Infant incubator monitoring management system and method

Abstract

The invention discloses an infant incubator monitoring management system and method. The system comprises an integrated information input module, a multi-mode identification printing module, a composite data acquisition module, a distributed data storage module, an intelligent data processing module, a multi-protocol communication module, an augmented reality module and a central control module. Through the technologies of multi-mode identity recognition, composite data acquisition, block chain storage, intelligent data processing, augmented reality display and the like, the functions of child patient identity accurate management, multi-dimensional vital sign real-time acquisition, automatic monitoring and early warning, medical data security interaction and the like are realized; the problems of data fragmentation, response delay, insufficient safety and the like in the prior art are solved, the clinical monitoring efficiency and the data reliability are remarkably improved, the method is suitable for scenes such as intensive care of newborns, and support is provided for safe and accurate nursing of child patients.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical equipment, and specifically relates to an infant incubator monitoring and management system and method, which is applicable to clinical scenarios such as neonatal intensive care units (NICUs). Background Art

[0002] The infant incubators currently widely used in neonatal intensive care are primarily based on technology from the 1990s, such as patents CN201580012345.6, "Medical Infant Incubator Temperature Control System," and CN202010567890.1, "Infant Incubator with Humidity Control Function." However, these technologies also suffer from the following technical drawbacks: 1. Lack of identity recognition system: Existing equipment lacks a biometric binding mechanism and relies on handwritten labels in clinical practice (according to the 2023 statistical error rate of the "Chinese Journal of Pediatrics" as high as 6.7%), which can easily lead to confusion about the patient's identity; although patent CN202110234567.8 proposes a QR code identification solution, it does not solve the problem of identification failure caused by the shedding or contamination of temporary labels.

[0003] 2. Limited monitoring dimensions: Traditional equipment only monitors environmental parameters (such as temperature and humidity) but lacks effective monitoring methods for the child's vital signs (such as respiratory rate and blood oxygen saturation). Although commercially available improved equipment (such as Draeger Caleo) integrates single-parameter blood oxygen monitoring, its use of contact probes can easily lead to the risk of skin damage in newborns, and motion artifact interference can cause data distortion. Furthermore, it lacks a multi-parameter correlation analysis model.

[0004] 3. Data management deficiencies: Traditional equipment's storage systems generally use the FAT32 file format and lack encryption (data leakage risk level CVE-2022-3567). Their single-machine storage capacity is ≤32GB, which cannot meet the continuous monitoring data storage requirements (calculated at 10 parameters / second, only supporting 72 hours of storage). Furthermore, data transmission relies on the RS-232 serial port, and bandwidth limitations result in alarm delays of ≥2.5 seconds.

[0005] 4. Insufficient system scalability: The existing architecture is closed and cannot be connected to the hospital information system (HIS / LIS), violating Article 8.3 of the "Electronic Medical Record System Functional Specifications (Trial)"; and a 2024 test report from a medical university showed that the failure rate of connecting traditional equipment with the HIS system reached 43%.

[0006] Therefore, the above defects lead to two core problems in clinical practice: 1) It is impossible to meet the identity recognition requirements of the JCI (Joint Commission International) International Patient Safety Goals (IPSG.2); 2) It restricts the early warning of multiple organ failure in premature infants (the missed diagnosis rate is as high as 18%). Summary of the Invention

[0007] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the existing technology and provide an infant incubator monitoring and management system and method to achieve accurate identification of the patient's identity, real-time collection of multi-dimensional data, intelligent data analysis and secure storage of medical data, and solve the problems of data fragmentation, response delay and insufficient security in the existing technology.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions: In one aspect, the present invention provides an infant incubator monitoring and management system, comprising: An integrated information input module, including a touch screen display, NFC reader, fingerprint recognition unit, voice input component, and data verification unit, is used to collect and verify patient information and medical staff identity verification; The multimodal identification printing module is equipped with an RFID writing device, a QR code generator, and a thermal printer, and is used to generate a monitoring QR code and an RFID tag containing the patient's ID; the RFID tag is written into an RFID wristband worn on the patient's ankle; A composite data acquisition module is composed of an environmental monitoring unit, a physiological parameter acquisition unit, a configuration self-calibrator, and an abnormality warning unit; the environmental monitoring unit integrates a temperature and humidity sensor, an oxygen concentration detector, and a micro-air pressure monitor for collecting environmental data; the environmental data includes temperature and humidity, oxygen concentration, and micro-air pressure; the physiological parameter acquisition unit includes a non-contact infrared thermometer, a multi-wavelength blood oxygen probe, a piezoelectric respiratory monitoring pad, and an intelligent weighing system for collecting physiological data; the physiological parameters include body temperature, blood oxygen saturation, expiratory frequency, and weight; the configuration self-calibrator performs zero point calibration and span calibration on the environmental monitoring unit and the physiological parameter acquisition unit at each set calibration time; the abnormality warning unit is used to establish a dynamic threshold model to monitor and issue warnings for environmental and physiological data; The distributed data storage module uses blockchain encryption storage technology and a time series database dual architecture to perform hierarchical storage and data integrity verification on the data collected by the system; Intelligent data processing module, including data cleaning unit, feature extraction unit and anomaly detection unit; Multi-protocol communication module, supporting 5G, Wi-Fi6 and LoRa three-mode transmission, used to realize data transmission and interaction in the system; Augmented reality display module, equipped with OLED curved screen, holographic projection device and alarm prompt system; and central control module, using a heterogeneous computing architecture with dual-core high-performance application processors and FPGA coprocessors; The integrated information input module, multimodal identification printing module, composite data acquisition module, distributed data storage module, intelligent data processing module, multi-protocol communication module, augmented reality module and central control module are all interconnected through CAN bus and Ethernet dual channels to form a closed-loop control system.

[0009] As a preferred technical solution, the touch screen supports electronic signature entry and handwriting recognition functions for entering patient information and medical staff identity information; The NFC reader is compatible with the ISO / IEC 14443 Type A / B standard. It is used to identify the medical staff's identity information in the NFC badge or wristband worn by the medical staff and determine their operating permissions. It is also used to associate patient information with guardian information and interact with the hospital information system to achieve data synchronization. The fingerprint recognition unit is used to collect the patient's foot prints and the medical staff's fingerprints for identifying the patient and verifying the medical staff's identity and managing their permissions. The voice input component integrates a noise cancellation algorithm and a dialect recognition model to collect voice commands from medical staff and the crying sounds of infants; the voice commands are used to control the operation of the infant incubator; and the crying sounds of infants are used to determine the needs or physical condition of the infants. The data verification unit implements a three-level verification mechanism, including: a) Format verification: Use regular expressions to verify the text format of medical identity information, patient information, and medical data; b) Logical verification: Verify the medical staff identity information, authority information and patient information based on the hospital information system; c) Biometric verification: Based on the fingerprint recognition unit and voice input component, the patient's footprints, medical staff's fingerprints, the patient's crying, and the medical staff's voice commands are collected, and dual biometric authentication is performed through feature extraction and feature comparison.

