Hand hygiene monitoring management system based on Internet of Things

Abstract

The invention discloses a hand hygiene monitoring management system based on the Internet of Things, which comprises an identity recognition module, a hand washing action monitoring module, a data acquisition and transmission module, a data storage module, a data analysis module, a reminding feedback module and the like, and is also provided with an equipment state monitoring module. The identity recognition module acquires personnel information through RFID tag reading and facial image acquisition; the hand washing action monitoring module collects hand washing process data by using various sensors; after data are collected and transmitted, the storage module stores the data in a classified mode, and the analysis module processes the data through algorithms such as a convolutional neural network and judges the hand washing compliance condition. And the reminding feedback module is used for reminding in an acousto-optic mode and presenting a result to a manager. The waterproof and dustproof performance of system hardware reaches the IP65 standard, the antibacterial rate of the shell is not lower than 99, and the normal operation of the hardware is guaranteed by the equipment state monitoring module. The system not only improves the efficiency and accuracy of hand hygiene management of medical staff, but also can effectively prevent disease transmission in a medical environment and ensure public health safety.

Description

Technical Field

[0001] This invention relates to the field of hygiene monitoring, and in particular to a hand hygiene monitoring and management system based on the Internet of Things. Background Technology

[0002] With the ongoing challenges of global public health events, especially the spread of the pandemic, hand hygiene, as a fundamental measure to prevent disease transmission, has become increasingly important in healthcare institutions. Hospitals and other medical settings face high risks of cross-infection, and the hand hygiene practices of healthcare workers directly impact patient safety and the overall hygiene level of the hospital. However, due to busy schedules, lack of supervision, and personal negligence among healthcare workers, handwashing practices are often not standardized or regulated, resulting in poor compliance. This situation not only increases the risk of hospital-acquired infections but also poses significant management challenges. Therefore, a hand hygiene monitoring and management system based on Internet of Things (IoT) technology has emerged, aiming to improve the hand hygiene compliance rate of healthcare workers through automated, real-time hand hygiene monitoring, thereby reducing the risk of infection within hospitals and enhancing public health safety.

[0003] Existing hand hygiene management technologies largely rely on manual inspection and traditional manual recording, which has significant drawbacks. First, manual monitoring is susceptible to subjective factors, making it difficult to ensure continuous, all-around supervision and resulting in insufficient monitoring coverage. Second, traditional methods also suffer from significant problems in the real-time nature and accuracy of data; hand hygiene data is often recorded retrospectively, lacking real-time feedback, making it impossible to promptly identify and intervene in non-standard handwashing behaviors. Furthermore, traditional hand hygiene management technologies provide rather crude assessments of handwashing techniques, failing to accurately analyze whether healthcare workers' handwashing processes conform to standard operating procedures, thus unable to provide precise improvement suggestions and data support. Overall, existing technologies cannot meet the needs for efficient and comprehensive hand hygiene monitoring. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a hand hygiene monitoring and management system based on the Internet of Things.

[0005] An IoT-based hand hygiene monitoring and management system includes: an identity recognition module, a handwashing action monitoring module, a data acquisition and transmission module, a data storage module, a data analysis module, and an alert and feedback module.

[0006] The identity recognition module is connected to the handwashing action monitoring module and the data acquisition and transmission module, respectively. It includes an RFID tag reading unit and a face recognition camera unit. The RFID tag reading unit is used to read the RFID tag information worn by medical staff to obtain the medical staff's identity. The face recognition camera unit captures facial images of medical staff entering the handwashing area and transmits the captured facial image information to the data acquisition and transmission module.

[0007] The handwashing action monitoring module is connected to the identity recognition module and the data acquisition and transmission module, and the internal sensors work together. The module is equipped with a hand sanitizer usage sensor, a faucet switch sensor, and a hand motion sensor. The hand sanitizer usage sensor is used to monitor whether the hand sanitizer has been squeezed out; the faucet switch sensor detects the open and closed state of the faucet; and the hand motion sensor monitors the actions of medical staff rubbing their hands and rinsing them. Each sensor transmits the monitoring data to the data acquisition and transmission module in real time.

[0008] Furthermore, the data acquisition and transmission module is connected to the identity recognition module, the handwashing action monitoring module, the data storage module, and the data analysis module. It is responsible for collecting the personnel identity information obtained by the identity recognition module and the handwashing process data collected by the handwashing action monitoring module. The data is packaged and transmitted to the data storage module and the data analysis module through Internet of Things (IoT) communication methods, including Wi-Fi, Bluetooth, or ZigBee.

