relative matching degree threshold, the warning information is determined according to the state relative matching degree and the specific matching features; Otherwise, the state transition probability from the real-time state feature set to the standard disease state feature set is calculated, and the warning information is determined according to the state transition probability.

9. The intelligent chip-based disabled patient tracking and analysis system according to claim 8, characterized in that: The transfer unit is specifically used to calculate the state transition probability from the real-time state feature set to the standard disease state feature set, and determine the warning information according to the state transition probability, which is specifically: The initial state transition probability from the real-time state feature set to the standard disease state feature set is calculated through the Markov chain, and the target state transition probability is determined according to the initial state transition probability and the relative state matching degree; ;in, is the target state transition probability, is the initial state transition probability, is the state transition probability correction, It represents the state transition probability correction amount obtained by mapping the relative matching degree of the state; The warning information is determined by the target state transition probability.

Images