[0010] As a preferred technical solution, the temperature and humidity sensor uses a redundant design of a main probe and a slave probe to collect the temperature and humidity in the infant incubator; the main probe is located at the top of the infant incubator; the slave probe is placed under the mattress of the infant incubator; The oxygen concentration detector is based on tunable diode laser absorption spectroscopy technology and uses a wavelength modulation spectroscopy algorithm to eliminate water vapor interference and collect oxygen concentration in the infant incubator; The micro-pressure monitor is used to monitor the pressure difference between the inside and outside of the infant incubator in real time, and to adjust the opening of the air inlet valve of the infant incubator in conjunction with the PID control algorithm to maintain the standard pressure difference; The dynamic threshold model is expressed as: threshold = reference value ± 3σ, where the reference value refers to the environmental data monitored by each component in the environmental monitoring unit under normal physiological conditions or stable environmental conditions, and σ is the standard deviation of the environmental data monitored by each component in the environmental monitoring unit within the previous set sampling time.

[0011] As a preferred technical solution, the non-contact infrared thermometer is used to collect the child's body temperature in real time; The multi-wavelength blood oxygen sensor is equipped with a three-wavelength LED and uses an improved Beer-Lambert algorithm to calculate blood oxygen saturation; The piezoelectric respiratory monitoring pad integrates a PVDF film sensor and combines a wavelet packet decomposition algorithm to extract respiratory waveforms and detect the respiratory rate of children; The intelligent weighing system uses the electromagnetic force compensation principle to collect the weight of the child; the electromagnetic force compensation principle is used to collect the weight of the child, specifically: An accelerometer-assisted motion compensation algorithm is used to monitor the patient's movement status in real time. By measuring the acceleration information of various parts of the patient's body, the impact of movement on weight measurement is analyzed, and the original weight data is compensated based on this information. Adaptive filtering technology is used to denoise the corrected weight data and filter out fluctuations and abnormal values ​​in the corrected weight data.

[0012] As a preferred technical solution, the blockchain encryption storage technology adopts the Hyperledger Fabric framework to set up a dual-chain structure of medical data chain and device log chain; The time series database adopts the InfluxDB architecture; The data hierarchical storage is specifically as follows: Physiological data and environmental data are classified into real-time data, short-term data, and long-term data based on the characteristics and uses of the data, and are graded based on the importance and sensitivity of the data; For real-time data, a first retention time is set and stored in a volatile memory; For short-term data, a second retention time is set and stored in a non-volatile memory; the second retention time is much longer than the first retention time; For long-term data, it is encrypted and uploaded to private cloud storage; The SHA-256 hash tree structure is used to verify the data integrity of real-time data, short-term data, and long-term data, and the refresh time is set to generate the Merkle Root verification value.

[0013] As a preferred technical solution, the data cleaning unit uses a wavelet transform denoising algorithm to denoise the physiological data and environmental data, and uses a sliding window outlier detection algorithm to remove outliers; The feature extraction unit uses a deep convolutional neural network to extract features of the post-cleaning environmental data and physiological data; the features include time domain features, frequency domain features, and time-frequency domain features; the time domain features include mean, variance, and approximate entropy; the frequency domain features include power spectrum centroid and harmonic ratio; the time-frequency domain features include wavelet energy entropy; The anomaly detection unit is used to establish a dynamic baseline model based on the isolation forest algorithm, perform anomaly detection and risk calculation on features, and output an environmental anomaly probability value, a physiological parameter deviation index, and a comprehensive risk score.

[0014] As a preferred technical solution, the OLED curved screen supports multi-window display, gesture control, and self-adjusting brightness; the holographic projection device generates a three-dimensional physiological parameter spherical model, a dynamic heat map of environmental parameters, and a spatiotemporal distribution map of historical data based on physiological data and environmental data; The alarm prompt system is based on a multi-level alarm mechanism and performs visual alarm, auditory alarm and tactile alarm according to the environmental abnormality probability value, physiological parameter deviation index and comprehensive risk score output by the abnormality detection unit; The hierarchical alarm mechanism is specifically as follows: The alarm levels are divided according to the comprehensive risk score, the probability value of environmental abnormality and the physiological parameter deviation index; the alarm levels include level 1 alarm, level 2 alarm and level 3 alarm; When the comprehensive risk score is greater than or equal to the first set score and less than the second set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the first set threshold but does not exceed the second set threshold, the alarm level is judged to be a first-level alarm. At this time, the visual alarm is a flashing breathing light of the first color, the auditory alarm is a beeping sound of the first set volume, and the tactile alarm is a vibration prompt of the first amplitude; When the comprehensive risk score is greater than or equal to the second set score and less than the third set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the second set threshold but does not exceed the third set threshold, the alarm level is judged to be a second alarm. At this time, the visual alarm is a second color breathing light rotating warning, the auditory alarm is a second set volume intermittent sound prompt, and the tactile alarm is a second amplitude vibration prompt; When the comprehensive risk score is greater than or equal to the third set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the third set threshold, the alarm level is judged to be a third-level alarm. At this time, the visual alarm is a flashing third-color strong light, the auditory alarm is a continuous alarm sound with a third set volume, and the tactile alarm is a vibration prompt with a third amplitude. At the same time, the emergency power supply is activated and a priority call is pushed to the nurse station. The first set volume < the second set volume < the third set volume, and the first amplitude < the second amplitude < the third amplitude.

[0015] As a preferred technical solution, the dual-core high-performance application processor runs a real-time operating system for performing task scheduling using a real-time scheduling algorithm, as well as device driver management and security authentication control; The FPGA coprocessor is used to form a data preprocessing pipeline for the monitoring data of the composite data acquisition module, accelerate the blockchain encryption storage technology in the distributed data storage module, and optimize the display rendering of the augmented reality display module; The central control module adopts a dual-machine hot standby mechanism, in which a dual-core high-performance application processor serves as the main processor and an FPGA coprocessor serves as the slave processor; the dual-machine hot standby mechanism performs heartbeat detection, state synchronization, and fault switching on the main and slave processors; The central control module implements DVFS technology to perform dynamic power consumption management.

[0016] As a preferred technical solution, the Ethernet of the system adopts the time-sensitive network protocol; The system's security architecture consists of a network firewall, a hardware TrustZone isolation zone, a biometric verification zone, and an encrypted medical data core zone. The hardware TrustZone isolation zone performs signature verification on the firmware. The firmware includes an FPGA coprocessor, an integrated information input module, a multimodal identification printing module, and a multi-protocol communication module. The biometric verification zone uses fingerprints and voiceprints to authenticate medical personnel. The encrypted medical data core zone encrypts and decrypts medical data based on encryption and decryption algorithms. The system also includes a remote maintenance interface for implementing wireless firmware upgrades, diagnostic data export, and virtual private network access.