[0009] The data storage module is connected to the data acquisition and transmission module and the data analysis module. It adopts a database storage method to classify and store the collected personnel identity information, handwashing time and handwashing action data, providing data support for the data analysis module.

[0010] The data analysis module is connected to the data acquisition and transmission module, the data storage module, and the reminder feedback module, respectively. It uses a preset analysis algorithm to process the stored data, including counting handwashing frequency, analyzing the compliance of handwashing steps, and transmitting the analysis results to the reminder feedback module.

[0011] The reminder feedback module is connected to the data analysis module. When the data analysis module determines that the medical staff's handwashing behavior is not compliant, it will issue a reminder signal through an audible and visual alarm. At the same time, the handwashing records and analysis results will be presented to the management personnel in the form of a visual interface for review and processing.

[0012] The data analysis module uses a convolutional neural network deep learning model to analyze handwashing action data. The model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. The input layer receives hand motion sensor data from the handwashing action monitoring module. The data dimension is [batch_size, sequence_length, feature_dim], where batch_size represents the number of samples in each input data, sequence_length represents the length of the action sequence, and feature_dim represents the dimension of each action feature.

[0013] Convolutional layers use A convolution kernel of size 1, stride 1, padding 1, is calculated using the following formula: ,in, It is the first Convolutional layers at position The output, It is the first The convolutional kernel weights of the layer, It is the first Layer in position Input, It is the first Layer bias;

[0014] Pooling layer adopts The maximum pooling size, with a step size of 2, is calculated using the following formula: For the l-th pooling layer at position The output;

[0015] The weight matrix of the fully connected layer is , bias is The calculation formula is: It is the output of the fully connected layer. It is the feature vector after processing by the pooling layer;

[0016] The output layer uses the Softmax function to convert the output of the fully connected layer into the probability of handwashing action classification. The formula for the Softmax function is: , It is the output vector of the fully connected layer. It represents the number of handwashing action categories.

[0017] Furthermore, the convolutional neural network model employs a stochastic gradient descent optimization algorithm during training, with a learning rate set to 0.001, a momentum parameter set to 0.9, and a cross-tab loss function used, the formula of which is: ,in, It is the sample size. It is the number of handwashing action categories. It is a sample The true label, The model predicts the sample. Category The probability is determined by validation, which is performed every 50 training batches during training. The validation set is a proportion of the total dataset. The total number of batches for model training is set to 300.

[0018] Furthermore, the RFID tag uses an ultra-high frequency band with a frequency range of 860-960MHz and a transmission power of no more than 2W. The RFID tag has a unique identification code with a code length of 64 bits, storing the medical staff's employee number, name, and department identity information.

[0019] Furthermore, the hand sanitizer uses an infrared sensing sensor with a detection distance range of 5-15 cm and a response time of no more than 0.1 seconds. When the liquid level in the hand sanitizer bottle is detected to be lower than a preset threshold, the sensor sends a low liquid level alarm signal to the data acquisition and transmission module. The preset threshold can be flexibly configured according to the capacity of the hand sanitizer bottle and the normal usage cycle, and the configuration range is 10%-20% of the total capacity in the bottle.

[0020] Furthermore, the data storage module adopts a distributed storage architecture, consisting of multiple storage nodes connected by a high-speed network. It uses a consistent hashing algorithm for data sharding and storage, distributing the data evenly across the storage nodes according to the hash value. This improves the scalability and fault tolerance of the storage system. When a storage node fails, the system automatically migrates the data to other normal nodes.

[0021] Furthermore, the audible and visual alarm in the reminder feedback module uses a combination of flashing lights and sounds of different frequencies to provide reminders. When non-compliance with handwashing procedures is detected, the light flashes 3 times per second and the sound frequency is 1000Hz; when the hand sanitizer level is low, the light flashes 2 times per second and the sound frequency is 800Hz. The volume of the audible and visual alarm is adaptively adjusted according to the noise level in the hospital environment, with an adjustment range of 40-80 decibels.

[0022] Furthermore, the hand movement sensor in the handwashing action monitoring module employs an inertial measurement unit, integrating an accelerometer and a gyroscope. The accelerometer's measurement range is... The gyroscope's measurement range is The sampling frequency is 100Hz. By fusing accelerometer and gyroscope data, a complementary filtering algorithm is used to calculate the posture angle of hand movements for monitoring hand rubbing and rinsing actions. The formula for the complementary filtering algorithm is: ,in, It is the first Attitude angle at any moment These are the filter coefficients, and their values ​​range from [value range missing]. It is the attitude angle measured by the accelerometer. It is the attitude angle measured by the gyroscope.

