Blood routine noninvasive monitoring patch, system and detection method

Abstract

The invention provides a blood routine noninvasive monitoring patch and system and a detection method, and relates to the technical field of medical detection. According to the blood routine monitoring patch, blood routine monitoring can be carried out by utilizing the patch, a noninvasive detection technical means is adopted, infrared spectrum is utilized to permeate and collect infrared spectrum signals of subcutaneous blood, and after analysis, blood monitoring data is obtained and wirelessly reported to a background server; and blood abnormal signal supervision can be further carried out on the background based on an AI algorithm. Therefore, by the adoption of the blood routine monitoring device, blood routine changes can be monitored in real time by wearing the patch on a patient, invasive operation is avoided, convenience is provided for the patient, and dependence on medical staff is reduced. The patient can carry out blood routine (blood and blood vessel forms) monitoring at home, background feedback can be obtained when abnormity occurs, and intelligent blood routine supervision and detection are achieved.

Description

Technical Field

[0001] The present invention relates to the field of medical detection technology, and in particular to a non-invasive blood routine monitoring patch, system and detection method. Background Art

[0002] Patients with blood diseases need to have their blood routine tested frequently during hospitalization and at home. Currently, patients need to go to the hospital for medical staff to use a blood collection needle or fingertip blood for blood routine testing. Therefore, almost all blood tests are invasive. Invasive blood collection operations have the following technical defects for patients and medical staff:

[0003] During a routine blood test, blood samples are sometimes collected frequently, which causes multiple invasive blood collection operations and pain for patients. For nurses, they need to frequently visit patients to collect blood, which takes up a lot of time and resources.

[0004] Traditional routine blood tests require the hospital to perform blood analysis before the patient can get a report, so real-time monitoring is not possible, and it is not convenient for patients to monitor their blood test values ​​at home after being discharged from the hospital;

[0005] In the traditional blood test, patients need to go to the hospital for a blood test every time, which not only takes time to travel, but also cannot quickly obtain information about blood changes, so it is not convenient and intelligent enough;

[0006] In addition, traditional blood tests mainly check abnormal blood data, but cannot perform in-depth vascular analysis, and cannot simultaneously monitor and provide feedback on abnormalities in the patient's blood and subcutaneous blood vessels. Summary of the invention

[0007] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0008] On the one hand, a non-invasive monitoring patch for routine blood test is provided, the non-invasive monitoring patch for routine blood test comprising a skin-friendly layer and a non-invasive monitoring module for routine blood test attached to the skin-friendly layer, wherein the non-invasive monitoring module for routine blood test comprises:

[0009] A controller, comprising:

[0010] An infrared matrix, composed of a plurality of infrared diodes arranged in a matrix, is used to collect infrared spectrum signals of blood and feed them back to the controller. After spectrum analysis by the controller, corresponding blood monitoring data is obtained and forwarded to the wireless communication module;

[0011] A wireless communication module is used to upload the blood monitoring data to a backend server, which records and writes the data into the patient's electronic medical record in the HIS system and sends it to the patient's APP simultaneously;

[0012] Power supply, used for power supply;

[0013] The infrared matrix, wireless communication module and power supply are electrically connected to the controller respectively.

[0014] Optionally, the controller further includes:

[0015] MCU, used for logic calculation and control;

[0016] An A / D converter is used for performing AD sampling and converting the collected blood infrared spectrum signal into a digital-to-analog signal to obtain a corresponding blood infrared spectrum electrical signal;

[0017] A signal separator, used for performing signal separation on the blood infrared spectrum electrical signal to obtain the blood spectrum signal therein, and obtaining the corresponding blood monitoring data after being processed by the MCU;

[0018] A blood signal module, used to record the blood monitoring data and save it in a memory, and forward the blood monitoring data to the wireless communication module through the MCU;

[0019] A timer, used for sending sampling signals at regular intervals;

[0020] A memory for sampling and storage;

[0021] An electronic screen for displaying the blood monitoring data;

[0022] A power management module, used to manage the power supply of the power supply;

[0023] The A / D converter, signal separator, blood signal module, timer, memory, electronic screen and power management module are electrically connected to the MCU respectively.

[0024] Optionally, the signal separator is further used for:

[0025] Separating the infrared spectrum signal of the subcutaneous blood vessels from the infrared spectrum electrical signal of the blood, including: position, distribution and morphology;

[0026] The MCU is used to process the infrared spectrum signal and generate corresponding blood vessel monitoring data, and forward it to the wireless communication module, which is then uploaded to the backend server by the wireless communication module, and recorded and written into the patient's HIS electronic medical record system by the backend server.

[0027] Optionally, the controller further includes:

[0028] The ISD voice chip is used to process the voice alarm signal sent by the background server and generate corresponding voice alarm data, wherein the voice alarm signal is generated in the following manner: the background server monitors in real time and determines whether the patient's blood monitoring data is abnormal:

[0029] If yes, the voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and received by the wireless communication module of the blood routine non-invasive monitoring patch and forwarded to the ISD voice chip;

[0030] Otherwise give up.

[0031] Optionally, the non-invasive blood routine monitoring patch further comprises:

[0032] A voice module, used for receiving and responding to the voice alarm data;

[0033] The voice module is electrically connected to the ISD voice chip.

[0034] On the other hand, a non-invasive blood routine monitoring system is provided, characterized in that it includes:

[0035] The non-invasive blood routine monitoring patch described above;

[0036] The backend server is used to manage the non-invasive blood routine monitoring patch worn by the patient; and collect and record the patient's blood monitoring data in real time, and determine whether the patient's blood monitoring data is abnormal. If abnormal, a voice alarm signal corresponding to the patient is generated and sent to the non-invasive blood routine monitoring patch worn by the patient;

[0037] Patient APP, used to log in to the backend server, receive and view the blood monitoring data;

[0038] The non-invasive blood routine monitoring patch and the patient APP are respectively connected to the backend server for communication.

[0039] Optionally, the background server is deployed with:

[0040] A processing system for providing data processing services based on multiple CPUs;

[0041] A data acquisition system, used to communicate with the blood routine non-invasive monitoring patch worn by each patient to collect the patient's blood monitoring data;

[0042] An HIS system, used for generating an electronic medical record of a patient and recording the blood monitoring data and / or vascular monitoring data of the patient;

[0043] A blood analysis system, used to write the patient's blood monitoring data and generate a corresponding monitoring curve;

[0044] MySQL database, used for background data storage;

[0045] API interface, used to provide background communication port;

[0046] A blood abnormality monitoring system is used to perform abnormal monitoring on the patient's blood monitoring data through a pre-deployed blood abnormality AI model to determine whether the patient's blood monitoring data has blood abnormality characteristics and / or abnormal vascular morphology characteristics identified by the model. If abnormal, a corresponding monitoring signal is output and the processing system generates the voice alarm signal corresponding to the patient; otherwise, monitoring continues;

[0047] The data acquisition system, HIS system, blood analysis system, MySQL database, API interface and blood abnormality monitoring system are respectively connected to the processing system for communication.