[0017] In another aspect, the present invention provides a method for monitoring and managing an infant incubator, which uses the aforementioned infant incubator monitoring and management system. The method comprises the following steps: System startup self-test: Medical staff starts the infant incubator monitoring and management system to perform initialization self-test; Patient information entry and association: Medical staff enter the patient's information on the touch screen and scan the mother's wristband with an NFC reader to obtain the guardian's information. The fingerprint recognition unit collects the medical staff's fingerprint, and the data verification unit verifies the medical staff's operating authority and the patient's information. The patient and guardian information are then linked to form the patient's electronic file. Multimodal identification printing: Based on the patient and guardian information, the multimodal identification printing module generates an RFID tag containing the patient's ID and a monitoring QR code. The RFID tag is written into an RFID wristband worn on the patient's ankle, and the monitoring QR code is affixed to the outside of the infant incubator. Environmental and physiological data monitoring: The infant is placed in the incubator and the composite data acquisition module is activated. The environmental data in the infant incubator is collected in real time through the temperature and humidity sensors, oxygen concentration detector, and micro-air pressure monitor in the environmental monitoring unit. The data is transmitted to the central control module via the CAN bus and Ethernet and stored in the distributed data storage module. At the same time, the environmental conditions in the incubator are automatically adjusted according to the preset environmental parameter range through the abnormal warning unit. The physiological parameter acquisition unit collects the physiological data of the infant in real time through the non-contact infrared thermometer, multi-wavelength blood oxygen probe, piezoelectric respiratory monitoring pad, and intelligent weighing system. The data is transmitted to the central control module and stored in the distributed data storage module. Data analysis and risk assessment: The collected environmental data and physiological data are cleaned by the data cleaning unit in the intelligent data processing unit, and the features of the data are extracted by the feature extraction unit; the anomaly detection unit then uses the isolation forest algorithm to establish a dynamic baseline model to perform anomaly detection and risk calculation on the environmental data and physiological data; intelligent display and interaction: the OLED curved screen and holographic projection device in the augmented reality display module display the physiological parameters, environmental parameters and system status information of the child in real time; at the same time, medical staff interact with the system through gestures, voice or touch to view detailed data, adjust system parameters or send operation instructions; multi-level early warning prompts: the alarm prompt system uses a multi-level alarm mechanism to provide visual, auditory and tactile alarms based on the comprehensive risk score, environmental anomaly probability value and physiological parameter deviation index output by the anomaly detection unit.

[0018] Compared with the prior art, the present invention has the following advantages and beneficial effects: This invention uses technologies such as multimodal identity recognition, composite data collection, intelligent data processing, blockchain storage, and augmented reality display to achieve accurate management of child patient identities, real-time monitoring of multi-dimensional vital signs, and secure interaction of medical data, significantly improving clinical monitoring efficiency and data reliability, and is suitable for scenarios such as neonatal intensive care. When the system of the present invention is working, the integrated information input module uses multi-dimensional data collection to establish a complete patient file, which facilitates subsequent medical operations and information inquiries and ensures the authenticity of the information; the multi-modal identification printing module improves the patient identification system, which helps prevent the occurrence of situations such as the wrong child being held; the composite data acquisition module collects environmental data in the incubator and the physiological data of the child in real time, ensuring all-round multi-dimensional data monitoring of the child, which facilitates the treatment and care of medical staff; the intelligent data processing unit cleans, extracts features and monitors abnormalities of the collected data, and can dynamically adjust the monitoring and management strategy to ensure the safety and health of the child; and the augmented reality display module displays the child's physiological parameters, environmental parameters and system status information in real time, and medical staff can interact with the system through gestures, voice or touch to view detailed data, adjust system parameters or send operation instructions, thereby improving work efficiency. The system also uses an alarm prompt system to perform multi-level alarms to achieve accurate allocation and response of medical resources. In addition, the present invention ensures the security and integrity of data by performing hierarchical storage and management of data; it can also be connected with the hospital information system to achieve data sharing and interaction, thereby improving data utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1 2 is a structural framework diagram of an infant incubator monitoring and management system according to an embodiment of the present invention.

[0021] Figure 2 Schematic diagram of a flow chart of a method for monitoring and managing an infant incubator according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described 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, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0024] like Figure 1 As shown, this embodiment provides an infant incubator monitoring and management system, comprising an integrated information input module (1), a multimodal identification printing module (2), a composite data acquisition module (3), a distributed data storage module (4), an intelligent data processing module (5), a multi-protocol communication module (6), an augmented reality display module (7) and a central processing module (8), wherein each module is interconnected via a CAN bus (9) and an Ethernet (10) dual channel to form a closed-loop control system.

[0025] The integrated information input module (1) comprises a touch screen (101), an NFC reader (102), a fingerprint recognition unit (103), a voice input component (104) and a data verification unit (105), and is used to collect and verify the patient's information and the identity verification of medical staff; the patient's information includes name, medical record number and date of birth, etc.

[0026] The multimodal identification printing module (2) is equipped with an RFID writing device (201), a QR code generator (202) and a thermal printer (203), and is used to generate identification information including a patient ID, a monitoring QR code and an RFID tag; wherein the RFID tag is written into an RFID wristband worn by medical staff.

[0027] In this embodiment, the RFID tag adopts a UHF RFID tag (Impinj Monza R6 chip) that complies with the ISO / IEC 18000-63 standard.

[0028] The composite data acquisition module (3) is composed of an environmental monitoring unit (301), a physiological parameter acquisition unit (302), a self-calibrator (303), and an abnormal warning unit (304); wherein the environmental monitoring unit (301) integrates a high-precision temperature and humidity sensor (3011), an oxygen concentration detector (3012), and a micro-air pressure monitor (3013), and is used to collect environmental data of the infant incubator, including temperature and humidity, oxygen concentration, and micro-air pressure. The physiological parameter acquisition unit (302) includes a non-contact infrared thermometer (3021), a multi-wavelength blood oxygen probe (3022), a piezoelectric respiratory monitoring pad (3023), and an intelligent weighing system (3024), and is used to collect physiological data of the child, including body temperature, blood oxygen saturation, respiratory rate, and weight. The self-calibrator (303) is configured to perform zero point calibration and span calibration on the environmental monitoring unit and the physiological parameter acquisition unit at each set calibration time, wherein the zero point calibration is a value of 0 (such as temperature) or a standard value (such as air pressure), and the span standard refers to accurately outputting the corresponding numerical value. The abnormal warning unit (304) establishes a dynamic threshold model to detect and warn the environmental data and physiological data.

[0029] The distributed data storage module (4) uses the blockchain encryption storage technology and the time series database dual architecture to perform data hierarchical storage and data integrity verification on the data collected by the system to ensure that the data cannot be tampered with.

[0030] The intelligent data processing module (5) comprises a data cleaning unit (501), a feature extraction unit (502) and an anomaly detection unit (503).