[0023] Furthermore, the system also includes a device status monitoring module, which is connected to the identity recognition module, handwashing action monitoring module, data acquisition and transmission module, data storage module, data analysis module, and reminder feedback module. The device status monitoring module monitors the operating status of each hardware device in real time, including power level, network connection status, and sensor working status. When a device malfunction is detected, it sends a device malfunction alarm information to the reminder feedback module and records the malfunction time and malfunction type information to the data storage module.

[0024] Furthermore, the hardware components of the system are all designed to be waterproof and dustproof, with an IP65 protection rating. The equipment casing is made of antibacterial materials with an antibacterial rate of no less than 99%, which is used to prevent bacterial growth.

[0025] Beneficial effects:

[0026] This invention proposes an IoT-based hand hygiene monitoring and management system. By combining multiple modules such as identity recognition, handwashing action monitoring, and data collection and transmission, it provides a comprehensive and efficient hand hygiene management solution. The identity recognition module uses RFID tags and facial recognition technology to ensure accurate identification of healthcare personnel, ensuring that each handwashing action can be associated with a specific individual. The handwashing action monitoring module uses multiple sensors, such as hand sanitizer usage sensors, faucet switch sensors, and hand motion sensors, to comprehensively monitor the entire handwashing process, collecting data in real time and transmitting it to the data storage and analysis modules. The data analysis module uses a convolutional neural network model to perform deep learning on hand movements, determining the compliance of handwashing behavior and providing timely feedback to management personnel. This system can monitor the handwashing behavior of healthcare personnel in real time, ensuring that hand hygiene standards are followed in hospitals or other medical environments, reducing the risk of cross-infection. Based on IoT technology, all data collection and transmission are efficient and automated, reducing the pressure and errors of manual monitoring. Furthermore, the system's alert and feedback function ensures immediate warnings when handwashing behavior is non-compliant, effectively improving handwashing compliance, and presents real-time data to management personnel through a visual interface, helping them make timely decisions. The equipment status monitoring module further enhances system reliability, ensuring that all hardware devices operate in optimal condition and preventing monitoring results from being affected by equipment failure. In summary, this system not only improves the efficiency and accuracy of hand hygiene management for healthcare workers but also effectively prevents the spread of diseases in the medical environment, ensuring public health safety. Attached Figure Description

[0027] Figure 1 This is a diagram showing the system module composition of the present invention;

[0028] Figure 2 This is a flowchart of the system operation of the present invention. Detailed Implementation

[0029] It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. The following describes the application in further detail with reference to the accompanying drawings and specific embodiments.

[0030] like Figure 1 As shown, the IoT-based hand hygiene monitoring and management system includes: an identity recognition module, a handwashing action monitoring module, a data acquisition and transmission module, a data storage module, a data analysis module, and an alert and feedback module.

[0031] Identity Recognition Module: This module is connected to the handwashing action monitoring module and the data acquisition and transmission module, respectively. It includes an RFID tag reading unit and a face recognition camera unit. The RFID tag reading unit is used to read the RFID tag information worn by medical staff to obtain the medical staff's identity. The face recognition camera unit captures facial images of medical staff entering the handwashing area and transmits the captured facial image information to the data acquisition and transmission module.

[0032] Specifically, the identification module is a core component of the entire hand hygiene monitoring and management system. Its main function is to identify the healthcare workers entering the handwashing area and it works in conjunction with the handwashing action monitoring module and the data acquisition and transmission module. This module combines two technologies: an RFID tag reader and a facial recognition camera unit. The RFID tag reader obtains a unique identifier by reading the RFID tags worn by healthcare workers. The RFID tags use the ultra-high frequency band, with a frequency range of 860-960MHz, ensuring fast and accurate information reading over short distances. Each RFID tag has a unique 64-bit identification code used to store the healthcare worker's employee number, name, and other identification information. The facial recognition camera unit captures facial images of healthcare workers entering the handwashing area and transmits the image information to the data acquisition and transmission module. Through these two identification technologies, the system effectively ensures that each handwashing action is linked to a specific healthcare worker, facilitating future tracking and management. In application scenarios, this module is particularly important in hospitals, clinics, and public health sectors, ensuring the compliance of handwashing practices and avoiding management loopholes caused by incorrect identification or monitoring omissions.

[0033] Handwashing Action Monitoring Module: Connected to the identity recognition module and data acquisition and transmission module, and with all internal sensors working together, this module is equipped with a hand sanitizer usage sensor, a faucet switch sensor, and a hand motion sensor. The hand sanitizer usage sensor monitors whether hand sanitizer has been dispensed; the faucet switch sensor detects the on / off state of the faucet; and the hand motion sensor monitors the actions of medical staff rubbing and rinsing their hands. All sensors transmit the monitoring data to the data acquisition and transmission module in real time.