[0048] Optionally, the method for generating the blood abnormality AI model includes:

[0049] Collect and pre-process the blood routine test data of several patients with abnormal blood routine test data;

[0050] A supervised learning algorithm is used to identify and extract the abnormal blood feature range values ​​in the abnormal blood data, and bind them with the corresponding examination items to generate corresponding blood abnormality feature data; CNN is used to extract the abnormal vascular morphological features in the abnormal subcutaneous vascular atlas and mark the parameter feature range values ​​of the corresponding abnormal blood vessels;

[0051] The abnormal blood characteristic data and the abnormal blood vessel morphological characteristics are respectively labeled, and the labeling information includes a supervisory signal corresponding to the abnormal blood characteristic data;

[0052] The annotated blood abnormality feature data and blood vessel abnormality morphology feature are statistically constructed into a feature set S: S={blood abnormality feature data, blood vessel abnormality morphology feature};

[0053] Dividing the feature set S into a training set and a validation set according to a preset ratio;

[0054] Inputting the training set into a preset random forest model to perform feature training and learning to generate the blood abnormality AI model;

[0055] The recognition performance of the blood abnormality AI model was verified using the verification set:

[0056] If the verification is successful, the blood abnormality AI model is deployed on the backend server;

[0057] If the verification fails, repeat the above steps to retrain the blood abnormality AI model.

[0058] On the other hand, a blood routine test method is provided, which is implemented based on the above-mentioned blood routine non-invasive monitoring system, and the method comprises the following steps:

[0059] The patient wears and activates the blood routine non-invasive monitoring patch to establish a communication connection with the backend server;

[0060] The blood monitoring data of the patient is collected in real time through the non-invasive blood routine monitoring patch worn by the patient and uploaded to the backend server;

[0061] The backend server records the patient's blood monitoring data in real time, and determines whether the patient's blood monitoring data is abnormal. If abnormal, a voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and the blood routine non-invasive monitoring patch responds to the voice alarm data to remind the patient of abnormal blood monitoring; wherein, determining whether the patient's blood monitoring data is abnormal includes determining whether abnormal blood features and / or abnormal vascular morphological features identified by the model appear in the patient's blood monitoring data;

[0062] The patient logs in to the backend server through the patient APP to receive and view the blood monitoring data.

[0063] On the other hand, an electronic device is provided, comprising: a processor; a memory, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, the above method is implemented.

[0064] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the above method.

[0065] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0066] The present invention provides a non-invasive monitoring patch for routine blood tests, which can be used for blood monitoring, using non-invasive detection technology, using infrared spectrum to penetrate and collect infrared spectrum signals of subcutaneous blood, and after analysis, blood monitoring data is obtained and reported to the background server wirelessly, and blood abnormality signal supervision can be further performed based on AI algorithm in the background. Therefore, by adopting the present invention, it is possible to monitor the changes in routine blood tests in real time by having the patient wear a patch, reduce invasive operations, provide convenience for patients, and reduce dependence on medical staff. And the present invention enables patients to monitor abnormal blood and subcutaneous vascular morphology at home, and can obtain background feedback when abnormalities occur, realizing intelligent blood supervision and detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0068] Figure 1 This is a schematic diagram of the application structure of a non-invasive blood routine monitoring patch provided by an embodiment of the present invention;

[0069] Figure 2 It is a schematic diagram of the structure of a surface mount integrated circuit provided by an embodiment of the present invention;

[0070] Figure 3 is a block diagram of a non-invasive blood routine monitoring system provided by an embodiment of the present invention;

[0071] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0073] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0074] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0075] In the embodiments of the present invention, sometimes a subscript such as W1 may be expressed in a non-subscript form such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.

[0076] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0077] The embodiment of the present invention provides a non-invasive blood routine monitoring patch. Figure 1 and Figure 2 The application structure diagram of the non-invasive monitoring patch for blood routine is shown, wherein the non-invasive monitoring patch for blood routine comprises a skin-friendly layer 200 and a non-invasive monitoring module for blood routine 100 attached to the skin-friendly layer 200, wherein the non-invasive monitoring module for blood routine 100 comprises:

[0078] Controller, including:

[0079] An infrared matrix, composed of a plurality of infrared diodes arranged in a matrix, is used to collect infrared spectrum signals of blood and feed them back to the controller. After spectrum analysis by the controller, corresponding blood monitoring data is obtained and forwarded to the wireless communication module;

[0080] A wireless communication module is used to upload the blood monitoring data to a backend server, which records and writes the data into the patient's electronic medical record in the HIS system and sends it to the patient's APP simultaneously;

[0081] Power supply, used for power supply;

[0082] The infrared matrix, wireless communication module and power supply are electrically connected to the controller respectively.

[0083] Near infrared spectroscopy technology is usually combined with specific equipment or instruments to monitor subcutaneous blood. Reflection or transmission spectral acquisition methods are used to obtain spectral information of subcutaneous blood. Then, the spectral data is processed and analyzed through professional algorithms and models to extract useful blood data, such as hemoglobin concentration, cell count, etc.

[0084] Because infrared spectroscopy is non-invasive, highly accurate, easy to operate, safe and reliable, near-infrared light can penetrate the skin to a certain depth, so it can capture information such as the location, distribution and morphology of subcutaneous microvessels. In addition, near-infrared spectroscopy can also be used to quantitatively detect and analyze specific components in the blood, such as hemoglobin concentration. Therefore, the present invention uses near-infrared spectroscopy technology to observe and analyze relevant data of subcutaneous blood through the skin surface.

[0085] The present invention designs a patch structure, and the skin-friendly layer 200 can be an adhesive layer of a skin-friendly material with a thickness of 3-15 mm, and the material is, for example:

[0086] The PVA and cyanoacrylate adhesive layer is a translucent film with small holes in the middle. It has a certain air permeability and is suitable for bandaging small wounds.