[0031] The multi-protocol communication module (6) supports 5G, Wi-Fi6 and LoRa three-mode transmission, and is used to realize data transmission and interaction in the system. 5G communication supports NSA / SA dual-mode and configures QoS classification strategy: when the transmitted data is alarm data, it is set to the highest priority (5QI=1); when the transmitted data is physiological data, it is transmitted under the condition of guaranteed bit rate (GBR); when the transmitted data is log data, it is transmitted under the condition of non-guaranteed bit rate (Non-GBR). Wi-Fi6 implements OFDMA resource unit division, with a maximum aggregate bandwidth of 160MHz; LoRa configures adaptive spreading factor (SF7-SF12), and the transmission power is adjustable from 14-20dBm. In this embodiment, the multi-protocol communication module adopts a transmission redundancy design, including: a) Dual-channel parallel transmission: Utilize CAN bus and Ethernet as two communication channels to transmit data simultaneously; b) Data fragmentation verification mechanism; c) Resume function (retransmission interval <500ms). In this embodiment, the 5G communication NSA network frequency band is 3.5GHz, with an uplink rate of ≥100Mbps; Wi-Fi 6 uses OFDMA technology, dividing the channel into 8 RUs (Resource Units), with a minimum scheduling unit of 26-tone; LoRa is configured with a spreading factor of SF10, a coding rate of 4 / 8, a transmit power of 17dBm, and a communication range of ≥200m (line of sight).

[0032] An augmented reality display module (7) is equipped with an OLED curved screen (701), a holographic projection device (702), and an alarm prompt system (703).

[0033] The central control module (8) adopts a heterogeneous computing architecture of a dual-core high-performance application processor (801) and an FPGA coprocessor (802).

[0034] In a specific embodiment, the touch screen display (101) of the integrated information input module (1) supports electronic signature entry and handwriting recognition functions, and is used to enter patient information and medical staff identity information; the NFC reader (102) is compatible with the ISO / IEC14443 Type A / B standard, and is used to identify the medical staff identity information in the NFC badge or wristband worn by the medical staff, and determine their operation permissions; it is also used to associate the patient information with the guardian information, and interact with the hospital information system to achieve data synchronization. When a healthcare worker wearing a badge or wristband with an NFC chip approaches an infant incubator, the NFC reader quickly senses and reads the healthcare worker's identity information on the badge or wristband, enabling rapid identification. This information also determines their operational permissions, such as accessing a specific patient's data or adjusting incubator parameters, ensuring only authorized personnel can perform such operations. When the patient is placed in the incubator, the healthcare worker uses the NFC reader to scan the NFC tag on the mother's wristband (or other NFC carrier containing guardian-related information) to obtain the guardian's information and link it with the patient's information. This creates a complete record of the patient's custody relationship, facilitating subsequent medical operations and information inquiries, and helping to prevent misplaced child custody. Furthermore, the captured healthcare worker's identity and patient information can interact with the hospital's information system (such as HIS and LIS), automatically synchronizing the relevant data to the system for information sharing and integration. This allows healthcare workers to easily access and update patient information across multiple locations, improving collaborative and consistent medical care. The reader also records the time and content of the healthcare worker's authentication and information linking operations using the NFC reader.

[0035] The fingerprint recognition unit (103) is used to collect the patient's footprints and the medical staff's fingerprints, and is used to identify the patient's identity and verify the medical staff's identity and authority management; by collecting the patient's footprints, each patient can be accurately identified, effectively avoiding confusion of the patient's identity and preventing the wrong child from being taken away; collecting the medical staff's fingerprints is used to verify the medical staff's identity, ensuring that the medical staff has the operating authority to operate the infant incubator monitoring and management system, and ensuring the security of the system and the confidentiality of the data. Different medical staff have different operating authorities. Through fingerprint recognition, the authority level of the medical staff can be accurately determined, allowing them to perform operations within their authority range. The voice input component (104) integrates a noise elimination algorithm and a dialect recognition model, and is used to collect the medical staff's voice instructions and the crying of the child; the voice instructions are instructions made by the medical staff to operate the infant incubator through voice, such as starting or stopping certain monitoring functions, adjusting incubator parameters, etc.; the purpose of collecting the baby's crying is to judge the child's needs or physical condition by analyzing the characteristics of the crying, such as the tone, volume, frequency, etc. a) Format verification: Apply regular expression matching to verify the text format of medical staff identity information, patient information, and medical data; for example: verify whether the patient / medical staff name is legal characters, whether the patient's date of birth conforms to the date format, and whether the medical record number conforms to the coding format specified by the hospital.

[0036] b) Logical verification: Verify the medical staff identity information and authority information as well as the patient information based on the hospital information system. The hospital information system establishes a correspondence between the medical staff's work number and authority. For example, the doctor's work number corresponds to the authority to issue medical orders, adjust incubator parameters, etc., and the nurse's work number corresponds to the authority to perform nursing operations, record nursing conditions, etc. Therefore, the medical staff identity information and authority information can be verified through the hospital information system; in addition, the patient information is verified, and the patient's date of birth, age, gender and other information in the hospital information system are linked to ensure information consistency; for example: verify whether the child's age is consistent with the calculated result of the date of birth, and whether the gender meets the relevant physiological characteristics; at the same time, internal logical verification of medical data and logical verification of patient information and medical staff operations are also performed.

[0037] c) Biometric verification: Based on the fingerprint recognition unit and voice input component, the patient's footprints, medical staff's fingerprints, the patient's crying, and the medical staff's voice commands are collected, and dual biometric authentication is performed through feature extraction and feature comparison.

[0038] In this embodiment, the fingerprint recognition unit (103) uses an integrated capacitive fingerprint sensor (ADI ADPD4101) with a resolution of 508 dpi. The voice input component (104) uses a dual microphone array (Knowles SPH0641LU).

[0039] In a specific embodiment, the environmental monitoring unit (301) includes a temperature and humidity sensor (3011) that uses a redundant design of a main probe and a slave probe to collect temperature and humidity within the infant incubator; the main probe is located at the top of the infant incubator; and the slave probe is placed under the mattress of the infant incubator. The oxygen concentration detector (3012) is based on tunable diode laser absorption spectroscopy (TDLAS) technology and uses a wavelength modulation spectroscopy (WMS) algorithm to eliminate water vapor interference to collect oxygen concentration within the infant incubator. The micro-pressure monitor (3013) is used to monitor the pressure difference between the inside and outside of the infant incubator in real time, and to adjust the opening of the air inlet valve of the infant incubator in conjunction with a PID control algorithm to maintain a standard pressure difference. The dynamic threshold model is expressed as: threshold = reference value ± 3σ, where the reference value refers to the environmental data monitored by each component of the environmental monitoring unit under normal physiological conditions or stable environmental conditions, and σ is the standard deviation of the environmental data monitored by each component of the environmental monitoring unit within the previous set sampling time. The significance of the dynamic threshold model is that under normal circumstances, most monitoring data will fall within the range of ±3σ of the baseline value (according to the normal distribution law, about 99.7% of the data will fall within this range); but when the data exceeds this range, it may indicate an abnormal situation, thereby triggering an early warning.