[0034] Specifically, the handwashing action monitoring module is a key component in monitoring hand hygiene behavior. This module integrates multiple sensors to monitor healthcare workers' handwashing process in real time, including a sanitary sanitizer usage sensor, a faucet switch sensor, and a hand motion sensor. The sanitary sanitizer usage sensor uses infrared sensing technology to detect whether sanitary sanitizer has been dispensed, ensuring that healthcare workers have used enough sanitizer for hand disinfection. The faucet switch sensor monitors the faucet's on / off status, ensuring that the faucet is turned on or off at the appropriate time to avoid wasting water. The hand motion sensor uses an inertial measurement unit (IMU) to detect healthcare workers' hand movements, including rubbing hands and rinsing, monitoring whether they conform to the correct handwashing steps. The hand motion sensor combines accelerometers and gyroscopes, fusing the data through a complementary filtering algorithm to accurately calculate the hand posture angle, ensuring that every movement is accurately captured and recorded. All monitoring data is transmitted in real time to the data acquisition and transmission module, providing a basis for subsequent data analysis and alerts. This module is particularly crucial in the medical environment, effectively monitoring handwashing procedures and reducing the occurrence of cross-infection within hospitals.

[0035] Data acquisition and transmission module: Connected to the identity recognition module, handwashing action monitoring module, data storage module, and data analysis module, it is responsible for collecting personnel identity information obtained by the identity recognition module and handwashing process data collected by the handwashing action monitoring module, and transmitting the data to the data storage module and data analysis module through Internet of Things communication technologies such as Wi-Fi, Bluetooth, or ZigBee.

[0036] Specifically, the data acquisition and transmission module acts as a bridge in the entire system, responsible for collecting data from the identification module and the handwashing action monitoring module, and transmitting the data to the data storage and data analysis modules via IoT communication technology. This module can use IoT communication methods such as Wi-Fi, Bluetooth, and ZigBee to package and transmit data, ensuring fast and stable data flow. The module first obtains the healthcare worker's identification information from the identification module and various data during the handwashing process (such as hand sanitizer usage, hand movements, and time) from the handwashing action monitoring module. All collected data is packaged in real time and sent to the data storage and analysis module via IoT technology according to a preset communication protocol. During transmission, to ensure data security and accuracy, the data may be protected using encryption algorithms to prevent data leakage or tampering. This module is widely used in hospitals, clinics, and public health facilities, ensuring efficient and accurate transmission of various hand hygiene data and providing timely data support for management personnel.

[0037] Data storage module: Connected to the data acquisition and transmission module and the data analysis module. Employing database storage technologies such as MySQL and Oracle, it categorizes and stores collected personnel identification information, handwashing time, and handwashing action data, providing data support for the data analysis module.

[0038] Specifically, the data storage module is responsible for storing all data collected by the data acquisition and transmission module, including medical staff's identity information, handwashing time, and handwashing action data. This module employs database storage technology, such as MySQL or Oracle, using an efficient storage structure to categorize and store the data. Database storage ensures fast access to large amounts of data under high concurrency while guaranteeing data integrity and consistency. Another important function of the data storage module is to provide stable data support, ensuring that the data analysis module can make analytical judgments based on comprehensive and accurate data. Handwashing records stored in the database can be queried and analyzed according to different conditions, helping hospital administrators to understand the handwashing behavior of medical staff in a timely manner. Furthermore, to improve the scalability and fault tolerance of the storage system, the data storage module adopts a distributed storage architecture. Multiple storage nodes use a consistent hashing algorithm for data sharding. When a storage node fails, the system can automatically migrate the data to other nodes, ensuring high data availability and system stability.

[0039] Data Analysis Module: Connects to the data acquisition and transmission module, data storage module, and reminder / feedback module. Using preset analysis algorithms, it processes the stored data, including calculating handwashing frequency and analyzing compliance with handwashing procedures, and transmits the analysis results to the reminder / feedback module.

[0040] Specifically, the data analysis module is responsible for processing and analyzing the large amount of data collected in the storage module. Using pre-defined analysis algorithms, it assesses the compliance of handwashing practices, tracks handwashing frequency, and determines the correctness of handwashing steps. This module primarily relies on deep learning analysis using convolutional neural networks (CNNs). By analyzing hand movement data, CNNs can identify each step in the handwashing process and determine whether it conforms to standard handwashing procedures. For example, it assesses whether the rubbing and rinsing movements are sufficient. This module can also track the handwashing frequency of healthcare workers, evaluate compliance over a period of time, and transmit the analysis results to the alert and feedback module for management reference. The application of the data analysis module not only helps improve the hand hygiene compliance of healthcare workers but also optimizes hospital hand hygiene management strategies through big data analysis, further improving the overall hygiene management level of the hospital. In practical applications, this module can provide real-time behavioral feedback in the medical environment, reducing oversights and errors in hand hygiene implementation.