[0087] Gauze strips have good moisture absorption and certain air permeability, and are suitable for situations where the wound area is irritated by foreign bodies or is severely contaminated;

[0088] Sterile dressings, such as vaseline gauze, transparent dressings, and foam dressings, have good air permeability and moisture permeability, which can effectively prevent external bacteria and other pathogens from entering the wound and promote wound healing;

[0089] Medical adhesive plaster (zinc oxide tape): Made of cotton cloth, with medical pressure-sensitive adhesive on the surface, it has good tensile strength, waterproof and anti-pollution properties, and breathable and skin-friendly properties. It is suitable for surgical fixed dressings and lightweight catheters, etc.

[0090] PE perforated breathable tape: The base material is polyethylene (PE) film, the surface layer is made of medical pressure-sensitive adhesive, the laser micropores are evenly distributed, the air permeability is strong, water vapor and sweat are easily discharged, and the adhesion is comfortable.

[0091] Microporous non-woven breathable tape: Made of high-quality non-woven material, the microporous structure makes it breathable, allowing the skin to breathe naturally, and is suitable for wounds of various shapes and locations.

[0092] Silk type breathable tape: It is made of artificial silk, medical hot melt adhesive and a core. The material is light, thin and soft, has good compliance, fits the skin more comfortably, and air can pass through it naturally, allowing the skin to breathe naturally.

[0093] A blood routine non-invasive monitoring module 100 is fixedly mounted thereon. The blood routine non-invasive monitoring module 100 adopts a sealed structure, an internal integrated circuit board, and an integrated structure on the circuit board such as Figure 2 shown.

[0094] The module is mainly a PCP board, which integrates a controller, an infrared matrix constructed by a matrix of several infrared diodes, a power module, a wireless communication module, and an electronic screen. The specific integrated installation and fixing structure of each electronic device on the PCP board can refer to the application structure of the circuit board in various existing electronic devices, and this embodiment is not limited. For each electronic device used, it can be configured according to demand, or it can be purchased from the market. For example, the specifications and models of infrared diodes can be selected according to demand.

[0095] In the present invention, the controller adopts an MCU chip, such as an STM32-bit or 64-bit single-chip MCU chip, etc. The power supply can be powered by a lithium battery or a charging interface, and the voice module and the corresponding voice chip can refer to the voice module configured on the existing electronic device. The wireless communication module can adopt Bluetooth or 5g communication module, etc.; the electronic screen can adopt an LED electronic screen, etc., which can be controlled by the MCU and display the corresponding monitoring data.

[0096] After the patch is activated, its wireless communication module can communicate with the background server. For example, if 5g communication is selected, the wireless communication module can send a 5g communication request when the patch is activated, and the background server will respond after receiving it to establish the corresponding data communication link channel. The specific wireless communication protocol and communication mode used can be understood in combination with the specific communication technology used.

[0097] In the present invention, the data communication between the patch worn by each patient and the background server can be managed according to the wireless communication mode configured by the hospital. The background server can perform background management on the patch worn by each patient based on the wireless communication mode. When a patch is activated, the background can establish a data interaction link with the patch to check the working status of each patch; at the same time, the corresponding sampling task or signal can be issued to perform data sampling through the patch to check the blood monitoring data of each patient. In the background, the identity ID or medical ID of each patient can also be bound to the patch device number worn by the patient, so that the background can use the patch to monitor the patient's blood data, and the data can be bound and stored in the background.

[0098] The infrared diode can emit corresponding infrared light. When the infrared matrix is ​​working, the diode of each diode needs to pass through the skin-friendly layer and aim at the skin. Therefore, several through holes should be set in advance on the skin-friendly layer to facilitate the infrared diode to emit infrared light and receive reflected signals. In the controller, the collected blood infrared spectrum signal (pre-processed first, such as signal filtering, op amp processing, etc.) can be spectrally analyzed, and the signal ad conversion, filtering, noise reduction, op amp processing, etc. can be performed according to the preset program. The specific signal processing circuit integrated in the controller can refer to the application circuit principle of the existing optoelectronic signal processor, and no limitation or redundant description of signal processing is made here.

[0099] After the controller analyzes and processes the collected infrared spectrum signals, the corresponding blood monitoring data can be obtained. The required spectrum analysis and processing program can be configured as an embedded program or the data can be reported to the background server after signal conversion processing, and the background server will analyze and classify the data to generate the corresponding blood monitoring data, such as the number of red blood cells, hemoglobin concentration, the number of white blood cells or their proportions and other monitoring data.

[0100] The backend can record and save the patient's blood routine monitoring and analysis data. The patient can log in to the backend and view the data through the patient APP and other application software. The administrator (attending doctor) can log in to the backend through a PC and view the patient's blood routine data.

[0101] like Figure 2 As shown, the controller further includes:

[0102] MCU, used for logic calculation and control;

[0103] An A / D converter is used for performing AD sampling and converting the collected blood infrared spectrum signal into a digital-to-analog signal to obtain a corresponding blood infrared spectrum electrical signal;

[0104] A signal separator, used for performing signal separation on the blood infrared spectrum electrical signal to obtain the blood spectrum signal therein, and obtaining the corresponding blood monitoring data after being processed by the MCU;

[0105] A blood signal module, used to record the blood monitoring data and save it in a memory, and forward the blood monitoring data to the wireless communication module through the MCU;

[0106] A timer, used for sending sampling signals at regular intervals;

[0107] A memory for sampling and storage;

[0108] An electronic screen for displaying the blood monitoring data;

[0109] A power management module, used to manage the power supply of the power supply;

[0110] The A / D converter, signal separator, blood signal module, timer, memory, electronic screen and power management module are electrically connected to the MCU respectively.

[0111] A / D converter (analog-to-digital converter, ADC for short) is an electronic component used to convert analog signals into digital signals. A / D converter is used to convert the collected blood infrared spectrum signal into digital-to-analog signal to obtain the corresponding blood infrared spectrum electrical signal. Using an infrared spectrometer or diode, etc., infrared light is emitted and infrared light reflected or transmitted by blood is received to obtain the infrared spectrum information of blood. The A / D converter is responsible for converting the collected continuous analog infrared spectrum signal into a discrete digital signal. This is because digital signals are easier to process, store and transmit.

[0112] The general process of A / D conversion:

[0113] Sampling: Periodically measuring the instantaneous value of an analog signal.

[0114] Hold: Between samples, the hold circuit stores the sampled values ​​for a period of time to ensure that the quantization encoding circuit can process these values ​​stably.

[0115] Quantization: Convert the amplitude value of the analog signal into a discrete digital value. This process will introduce a certain amount of quantization error.

[0116] Encoding: Represent the quantized result with a digital code (usually a binary number).