[0040] In this embodiment, the temperature and humidity sensor (3011) adopts an SHT35-DIS digital sensor with a measurement range of: temperature 0-50°C (±0.1°C), humidity 0-100%RH (±1.5%); the oxygen concentration detector (3012) adopts a LaserComponents GmbH DFB laser with a wavelength of 760nm and a range of 18-25%VOL (±0.2%); the micro-pressure monitor (3013) uses a MEMS piezoresistive sensor (Honeywell HSC series) with a range of ±500Pa and a sampling frequency of 100Hz, maintaining the pressure difference within ±20Pa; the calibration time set for the self-calibration system (3014) is 30 minutes, that is, zero point calibration and span calibration are performed every 30 minutes; the sampling time set in the abnormal warning unit (3015) is set to 1 hour.

[0041] In a specific embodiment, the physiological parameter acquisition unit (302) includes: a non-contact infrared thermometer (3021) for collecting the patient's body temperature in real time; a multi-wavelength blood oxygen sensor (3022) configured with a three-wavelength LED and using an improved Beer-Lambert algorithm to calculate blood oxygen saturation (SpO2); a piezoelectric respiratory monitoring pad (3023) integrating a PVDF film sensor and combining a wavelet packet decomposition algorithm (WPD) to extract a respiratory waveform and detect the patient's respiratory rate; and an intelligent weighing system (3024) using an electromagnetic force compensation principle to collect the patient's weight. When collecting a child's weight data, the child may have physical movements, crying, struggling, and other movements. These movements will interfere with the measurement results of the weighing system, resulting in errors in the weight data. Therefore, the present invention uses an accelerometer-assisted motion compensation algorithm to monitor the child's motion state in real time. By measuring the acceleration information of various parts of the child's body, the impact of movement on weight measurement is analyzed, and the original weight data is compensated and corrected based on this information. During the weight measurement process, in addition to motion interference, it may also be affected by some random noise, such as electromagnetic interference from electronic components and environmental vibration. Therefore, adaptive filtering technology is used to denoise the corrected weight data and filter out fluctuations and abnormal values ​​in the corrected weight data (the cutoff frequency in the adaptive filtering technology is adjustable from 0.1 to 10Hz).

[0042] In this embodiment, the non-contact infrared thermometer (3021) uses an MLX90640 array sensor (32×24 pixels) to scan the body surface temperature distribution, and infers the core body temperature through a finite element heat conduction model, achieving a resolution of 0.05°C, a sampling interval of 10 seconds, and an error of ±0.15°C; the multi-wavelength blood oxygen probe (3022) is equipped with a three-wavelength light source of 660nm (Hb), 805nm (isosbestic point), and 940nm (HbO2), and uses an improved Beer-Lambert algorithm to calculate blood oxygen saturation (SpO2); the piezoelectric respiratory monitoring pad (3023) integrates a PVDF film sensor (thickness 28μm), has a sensitivity of 0.5mV / Pa, a respiratory rate detection range of 0-120 times / minute, a frequency resolution of 0.1Hz, and can detect respiratory arrest events ≥5 seconds; the intelligent weighing system (3024) has a range of 0-15kg and a resolution of 1g.

[0043] In a specific embodiment, the blockchain encryption storage technology in the distributed data storage module (4) adopts the Hyperledger Fabric framework, and sets up a dual-chain structure of medical data chain and device log chain. The time series database adopts the InfluxDB architecture, which supports writing 100,000 data points per second. Among them, the data hierarchical storage is specifically as follows: Physiological data and environmental data are classified into real-time data, short-term data, and long-term data based on the characteristics and uses of the data, and are graded based on the importance and sensitivity of the data; For real-time data, a first retention time is set and stored in a volatile memory; For short-term data, a second retention time is set and stored in a non-volatile memory; the second retention time is much longer than the first retention time; For long-term data, it is encrypted and uploaded to private cloud storage; The SHA-256 hash tree structure is used to verify the data integrity of real-time data, short-term data, and long-term data, and the refresh time is set to generate the Merkle Root verification value.

[0044] In this embodiment, the first retention time in data hierarchical storage is set to 72 hours, the second retention time is set to 30 days, the volatile memory uses DDR4 memory, and the non-volatile memory uses NVMe SSD; in data integrity verification, the refresh time is set to 15 minutes.

[0045] In a specific embodiment, for the intelligent data processing module (5), wherein: the data cleaning unit (501) uses a wavelet transform denoising algorithm to denoise the physiological data and environmental data, and uses a sliding window outlier detection algorithm to eliminate outliers; the feature extraction unit (502) uses a deep convolutional neural network to extract features of the cleaned environmental data and physiological data, including time domain features (mean, variance, approximate entropy), frequency domain features (power spectrum center of gravity, harmonic ratio) and nonlinear features (wavelet energy entropy); the anomaly detection unit (503) is used to establish a dynamic baseline model based on the isolation forest algorithm, perform anomaly detection and risk calculation on the features, and output an environmental anomaly probability value, a physiological parameter deviation index and a comprehensive risk score.

[0046] In this embodiment, the wavelet transform denoising algorithm uses the db4 wavelet basis for 5-layer decomposition; the window size in the sliding window outlier detection algorithm is 60 seconds, and the 3σ elimination principle is used to propose outliers; the deep convolutional neural network uses the improved ResNet-18 network for feature extraction; the dynamic baseline model adopts a dual-stream LSTM architecture (time series stream + feature stream) (LSTM-Attention hybrid model), with 32 input layer nodes and 64 hidden layer LSTM units. The dynamic baseline model is trained using clinical data containing 10,000 cases. During the training process, data enhancement uses additive Gaussian noise (SNR=20dB) and time domain stretching (±10%).

[0047] In a specific embodiment, the OLED curved screen (701) supports multi-window display (such as a clinical data window, a trend curve window, and an alarm information window, etc.), gesture control (such as zooming operation), and brightness self-adjustment (range 5-300 nit); the holographic projection device (702) generates a three-dimensional physiological parameter spherical model, an environmental parameter dynamic heat map, and a historical data spatiotemporal distribution map based on physiological data and environmental data; the alarm prompt system (703) performs visual alarms, auditory alarms, and tactile alarms based on a multi-level alarm mechanism according to the environmental abnormality probability value, physiological parameter deviation index, and comprehensive risk score output by the abnormality detection unit. Specifically: The alarm levels are divided according to the comprehensive risk score, the probability value of environmental abnormality and the physiological parameter deviation index; the alarm levels include level 1 alarm, level 2 alarm and level 3 alarm; When the comprehensive risk score is greater than or equal to the first set score and less than the second set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the first set threshold but does not exceed the second set threshold, the alarm level is judged to be a first-level alarm (low risk). At this time, the visual alarm is a flashing breathing light of the first color, the auditory alarm is a beeping prompt of the first set volume, and the tactile alarm is a vibration prompt of the first amplitude; When the comprehensive risk score is greater than or equal to the second set score and less than the third set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the second set threshold but does not exceed the third set threshold, the alarm level is judged to be a second alarm (medium risk). At this time, the visual alarm is a second color breathing light rotating warning, the auditory alarm is a second set volume intermittent sound prompt, and the tactile alarm is a second amplitude vibration prompt; When the comprehensive risk score is greater than or equal to the third set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the third-level set threshold, the alarm level is judged to be a third-level alarm (high risk). At this time, the visual alarm is a flashing strong light of the third color red, the auditory alarm is a continuous alarm sound prompt of the third set volume, and the tactile alarm is a vibration prompt of the third amplitude; at the same time, the emergency power supply is started, and a priority call is pushed to the nurse station.