[0117] After the infrared spectrum signals are collected, the A / D converter converts these analog signals into digital signals according to the above process. These digital signals can then be further analyzed and processed by a computer or other digital processing system to extract information about the blood components or status.

[0118] The collected blood infrared spectrum electrical signal contains several types of digital signals, such as infrared spectrum signals of blood vessels and blood. Therefore, it is necessary to perform signal separation to extract the blood spectrum signal and obtain the corresponding blood monitoring data after processing by the MCU.

[0119] By analyzing the abnormal signal characteristics of blood vessels and blood shown in the blood infrared spectrum electrical signal, the signal separator is used to identify and separate the signal. The signal separator (signal classification and extraction) plays an important role in the processing of blood infrared spectrum electrical signals. It can be used as an electronic device to separate mixed signals into different components. Its main function is to decompose complex mixed signals into multiple independent signals for independent processing or analysis. In the scenario of blood infrared spectrum electrical signal processing, the signal separator is used to extract the blood spectrum signal from the mixed signal.

[0120] When the infrared spectrum electrical signals of blood are collected and converted into digital signals, these signals may contain multiple components, such as blood spectrum signals, noise signals, interference signals, etc. The signal separator decomposes these mixed signals into independent signal components through some method (such as frequency diversity, phase detection, etc.).

[0121] During this process, the signal separator will identify and extract the blood spectral signal, which is a key step in subsequent analysis and processing.

[0122] The extracted blood spectrum signal is then sent to the MCU (microcontroller) for processing. The MCU is an integrated circuit chip that can perform various computing and control tasks. In this scenario, the MCU will further analyze and process the blood spectrum signal to extract key information about the blood, such as blood composition, blood oxygen saturation, blood sugar level, etc. The MCU can also convert the processed data into a format that is easy to understand and analyze, such as numbers, charts, or images. After being processed by the MCU, the blood monitoring data obtained can be used in a variety of scenarios such as clinical diagnosis, health monitoring, and disease prediction. These data provide doctors with detailed information about the patient's blood condition, which helps to formulate more accurate treatment plans.

[0123] When using the signal separator and MCU to process the blood infrared spectrum electrical signal, it is necessary to consider a variety of technical requirements and factors, such as the resolution, dynamic range, linearity and other performance indicators of the signal separator, as well as the processing speed, memory capacity, interface type, etc. of the MCU, which can be set by the administrator. It is also necessary to pay attention to the noise and interference of the signal (which can be pre-processed by the built-in filtering circuit of the MCU), as well as the accuracy and reliability of signal processing and data analysis.

[0124] Therefore, based on the signal characteristics of blood and blood vessels, signal separation can be performed to obtain and obtain blood spectrum signals, and the corresponding blood monitoring data can be obtained after processing by the MCU. The data of various blood test items, such as the number of red blood cells, the number of white blood cells or their proportion, can be configured and applied in an embedded program (such as configuring some algorithm models) in combination with the statistical method of blood routine analysis. For example:

[0125] The collected spectral signal may contain noise, interference or other irrelevant information. Therefore, before further analysis, the signal needs to be preprocessed, such as filtering, denoising, baseline correction, etc., to ensure the accuracy and reliability of the signal.

[0126] Extract features related to blood routine indicators from the preprocessed spectral signals. This may involve analyzing the light absorption, scattering or reflection characteristics at specific wavelengths, and the relationship between these characteristics and different components in the blood (such as red blood cells, white blood cells, platelets, etc.).

[0127] In order to link spectral features with blood routine indicators, a mathematical model needs to be established. This usually requires using blood samples with known blood routine results as training sets and using machine learning or statistical methods to find the mapping relationship between spectral features and blood routine indicators. At the same time, calibration is also required to ensure the accuracy and reliability of the model.

[0128] The established model is used to predict new blood spectral signals to obtain blood routine indicators (such as white blood cell count, red blood cell count, platelet count, etc.). These indicators can then be analyzed and interpreted to assess the patient's blood health status.

[0129] The blood monitoring data obtained after MCU processing, such as the number and composition of red blood cells, can be recorded by the blood signal module and saved in the memory, and the blood monitoring data can be forwarded to the wireless communication module through the MCU; the memory can be an SD memory. The timer can send out sampling signals at regular intervals for sampling, which can be the internal clock of the MCU. The electronic screen uses LED to display the blood monitoring data; the power management module manages the power supply of the power supply, and the corresponding management circuit can be a power circuit on a conventional circuit board.

[0130] Optionally, the signal separator is further used for:

[0131] Separating the infrared spectrum signal of the subcutaneous blood vessels from the infrared spectrum electrical signal of the blood, including: position, distribution and morphology;

[0132] The MCU is used to process the infrared spectrum signal and generate corresponding blood vessel monitoring data, and forward it to the wireless communication module, which is then uploaded to the backend server by the wireless communication module, and recorded and written into the patient's HIS electronic medical record system by the backend server.

[0133] Under infrared spectrum analysis, the signal separator can also separate the location and morphology of blood vessels from the collected signals, and further analyze the patient's vascular characteristics (such as whether the morphology has changed). The steps for collecting blood vessel signals from subcutaneous infrared signals and obtaining information such as blood vessel location can be referred to as follows:

[0134] Select the appropriate imaging mode according to the patient's vascular condition, which may include basic mode, depth recognition mode, fine mode, etc. The depth recognition mode can detect the depth of the blood vessel and provide a reference for the needle insertion angle for subsequent puncture operations.

[0135] Infrared diodes emit infrared rays to illuminate the skin. Infrared rays can penetrate the skin and be absorbed by blood vessels to form an infrared image of blood vessels. Through specific technical processing, such as filtering, image enhancement (the infrared light signals reflected by blood and blood vessels are different), etc., the distribution of blood vessels under the skin can be clearly displayed. By observing the infrared image, you can intuitively see the direction and distribution of blood vessels under the skin. If needed, the depth, diameter and other information of the blood vessels can be further analyzed to provide accurate positioning for subsequent medical operations (such as puncture). The collected blood vessel information can be recorded and used to guide subsequent medical operations or as part of medical records. At the same time, this information can also be used in scientific research, teaching and other fields to provide data support for related research.

[0136] Here, after the patch reports the monitoring signal to the background server, the background can further analyze the various data in the patient's blood monitoring data to determine whether there are any abnormal items. If there are corresponding abnormal items, a corresponding voice alarm signal will be issued and sent to the patch worn by the patient to make a voice alarm, such as playing a voice message of "Abnormal blood routine, please go to the doctor" to remind the patient to go to the hospital for clinical treatment.

[0137] Optionally, the controller further includes:

[0138] The ISD voice chip is used to process the voice alarm signal sent by the background server and generate corresponding voice alarm data, wherein the voice alarm signal is generated in the following manner: the background server monitors in real time and determines whether the patient's blood monitoring data is abnormal:

[0139] If yes, the voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and received by the wireless communication module of the blood routine non-invasive monitoring patch and forwarded to the ISD voice chip;

[0140] Otherwise, the generation of the voice warning signal is abandoned.

[0141] Optionally, the non-invasive blood routine monitoring patch further comprises:

[0142] A voice module, used for receiving and responding to the voice alarm data;

[0143] The voice module is electrically connected to the ISD voice chip.

[0144] Specifically, after the patch reports the patient's blood monitoring data to the background, the background will analyze the patient's blood monitoring data in the blood analysis system to generate the inspection data under each blood routine examination item, and can generate the patient's corresponding blood routine examination information, such as the number of red blood cells, white blood cells and other components, etc. The specific understanding can be combined with the blood routine analysis algorithm or the program in the analysis system. This blood analysis system can refer to the existing clinical hematology application software.

[0145] As for the various data of the patient, the background can traverse and identify whether there are abnormal items in the various blood routine examination data. If there are corresponding abnormal items, the background will generate a corresponding voice alarm signal for the patient and send it to the patch worn by the patient. The wireless communication module of the patch receives and converts it to the controller. The ISD voice chip (a series of voice chips) in the controller performs voice processing and generates corresponding voice alarm data, which is sent to the voice module for alarm. Please refer to the voice modules on existing terminals such as smart bracelets for this voice module, which can play voice and collect the patient's voice information to realize dialogue. The ISD series chips in the present invention can be configured by the administrator, and they can communicate with the MCU chip. The MCU chip controls the voice chip according to the budget logic to perform voice processing and forwarding, etc.

[0146] Therefore, using the above method, when the corresponding abnormal blood routine data is found in the background, a corresponding alarm can be issued to remind the patient to go to the hospital for the corresponding diagnosis. The background blood routine abnormality examination uses an AI algorithm. For details, please refer to the application of the AI ​​algorithm model below.

[0147] like Figure 3 As shown, on the other hand, a non-invasive monitoring system for routine blood tests is provided, characterized in that it includes:

[0148] The non-invasive blood routine monitoring patch described above;

[0149] The backend server is used to manage the non-invasive blood routine monitoring patch worn by the patient; and collect and record the patient's blood monitoring data in real time, and determine whether the patient's blood monitoring data is abnormal. If abnormal, a voice alarm signal corresponding to the patient is generated and sent to the non-invasive blood routine monitoring patch worn by the patient;

[0150] Patient APP, used to log in to the backend server, receive and view the blood monitoring data;

[0151] The non-invasive blood routine monitoring patch and the patient APP are respectively connected to the backend server for communication.

[0152] Optionally, the background server is deployed with:

[0153] A processing system for providing data processing services based on multiple CPUs;

[0154] A data acquisition system, used to communicate with the blood routine non-invasive monitoring patch worn by each patient to collect the patient's blood monitoring data;

[0155] An HIS system, used for generating an electronic medical record of a patient and recording the blood monitoring data and / or vascular monitoring data of the patient;

[0156] A blood analysis system, used to write the patient's blood monitoring data and generate a corresponding monitoring curve;

[0157] MySQL database, used for background data storage;

[0158] API interface, used to provide background communication port;

[0159] A blood abnormality monitoring system is used to perform abnormal monitoring on the patient's blood monitoring data through a pre-deployed blood abnormality AI model to determine whether the patient's blood monitoring data has blood abnormality characteristics and / or abnormal vascular morphology characteristics identified by the model. If abnormal, a corresponding monitoring signal is output and the processing system generates the voice alarm signal corresponding to the patient; otherwise, monitoring continues;

[0160] The data acquisition system, HIS system, blood analysis system, MySQL database, API interface and blood abnormality monitoring system are respectively connected to the processing system for communication.

[0161] Optionally, the method for generating the blood abnormality AI model includes:

[0162] Collect and pre-process the blood routine test data of several patients with abnormal blood routine test data;

[0163] A supervised learning algorithm is used to identify and extract the abnormal blood feature range values ​​in the abnormal blood data, and bind them with the corresponding examination items to generate corresponding blood abnormality feature data; CNN is used to extract the abnormal vascular morphological features in the abnormal subcutaneous vascular atlas and mark the parameter feature range values ​​of the corresponding abnormal blood vessels;

[0164] The abnormal blood characteristic data and the abnormal blood vessel morphological characteristics are respectively labeled, and the labeling information includes a supervisory signal corresponding to the abnormal blood characteristic data;

[0165] The annotated blood abnormality feature data and blood vessel abnormality morphology feature are statistically constructed into a feature set S: S={blood abnormality feature data, blood vessel abnormality morphology feature};

[0166] Dividing the feature set S: S={blood abnormality feature data, vascular abnormality morphology feature} into a training set and a validation set according to a preset ratio;

[0167] Inputting the training set into a preset random forest model to perform feature training and learning to generate the blood abnormality AI model;

[0168] The recognition performance of the blood abnormality AI model was verified using the verification set:

[0169] If the verification is successful, the blood abnormality AI model is deployed on the backend server;

[0170] If the verification fails, repeat the above steps to retrain the blood abnormality AI model.

[0171] An API interface is deployed on the background server, which allows the background to call the corresponding model, such as a large language model, etc., to provide auxiliary analysis (for example, using a third-party large language model to assist in the data extraction of various blood routine examination items, reading each blood routine test value from the patient's blood routine test data in turn and entering the blood abnormality monitoring system). The HIS system, database, blood analysis system, data acquisition system, etc. deployed on the background, as well as the CPU data processing service provided by the processing system, can refer to the corresponding deployment on the existing hospital background server; and patient users can also log in to the background through the patient APP to realize the corresponding data sharing and viewing, for example, they can view the corresponding daily blood routine monitoring data, etc.

[0172] In the blood analysis system, the detection curve of the corresponding blood routine examination items can also be generated according to the blood monitoring data monitored by the patient each time. For example, for the white blood cell count change curve, the white blood cell count monitored each time can be read, with time as the horizontal axis and the white blood cell count as the vertical axis, so as to generate a monitoring curve of the patient's white blood cell count changing with time. This can be specifically implemented in combination with the existing hematology blood analysis system software.