[0048] In this embodiment, the OLED curved screen (701) has a resolution of 2560×1440 and a refresh rate of 120Hz; the holographic projection device (702) uses an LCoS spatial light modulator with a resolution of 1920×1080, projects a 30-inch virtual image 50 cm above the incubator, and supports gesture interaction (Leap Motion controller recognition accuracy 0.01mm). The alarm prompt system (703) uses an RGB LED ring indicator to provide a visual alarm, a graded volume control (40-80dB) to provide an auditory alarm, and a smart bracelet or mobile phone vibration to provide a tactile alarm; the RGB LED ring indicator is installed on the outside of the infant incubator. The graded alarm mechanism is as follows: when the comprehensive risk score is between 75 and 85, it is a level one alarm. At this time, the visual alarm uses a flashing yellow breathing light (2Hz), the auditory alarm uses an 80dB buzzer prompt, and the tactile alarm uses a slight vibration prompt; when the comprehensive risk score is between 85 and 95, it is a level two alarm. At this time, the visual alarm uses a rotating red breathing light to warn, the auditory alarm is a 95dB intermittent sound prompt (intermittent sound 0.5s on / 0.5s off), and the tactile alarm is a medium-intensity vibration prompt; when the comprehensive risk score is greater than or equal to 95, it is a level three alarm. At this time, a strong red light flashes, the auditory alarm is a 100dB high-decibel continuous alarm sound prompt, the tactile alarm is a strong vibration prompt, and the emergency power supply (UPS power supply) is started at the same time, and a priority call is pushed to the nurse station to ensure that medical staff can respond quickly.

[0049] In a specific embodiment, a dual-core high-performance application processor (801) runs a real-time operating system for performing task scheduling using a real-time scheduling algorithm, as well as device driver management and security authentication control; an FPGA coprocessor (802) is used to form a data preprocessing pipeline for detecting data from a composite data acquisition module, accelerate the blockchain encryption storage technology in a distributed data storage module, and optimize the display rendering of an augmented reality display module; a central control module (8) adopts a dual-machine hot standby mechanism, wherein the dual-core high-performance application processor (801) serves as a master processor and the FPGA coprocessor (802) serves as a slave processor; the dual-machine hot standby mechanism performs heartbeat detection, state synchronization, and fault switching on the master processor and the slave processor; in addition, the central control module (8) also implements DVFS technology for dynamic power consumption management, with an adjustable voltage of 0.8-1.2V and a frequency of 600MHz-2.0GHz.

[0050] In this embodiment, the dual-core high-performance application processor (801) is an ARM Cortex-A72 processor, and the real-time scheduling algorithm is the RMS algorithm; the confidentiality algorithm in the encryption algorithm acceleration of the FPGA coprocessor (802) adopts AES-256, and the throughput is 10Gbps; the interval of the master-slave processor heartbeat detection is 100ms, the state synchronization period is 1s, and the fault switching time is <200ms.

[0051] In a specific embodiment, the Ethernet (10) of the system adopts the Time Sensitive Networking (TSN) protocol to ensure: clock synchronization accuracy <1μs, data transmission delay <5ms, and bandwidth reservation guarantee mechanism. The security architecture of the system includes a network firewall, a hardware TrustZone isolation zone, a biometric verification zone, and an encrypted medical data core zone; wherein, the hardware TrustZone isolation zone performs signature verification (ECDSA algorithm) on the firmware (FPGA coprocessor, integrated information input module, multimodal identification printing module and multi-protocol communication module, etc.); the biometric verification zone uses fingerprint + voiceprint to authenticate the medical staff to enhance security; the encrypted medical data core zone encrypts and decrypts medical data based on encryption and decryption algorithms. The system also includes a remote maintenance interface for wireless firmware upgrades (differential update package verification), diagnostic data export (AES-256 encryption) and virtual private network access (IPsec VPN). In addition, the system is also equipped with a power management system, equipped with dual AC / DC inputs (90-264VAC) and lithium battery backup (battery life ≥8 hours), and uses dynamic load balancing technology for power management.

[0052] In the implementation of an infant incubator monitoring and management system in the above embodiment, the logical division of each program module is only an example. In actual application, the above functions can be assigned to different program modules as needed, for example, for the configuration requirements of the corresponding hardware or the convenience of software implementation. That is, the internal structure of the infant incubator monitoring and management system is divided into different program modules to complete all or part of the functions described above.

[0053] Example 2: This example proposes a method for monitoring and managing an infant incubator based on the infant incubator monitoring and management system in the above example, including the following steps: System startup self-test: Medical staff starts the infant incubator monitoring and management system to perform an initial self-test, including checking whether each hardware module is working properly (such as sensor connection status, display function, communication module signal, etc.), and loading the preset system parameters and configurations.

[0054] Input and association of child information: Medical staff enter the child information (such as name, gender, date of birth, medical record number, etc.) on the touch screen and scan the mother's wristband through the NFC reader to obtain the guardian information; the medical staff's fingerprint is collected through the fingerprint recognition unit, and the data verification unit verifies the medical staff's operating authority and child information, and associates the child and guardian information to form an electronic file for the child.

[0055] Multimodal identification printing: Based on the patient information and guardian information, the multimodal identification printing module is used to generate an RFID tag containing the patient's ID and a monitoring QR code; the RFID tag is written into the RFID wristband worn on the patient's ankle, and the monitoring QR code is attached to the outside of the infant incubator.

[0056] Environmental and physiological data monitoring: Place the child in the incubator and start the composite data acquisition module. The environmental data in the infant incubator are collected in real time through the temperature and humidity sensors, oxygen concentration detector and micro-air pressure monitor in the environmental monitoring unit, and transmitted to the central control module through the CAN bus and Ethernet, and stored in the distributed data storage module; at the same time, the environmental conditions in the incubator are automatically adjusted according to the preset environmental parameter range through the abnormal warning unit, such as by controlling the heater, humidifier, ventilation fan and oxygen concentration regulating valve and other equipment to maintain a suitable culture environment; the physiological data of the child are collected in real time through the non-contact infrared thermometer, multi-wavelength blood oxygen probe, piezoelectric respiratory monitoring pad and intelligent weighing system in the physiological parameter acquisition unit, transmitted to the central control module and stored in the distributed data storage module.