[0173] On the background server, in order to realize intelligent blood routine abnormality supervision and reminder, the present invention configures a blood abnormality supervision system on the background server. A blood abnormality AI model is deployed in the supervision system. The model is based on the blood routine abnormality big data and is constructed through AI algorithm training. It can identify and judge the abnormalities of various blood routine monitoring data of patients. When it is identified that a certain data in the patient's blood monitoring data contains abnormal blood abnormality characteristic data, a corresponding supervision signal will be issued (for example, when facing an abnormal white blood cell count value, a corresponding inflammation supervision signal will be generated to remind that there is inflammation). The background system generates a corresponding voice alarm signal according to the signal type in the supervision signal (for example, calling the LLM large language model to automatically generate voice alarm data about inflammation), so as to realize intelligent blood abnormality supervision and reminder.

[0174] The deployed AI model for blood abnormality uses the random forest algorithm for model training and construction. Through big data supervised learning and extraction of corresponding blood abnormality feature data, the data is annotated with information, and the supervisory signals corresponding to different blood abnormality feature data are annotated to construct the feature set S: S={blood abnormality feature data, vascular abnormality morphological features}, and then the model is trained and verified according to the training and construction method of the random forest model. Finally, the verified AI model for blood abnormality is deployed on the background server, specifically in the blood abnormality supervision system, for intelligent identification of abnormal blood routine monitoring data of patients. The blood abnormality supervision system can read the blood supervision monitoring data uploaded by the patient's worn patch from the HIS system at regular intervals, read the various blood monitoring values ​​in the data and input them into the AI ​​model, and the AI ​​model identifies and determines whether there is corresponding blood abnormality feature data. If so, the corresponding supervision signal is output for the blood test item, and the background generates a corresponding voice alarm signal for the supervision signal, and issues a reminder to the patient. The supervision signal or supervision data can be supervision suggestions or supervision strategies (such as inflammation and / or inflammation-related treatment recommendation keywords) corresponding to the blood abnormality feature data and other annotation information. The backend can then generate accurate voice alarm signals based on the annotations of the corresponding abnormal features. For example, when the number of white blood cells increases, a corresponding verification supervision signal can be generated, and the backend can generate corresponding infection and inflammation alarm signals to remind patients of possible inflammation. Specifically, for the labeled supervision signals and the alarm signals generated by the backend, the administrator can configure the voice alarm signals corresponding to different supervision signals in the backend database in advance and save the corresponding mapping relationships. The corresponding voice alarm signals can then be pulled based on the supervision signals output by the model.

[0175] The random forest algorithm can be used to generate a model to supervise the abnormal blood routine data, which can be carried out according to the following steps:

[0176] Data preprocessing:

[0177] Data cleaning: Process missing values, outliers, and duplicate data to ensure data quality.

[0178] Data standardization or normalization: According to the distribution and characteristics of the data, the data is standardized or normalized to facilitate algorithm processing.

[0179] Data partitioning: Divide the dataset into training and test sets, usually with a ratio of 70% for training and 30% for testing.

[0180] Feature Selection:

[0181] According to the characteristics of routine blood data, appropriate features are selected as input. These features may include abnormal range values ​​such as white blood cell count, red blood cell count, hemoglobin concentration, platelet count, etc. Feature annotation can also be performed, and corresponding blood treatment measures are annotated for different features (for example, if leukocytosis occurs (the monitoring value is within the abnormal range of white blood cells learned in model training, the abnormal value is detected and output, and the corresponding measures are output), then the corresponding features after annotation are obtained and used for training).

[0182] The feature importance evaluation methods of random forests (such as the Permute Variable method) can be used to select the most relevant features for modeling and prediction.

[0183] Model training:

[0184] Build a random forest model: Set the parameters of the random forest, such as the number of decision trees, the maximum depth of each tree, the maximum number of features, etc. These parameters can be adjusted according to the specific data set and application scenario.

[0185] Training model: Use the training set data to train the random forest model. During the training process, the random forest constructs multiple decision trees and integrates the prediction results of each decision tree by voting or averaging.

[0186] Model Evaluation:

[0187] Performance evaluation: Use the test set data to evaluate the performance of the model. Common evaluation indicators include accuracy, recall, F1 score, etc. The administrator can verify it.

[0188] Outlier detection: The outlier threshold is determined by the anomaly score output by the model. Data points with scores higher than the threshold can be considered outliers.

[0189] Model optimization:

[0190] Based on the results of the model evaluation, adjust the parameters of the random forest or select different features for training to optimize the performance of the model.

[0191] Methods such as cross-validation can be used to further evaluate and optimize the model.

[0192] Model Application:

[0193] The trained random forest model is applied to the actual abnormality detection of blood routine data to predict and analyze new data.

[0194] Therefore, the present invention can realize intelligent judgment and reminder of blood routine based on the above AI model, and realize intelligent home supervision of blood routine.

[0195] The convolutional neural network (CNN) is used to extract the abnormal morphological features of blood vessels in the abnormal subcutaneous blood vessel map and mark the parameter feature range values ​​of the corresponding abnormal blood vessels. The following steps can be referred to:

[0196] 1. Data collection and annotation

[0197] Collect data: Collect image data containing abnormal subcutaneous blood vessels. These data should come from reliable medical image databases or medical institutions.

[0198] In the present invention, infrared spectroscopy technology is also used to obtain subcutaneous vascular images of patients. The following steps can be referred to: the main steps of generating subcutaneous vascular images using infrared spectroscopy signals include: collecting infrared spectroscopy signals, image processing and analysis, and in-situ projection display of vascular images.

[0199] Collect infrared spectrum signals:

[0200] Using an infrared vascular imager, the device emits infrared light to illuminate the skin area to be observed. Infrared light can penetrate the surface of the skin, be absorbed by the blood vessels under the skin, and scatter back.

[0201] In order to obtain a clearer image, multiple infrared light sources may be used to form a light array to illuminate different parts of the area to be collected. The computer program analyzes the brightness of the current image and automatically adjusts the intensity of each light source until the collected image meets the requirements.

[0202] Image processing and analysis:

[0203] After the imager collects the scattered infrared light, it processes these signals through image processing technology. This may include steps such as removing noise and enhancing contrast to extract a clear image of the blood vessels.

[0204] In some advanced applications, functional near-infrared spectroscopy (fNIRS) may be used to calculate the concentration of oxygenated and deoxygenated hemoglobin in blood vessels by monitoring changes in optical density, and further analyze the functional status of blood vessels. However, this step is not usually used for simple vascular imaging.