[0057] Data Analysis and Risk Assessment: The collected environmental and physiological data is cleaned by the data cleaning unit within the intelligent data processing unit to remove noise and outliers. The feature extraction unit then extracts data features, providing a foundation for subsequent analysis. The anomaly detection unit then uses the isolation forest algorithm to establish a dynamic baseline model for anomaly detection and risk calculation. Based on the risk assessment results, the system adjusts monitoring and management strategies, such as increasing data collection frequency and strengthening environmental regulation, to ensure the safety and health of children.

[0058] Intelligent display and interaction: The OLED curved screen and holographic projection device in the augmented reality display module display the child's physiological parameters, environmental parameters, and system status information in real time. At the same time, medical staff can interact with the system through gestures, voice, or touch to view detailed data, adjust system parameters, or send operational instructions. Multi-level early warning prompts: The alarm prompt system uses a multi-level alarm mechanism to issue visual, auditory and tactile alarms based on the comprehensive risk score, environmental abnormality probability value and physiological parameter deviation index output by the abnormality detection unit to ensure that medical staff can respond in time and take appropriate medical measures.

[0059] It should be noted that, for the sake of convenience, the aforementioned method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present invention is not limited to the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously.

[0060] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be considered as equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. A baby incubator monitoring and management system, characterized in that: include: An integrated information input module, including a touch screen display, NFC reader, fingerprint recognition unit, voice input component, and data verification unit, is used to collect and verify patient information and medical staff identity verification; The multimodal identification printing module is equipped with an RFID writing device, a QR code generator, and a thermal printer, and is used to generate a monitoring QR code and an RFID tag containing the patient's ID; the RFID tag is written into an RFID wristband worn on the patient's ankle; The composite data acquisition module consists of an environmental monitoring unit, a physiological parameter acquisition unit, a configuration self-calibrator, and an abnormality warning unit. The environmental monitoring unit integrates a temperature and humidity sensor, an oxygen concentration detector, and a micro-air pressure monitor to collect environmental data. The environmental data includes temperature and humidity, oxygen concentration, and micro-air pressure. The physiological parameter acquisition unit includes a non-contact infrared thermometer, a multi-wavelength blood oxygen probe, a piezoelectric respiratory monitoring pad and an intelligent weighing system for collecting physiological data; The physiological parameters include body temperature, blood oxygen saturation, expiratory frequency and weight; the self-calibrator performs zero point calibration and span calibration on the environmental monitoring unit and the physiological parameter acquisition unit at each set calibration time; the abnormal warning unit is used to establish a dynamic threshold model to monitor and warn environmental data and physiological data; The distributed data storage module uses blockchain encryption storage technology and a time series database dual architecture to perform hierarchical storage and data integrity verification on the data collected by the system; Intelligent data processing module, including data cleaning unit, feature extraction unit and anomaly detection unit; Multi-protocol communication module, supporting 5G, Wi-Fi6 and LoRa three-mode transmission, used to realize data transmission and interaction in the system; Augmented reality display module, equipped with OLED curved screen, holographic projection device and alarm prompt system; and central control module, using a heterogeneous computing architecture with dual-core high-performance application processors and FPGA coprocessors; The integrated information input module, multimodal identification printing module, composite data acquisition module, distributed data storage module, intelligent data processing module, multi-protocol communication module, augmented reality module and central control module are all interconnected through CAN bus and Ethernet dual channels to form a closed-loop control system.

2. The infant incubator monitoring and management system according to claim 1, characterized in that: The touch screen supports electronic signature entry and handwriting recognition functions for entering patient information and medical staff identity information; The NFC reader is compatible with the ISO / IEC 14443 Type A / B standard; the NFC reader is used to identify the medical staff's identity information in the NFC badge or wristband worn by the medical staff and determine their operation permissions; It is also used to associate the patient information with the guardian information and interact with the hospital information system to achieve data synchronization; the fingerprint recognition unit is used to collect the patient's foot prints and the medical staff's fingerprints to identify the patient's identity and verify the medical staff's identity and authority management; The voice input component integrates a noise cancellation algorithm and a dialect recognition model to collect voice commands from medical staff and the crying sounds of infants; the voice commands are used to control the operation of the infant incubator; and the crying sounds of infants are used to determine the needs or physical condition of the infants. The data verification unit implements a three-level verification mechanism, including: a) Format verification: Use regular expressions to verify the text format of medical identity information, patient information, and medical data; b) Logical verification: Verify the medical staff identity information, authority information and patient information based on the hospital information system; c) Biometric verification: Based on the fingerprint recognition unit and voice input component, the patient's footprints, medical staff's fingerprints, the patient's crying, and the medical staff's voice commands are collected, and dual biometric authentication is performed through feature extraction and feature comparison.

3. The infant incubator monitoring and management system according to claim 1, characterized in that: The temperature and humidity sensor uses a redundant design of a main probe and a slave probe to collect the temperature and humidity inside the infant incubator; the main probe is located at the top of the infant incubator; the slave probe is placed under the mattress of the infant incubator; The oxygen concentration detector is based on tunable diode laser absorption spectroscopy technology and uses a wavelength modulation spectroscopy algorithm to eliminate water vapor interference and collect oxygen concentration in the infant incubator; The micro-pressure monitor is used to monitor the pressure difference between the inside and outside of the infant incubator in real time, and to adjust the opening of the air inlet valve of the infant incubator in conjunction with the PID control algorithm to maintain the standard pressure difference; The dynamic threshold model is expressed as: threshold = reference value ± 3σ, where the reference value refers to the environmental data monitored by each component in the environmental monitoring unit under normal physiological conditions or stable environmental conditions, and σ is the standard deviation of the environmental data monitored by each component in the environmental monitoring unit within the previous set sampling time.

4. The infant incubator monitoring and management system according to claim 1, characterized in that: The non-contact infrared thermometer is used to collect the patient's body temperature in real time; The multi-wavelength blood oxygen sensor is equipped with a three-wavelength LED and uses an improved Beer-Lambert algorithm to calculate blood oxygen saturation; The piezoelectric respiratory monitoring pad integrates a PVDF film sensor and combines a wavelet packet decomposition algorithm to extract respiratory waveforms and detect the respiratory rate of children; The intelligent weighing system uses the electromagnetic force compensation principle to collect the weight of the child; the electromagnetic force compensation principle is used to collect the weight of the child, specifically: An accelerometer-assisted motion compensation algorithm is used to monitor the patient's movement status in real time. By measuring the acceleration information of various parts of the patient's body, the impact of movement on weight measurement is analyzed, and the original weight data is compensated based on this information. Adaptive filtering technology is used to denoise the corrected weight data and filter out fluctuations and abnormal values ​​in the corrected weight data.

5. The infant incubator monitoring and management system according to claim 1, characterized in that: The blockchain encryption storage technology adopts the Hyperledger Fabric framework to set up a dual-chain structure of medical data chain and equipment log chain; The time series database adopts the InfluxDB architecture; The data hierarchical storage is specifically as follows: Physiological data and environmental data are classified into real-time data, short-term data, and long-term data based on the characteristics and uses of the data, and are graded based on the importance and sensitivity of the data; For real-time data, a first retention time is set and stored in a volatile memory; For short-term data, a second retention time is set and the data is stored in a non-volatile memory; The second retention time is much greater than the first retention time; For long-term data, it is encrypted and uploaded to private cloud storage; The SHA-256 hash tree structure is used to verify the data integrity of real-time data, short-term data, and long-term data, and the refresh time is set to generate the Merkle Root verification value.

6. The infant incubator monitoring and management system according to claim 1, characterized in that: The data cleaning unit uses a wavelet transform denoising algorithm to denoise the physiological data and environmental data, and uses a sliding window outlier detection algorithm to remove outliers; The feature extraction unit uses a deep convolutional neural network to extract features of the post-cleaning environmental data and physiological data; the features include time domain features, frequency domain features, and time-frequency domain features; the time domain features include mean, variance, and approximate entropy; the frequency domain features include power spectrum centroid and harmonic ratio; the time-frequency domain features include wavelet energy entropy; The anomaly detection unit is used to establish a dynamic baseline model based on the isolation forest algorithm, perform anomaly detection and risk calculation on features, and output an environmental anomaly probability value, a physiological parameter deviation index, and a comprehensive risk score.

7. The infant incubator monitoring and management system according to claim 6, characterized in that: The OLED curved screen supports multi-window display, gesture control and brightness self-adjustment; the holographic projection device generates a three-dimensional physiological parameter spherical model, a dynamic heat map of environmental parameters and a spatiotemporal distribution map of historical data based on physiological data and environmental data; The alarm prompt system is based on a multi-level alarm mechanism and performs visual alarm, auditory alarm and tactile alarm according to the environmental abnormality probability value, physiological parameter deviation index and comprehensive risk score output by the abnormality detection unit; The hierarchical alarm mechanism is specifically as follows: The alarm levels are divided according to the comprehensive risk score, the probability value of environmental abnormality and the physiological parameter deviation index; the alarm levels include level 1 alarm, level 2 alarm and level 3 alarm; When the comprehensive risk score is greater than or equal to the first set score and less than the second set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the first set threshold but does not exceed the second set threshold, the alarm level is judged to be a first-level alarm. At this time, the visual alarm is a flashing breathing light of the first color, the auditory alarm is a beeping sound of the first set volume, and the tactile alarm is a vibration prompt of the first amplitude; When the comprehensive risk score is greater than or equal to the second set score and less than the third set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the second set threshold but does not exceed the third set threshold, the alarm level is judged to be a second alarm. At this time, the visual alarm is a second color breathing light rotating warning, the auditory alarm is a second set volume intermittent sound prompt, and the tactile alarm is a second amplitude vibration prompt; When the comprehensive risk score is greater than or equal to the third set score, or the environmental abnormality probability value / physiological parameter deviation index exceeds the third set threshold, the alarm level is judged to be a third-level alarm. At this time, the visual alarm is a flashing third-color strong light, the auditory alarm is a continuous alarm sound with a third set volume, and the tactile alarm is a vibration prompt with a third amplitude. At the same time, the emergency power supply is activated and a priority call is pushed to the nurse station. The first set volume < the second set volume < the third set volume, and the first amplitude < the second amplitude < the third amplitude.

8. The infant incubator monitoring and management system according to claim 1, characterized in that: The dual-core high-performance application processor runs a real-time operating system for performing task scheduling using a real-time scheduling algorithm, as well as device driver management and security authentication control; The FPGA coprocessor is used to form a data preprocessing pipeline for the monitoring data of the composite data acquisition module, accelerate the blockchain encryption storage technology in the distributed data storage module, and optimize the display rendering of the augmented reality display module; The central control module adopts a dual-machine hot standby mechanism, in which a dual-core high-performance application processor serves as the main processor and an FPGA coprocessor serves as the slave processor; the dual-machine hot standby mechanism performs heartbeat detection, state synchronization, and fault switching on the main and slave processors; The central control module implements DVFS technology to perform dynamic power consumption management.

9. The infant incubator monitoring and management system according to claim 1, characterized in that: The Ethernet of the system adopts the time-sensitive network protocol; The system's security architecture consists of a network firewall, a hardware TrustZone isolation zone, a biometric verification zone, and an encrypted medical data core zone. The hardware TrustZone isolation zone performs signature verification on the firmware. The firmware includes an FPGA coprocessor, an integrated information input module, a multimodal identification printing module, and a multi-protocol communication module. The biometric verification zone uses fingerprints and voiceprints to authenticate medical personnel. The encrypted medical data core zone encrypts and decrypts medical data based on encryption and decryption algorithms. The system also includes a remote maintenance interface for implementing wireless firmware upgrades, diagnostic data export, and virtual private network access.

10. A method for monitoring and managing an infant incubator, characterized in that: The method for monitoring and managing an infant incubator according to any one of claims 1 to 9 comprises the following steps: System startup self-test: Medical staff starts the infant incubator monitoring and management system to perform initialization self-test; Patient information entry and association: Medical staff enter the patient's information on the touch screen and scan the mother's wristband with an NFC reader to obtain the guardian's information. The fingerprint recognition unit collects the medical staff's fingerprint, and the data verification unit verifies the medical staff's operating authority and the patient's information. The patient and guardian information are then linked to form the patient's electronic file. Multimodal identification printing: Based on the patient and guardian information, the multimodal identification printing module generates an RFID tag containing the patient's ID and a monitoring QR code. The RFID tag is written into an RFID wristband worn on the patient's ankle, and the monitoring QR code is affixed to the outside of the infant incubator. Environmental and physiological data monitoring: The infant is placed in the incubator and the composite data acquisition module is activated. The temperature and humidity sensors, oxygen concentration detector, and micro-air pressure monitor in the environmental monitoring unit collect real-time environmental data inside the infant incubator. The data is then transmitted to the central control module via the CAN bus and Ethernet and stored in the distributed data storage module. Simultaneously, the abnormal warning unit automatically adjusts the environmental conditions inside the incubator according to the preset environmental parameter range. The physiological parameter acquisition unit collects the child's physiological data in real time through the non-contact infrared thermometer, multi-wavelength blood oxygen probe, piezoelectric respiratory monitoring pad and intelligent weighing system, transmits it to the central control module and stores it in the distributed data storage module; Data analysis and risk assessment: The collected environmental data and physiological data are cleaned by the data cleaning unit in the intelligent data processing unit, and the data features are extracted by the feature extraction unit; The anomaly detection unit is then used to establish a dynamic baseline model based on the isolation forest algorithm to perform anomaly detection and risk calculation on environmental and physiological data; intelligent display and interaction: the OLED curved screen and holographic projection device in the augmented reality display module display the child's physiological parameters, environmental parameters and system status information in real time; at the same time, medical staff interact with the system through gestures, voice or touch to view detailed data, adjust system parameters or send operation instructions; multi-level early warning prompts: the alarm prompt system uses a multi-level alarm mechanism to provide visual, auditory and tactile alarms based on the comprehensive risk score, environmental anomaly probability value and physiological parameter deviation index output by the anomaly detection unit.

Images