[0205] In situ projection displays vascular images:

[0206] The processed vascular images are projected onto the skin surface in situ using micro-projection technology, so that medical staff can directly observe the subcutaneous vascular conditions without any invasive operation.

[0207] This in-situ projection technology makes the vascular image completely correspond to the patient's actual skin position, greatly improving the accuracy and convenience of vascular positioning.

[0208] The process of using infrared spectral signals to generate subcutaneous blood vessel images is a comprehensive application that combines infrared spectroscopy technology, image processing technology, and micro-projection technology. This technology provides medical staff with a more intuitive and accurate means of locating blood vessels, which helps improve the efficiency and safety of medical operations.

[0209] Annotated data: Professional medical imaging experts annotate the images to mark the location, morphology and corresponding parameter feature range values ​​of abnormal blood vessels (such as blood vessel diameter, length, curvature, etc.).

[0210] 2. Data Preprocessing

[0211] Image denoising: Use filtering and other methods to remove noise from images and improve image quality.

[0212] Image standardization: normalize the image, enhance the contrast, and other processing to make the image data consistent and comparable.

[0213] Data enhancement: Increase the amount of data by rotating, scaling, flipping, etc. to improve the generalization ability of the model.

[0214] 3. CNN model construction

[0215] Select architecture: Select a suitable CNN architecture according to the specific task, such as AlexNet, VGGNet, ResNet, etc., and make necessary adjustments to adapt to the characteristics of medical images.

[0216] Define layer structure: Build a CNN model, including convolutional layers, pooling layers, fully connected layers, etc., to automatically extract features from images.

[0217] Loss function and optimization algorithm: Select appropriate loss function and optimization algorithm for model training.

[0218] 4. Model Training

[0219] Training model: The preprocessed image data is input into the CNN model for training, so that the model can learn how to identify the morphological characteristics of abnormal blood vessels.

[0220] Avoid overfitting: Use dropout, data augmentation and other techniques to avoid model overfitting.

[0221] Performance evaluation: Use indicators such as accuracy, recall, precision, and F1 score to evaluate model performance.

[0222] 5. Feature extraction and annotation

[0223] Feature extraction: Use the trained CNN model to extract features from new abnormal subcutaneous vascular images to obtain the morphological characteristics of the blood vessels.

[0224] Parameter calculation: Calculate the parameter feature range values ​​of abnormal blood vessels based on the extracted features, such as blood vessel diameter, length, curvature, etc.

[0225] Annotation results: The calculated parameter feature range values ​​are annotated on the image to facilitate further analysis and diagnosis by medical experts.

[0226] 6. Explanation and visualization

[0227] Explain model decisions: Use techniques such as Grad-CAM to visualize the decision-making process of the CNN model to help doctors understand and trust the model's predictions.

[0228] Result display: The annotated images and parameter feature range values ​​are displayed to doctors to assist them in making diagnosis and treatment decisions.

[0229] Precautions

[0230] Data Privacy: Ensure that the processing and storage of image data complies with data privacy and regulatory requirements.

[0231] Medical knowledge: When annotating and interpreting model results, certain medical knowledge is required to accurately understand and evaluate the morphological characteristics of abnormal blood vessels.

[0232] Model validation: Before actual application, the model should be fully validated and tested to ensure its accuracy and reliability.

[0233] The above steps are for reference only. The specific implementation may need to be adjusted and optimized according to the actual situation. In addition, tasks involving the medical field require special caution and are recommended to be performed under the guidance of a professional doctor.

[0234] Therefore, through model training and deployment, the blood monitoring data of the patient can be supervised for abnormalities through the pre-deployed blood abnormality AI model to determine whether the blood abnormality characteristics and / or abnormal vascular morphology characteristics identified by the model appear in the patient's blood monitoring data. If abnormal, the corresponding supervision signal is output and the processing system generates the voice alarm signal corresponding to the patient; otherwise, supervision continues. In this embodiment, as long as the model recognizes the abnormal blood characteristics or abnormal vascular morphology characteristics or recognizes both abnormalities at the same time, the corresponding supervision signal can be output and an alarm can be issued.

[0235] On the other hand, a blood routine test method is provided, which is implemented based on the above-mentioned blood routine non-invasive monitoring system, and the method comprises the following steps:

[0236] The patient wears and activates the blood routine non-invasive monitoring patch to establish a communication connection with the backend server;

[0237] The blood monitoring data of the patient is collected in real time through the non-invasive blood routine monitoring patch worn by the patient and uploaded to the backend server;

[0238] The backend server records the patient's blood monitoring data in real time, and determines whether the patient's blood monitoring data is abnormal. If abnormal, a voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and the blood routine non-invasive monitoring patch responds to the voice alarm data to remind the patient of abnormal blood monitoring; wherein, determining whether the patient's blood monitoring data is abnormal includes determining whether abnormal blood features and / or abnormal vascular morphological features identified by the model appear in the patient's blood monitoring data;

[0239] The patient logs in to the backend server through the patient APP to receive and view the blood monitoring data.

[0240] Please understand and implement the above method in conjunction with the description of the aforementioned patch and system, and will not be repeated here.

[0241] Figure 4 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, the electronic device 410 may include a first processor 2001 .

[0242] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0243] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0244] Combine the following Figure 4The components of the electronic device 410 are described in detail:

[0245] The first processor 2001 is the control center of the electronic device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or may be application specific integrated circuits (ASICs), or may be one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs).

[0246] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0247] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.

[0248] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 4 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0249] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0250] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0251] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0252] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0253] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0254] It should be noted that Figure 4 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than those shown in the figure, or combine certain components, or arrange the components differently.

[0255] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the methods described in the above method embodiments, which will not be repeated here.

[0256] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0257] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0258] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0259] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0260] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0261] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0262] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0263] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0264] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0265] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0266] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0267] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0268] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A non-invasive blood routine monitoring system, characterized in that: include: Non-invasive blood routine monitoring patch, used to monitor and report patients' blood monitoring data in real time; A backend server for managing the non-invasive blood routine monitoring patch worn by the patient; And, the patient's blood monitoring data is collected and recorded in real time, and the patient's blood monitoring data is supervised for abnormalities through a pre-deployed blood abnormality AI model to determine whether the patient's blood monitoring data is abnormal. If abnormal, a voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and the blood routine non-invasive monitoring patch responds to the voice alarm data to remind the patient of blood monitoring abnormalities; wherein, the determination of whether the patient's blood monitoring data is abnormal includes determining whether abnormal blood features and / or abnormal vascular morphological features identified by the blood abnormality AI model appear in the patient's blood monitoring data; Patient APP, used to log in to the backend server, receive and view the blood monitoring data; The non-invasive blood routine monitoring patch and the patient APP are respectively connected to the backend server for communication; The method for generating the blood abnormality AI model includes: Collecting and preprocessing blood routine test data of several patients with abnormal blood routine test data, wherein the blood routine test data includes abnormal blood data and abnormal subcutaneous blood vessel maps; A supervised learning algorithm is used to identify and extract the abnormal blood feature range values ​​in the abnormal blood data, and bind them with the corresponding examination items to generate corresponding blood abnormality feature data; CNN is used to extract the abnormal vascular morphological features in the abnormal subcutaneous vascular atlas and mark the parameter feature range values ​​of the corresponding abnormal blood vessels; The abnormal blood characteristic data and the abnormal blood vessel morphological characteristics are respectively labeled, and the labeling information includes a supervisory signal corresponding to the abnormal blood characteristic data; The annotated blood abnormality feature data and the blood vessel abnormality morphology features are statistically constructed into a feature set S: S={blood abnormality characteristic data, vascular abnormality morphological characteristics}; Dividing the feature set S into a training set and a validation set according to a preset ratio; Inputting the training set into a preset random forest model to perform feature training and learning to generate the blood abnormality AI model; The recognition performance of the blood abnormality AI model was verified using the verification set: If the verification is successful, the blood abnormality AI model is deployed on the backend server; If the verification fails, repeat the above steps and retrain the blood abnormality AI model.

2. The non-invasive blood routine monitoring system according to claim 1, characterized in that: Deployed on the backend server are: A processing system for providing data processing services based on multiple CPUs; The data collection system is used to communicate with the blood routine non-invasive monitoring patch worn by each patient to collect the patient's blood monitoring data; HIS system, used to generate electronic medical records of patients and record blood monitoring data and / or vascular monitoring data of patients; A blood analysis system, used to write the patient's blood monitoring data and generate the corresponding monitoring curve; MySQL database, used for background data storage; API interface, used to provide background communication port; The blood abnormality monitoring system is used to perform abnormal monitoring on the patient's blood monitoring data through a pre-deployed blood abnormality AI model to determine whether the blood abnormality characteristics and / or abnormal vascular morphology characteristics identified by the model appear in the patient's blood monitoring data. If abnormal, a corresponding monitoring signal is output and the processing system generates the voice alarm signal corresponding to the patient; otherwise, the monitoring continues; The data acquisition system, HIS system, blood analysis system, MySQL database, API interface and blood abnormality monitoring system are respectively connected to the processing system for communication.

3. A non-invasive blood routine monitoring patch as claimed in claim 1, characterized in that: The non-invasive blood routine monitoring patch comprises a skin-friendly layer and a non-invasive blood routine monitoring module attached to the skin-friendly layer, wherein the non-invasive blood routine monitoring module comprises: The infrared matrix is ​​composed of a number of infrared diodes arranged in a matrix, and is used to collect infrared spectrum signals of blood and feed them back to the controller. After the controller performs spectrum analysis, the corresponding blood monitoring data is obtained and forwarded to the wireless communication module; The controller includes: an MCU for logic calculation and control; an A / D converter for AD sampling, performing digital-to-analog signal conversion on the collected blood infrared spectrum signal to obtain a corresponding blood infrared spectrum electrical signal; a signal separator for performing signal separation on the blood infrared spectrum electrical signal to obtain the blood spectrum signal and the infrared spectrum signal of the subcutaneous blood vessel including: position, distribution and morphology, and the corresponding blood monitoring data and blood vessel monitoring data are obtained after being processed by the MCU; a blood signal module for recording the blood monitoring data and saving it in a memory, and forwarding the blood monitoring data and the blood vessel monitoring data to a wireless communication module through the MCU; A wireless communication module is used to upload the blood monitoring data to a backend server, which records and writes the data into the patient's electronic medical record in the HIS system and sends it to the patient's APP simultaneously; Power supply, used for power supply; The infrared matrix, wireless communication module and power supply are electrically connected to the controller respectively.

4. The non-invasive blood routine monitoring patch according to claim 3, characterized in that: The controller further includes: A timer, used for sending sampling signals at regular intervals; A memory for sampling and storage; An electronic screen for displaying the blood monitoring data; A power management module, used to manage the power supply of the power supply; The A / D converter, signal separator, blood signal module, timer, memory, electronic screen and power management module are electrically connected to the MCU respectively.

5. The non-invasive blood routine monitoring patch according to claim 4, characterized in that: The MCU is also used to process the blood infrared spectrum signal and generate corresponding blood vessel monitoring data, and forward it to the wireless communication module, which uploads it to the background server, and the background server records and writes it into the patient's HIS electronic medical record system.

6. The non-invasive blood routine monitoring patch according to claim 4, characterized in that: The controller further includes: The ISD voice chip is used to process the voice alarm signal sent by the background server and generate corresponding voice alarm data, wherein the generation method of the voice alarm signal includes: the background server monitors in real time and determines whether the patient's blood monitoring data is abnormal: If so, the voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and received by the wireless communication module of the blood routine non-invasive monitoring patch and forwarded to the ISD voice chip; Otherwise, the generation of the voice warning signal is abandoned.

7. The non-invasive blood routine monitoring patch according to claim 6, characterized in that: The non-invasive blood routine monitoring patch also includes: A voice module, used for receiving and responding to the voice alarm data; The voice module is electrically connected to the ISD voice chip.

8. A blood routine test method, implemented based on the blood routine non-invasive monitoring system according to any one of claims 1-2, characterized in that: The method comprises the following steps: The patient wears and activates the blood routine non-invasive monitoring patch to establish a communication connection with the backend server; The blood monitoring data of the patient is collected in real time through the non-invasive blood routine monitoring patch worn by the patient and uploaded to the backend server; The backend server records the patient's blood monitoring data in real time, and determines whether the patient's blood monitoring data is abnormal. If abnormal, a voice alarm signal corresponding to the patient is generated and sent to the blood routine non-invasive monitoring patch worn by the patient, and the blood routine non-invasive monitoring patch responds to the voice alarm data to remind the patient of abnormal blood monitoring; wherein, determining whether the patient's blood monitoring data is abnormal includes determining whether abnormal blood features and / or abnormal vascular morphological features identified by the model appear in the patient's blood monitoring data; The patient logs in to the backend server through the patient APP to receive and view the blood monitoring data.

9. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to claim 8 is implemented.

Images