[0041] The alert and feedback module is connected to the data analysis module. When the data analysis module determines that medical staff's handwashing behavior is not compliant, it will issue an alert signal through an audible and visual alarm; at the same time, the handwashing records and analysis results will be presented to the management personnel in a visual interface for easy viewing and processing.

[0042] Specifically, the reminder and feedback module is a crucial part of the system, used to promptly remind medical staff and administrators. This module connects to the data analysis module. When the analysis module determines that a medical staff member's handwashing behavior is non-compliant, the reminder and feedback module issues a warning signal via an audible and visual alarm to remind the staff member to immediately correct their handwashing procedure. The audible and visual alarm uses a combination of flashing lights and sounds at different frequencies to effectively attract the attention of medical staff and indicate whether their handwashing steps are correct. For example, when the handwashing steps are non-compliant, the lights flash three times per second, and the sound frequency is 1000Hz; while when the hand sanitizer level is low, the light and sound frequencies change. The reminder and feedback module not only guides medical staff to correct non-compliant behavior through alarms but also presents handwashing records and analysis results to administrators in a visual interface, facilitating supervision and management. The application of this module is crucial for improving hand hygiene implementation and optimizing hospital management, especially in large-scale medical facilities, where it effectively ensures public health safety.

[0043] Preferably, the data analysis module incorporates a Convolutional Neural Network (CNN) deep learning model to analyze handwashing action data. This model includes an input layer, convolutional layers, pooling layers, fully connected layers, and an output layer. The input layer receives hand motion sensor data from the handwashing action monitoring module, with data dimensions of [batch_size, sequence_length, feature_dim], where batch_size represents the number of samples in each input, sequence_length represents the length of the action sequence, and feature_dim represents the dimension of each action feature. The convolutional layers employ... A convolution kernel of size 1, stride 1, padding 1, is calculated using the following formula: ,in, It is the first Convolutional layers at position The output, It is the first The convolutional kernel weights of the layer, It is the first Layer in position Input, It is the first Layer bias. Pooling layers use... The maximum pooling size, with a step size of 2, is calculated using the following formula: For the l-th pooling layer at position The output of the fully connected layer is... , bias is The calculation formula is: It is the output of the fully connected layer. This is the feature vector after pooling. The output layer uses the Softmax function to convert the output of the fully connected layer into the probability of handwashing action classification. The formula for the Softmax function is: , It is the output vector of the fully connected layer. It represents the number of handwashing action categories.

[0044] Specifically, convolutional neural network (CNN) deep learning models play a crucial role in handwashing action data analysis. From a model structure perspective, the input layer receives hand motion sensor data from the handwashing action monitoring module. The data dimensions include the number of samples, the length of the action sequence, and the dimension of each action feature. For example, in a real-world handwashing monitoring scenario, the number of samples could be the amount of handwashing action data from multiple people being monitored simultaneously; the action sequence length represents the duration or number of steps in a series of recorded actions from start to finish; and the action feature dimension covers various aspects such as hand position, angle, and speed. The convolutional layer uses a 3×3 kernel with a stride of 1 and padding of 1. It acts like an intelligent feature extractor, capturing key local features in the hand motion data, such as subtle movements like finger bending and palm rotation, through convolution operations on the input data. The pooling layer uses 2×2 max pooling with a stride of 2. Its function is to reduce the dimensionality of the data without losing key information, thereby reducing the amount of data and computation, while highlighting the main features. For example, in handwashing action data, subtle interference fluctuations are filtered out, retaining the most representative action features. The fully connected layer integrates the feature vectors processed by the pooling layer, associating different features through weight matrix and bias operations. The output layer uses the Softmax function to convert the output of the fully connected layer into handwashing action classification probabilities. In practical applications, such as restrooms in schools and hospitals, this model can determine whether people's handwashing actions are correct, thus reminding and guiding them to wash their hands properly, aiding in hygiene management, and effectively preventing the spread of disease.

[0045] Preferably, the convolutional neural network model employs the Stochastic Gradient Descent (SGD) optimization algorithm during training, with a learning rate set to 0.001 and a momentum parameter set to 0.9. The loss function used is the cross-tab loss function, with the following formula: ,in, It is the sample size. It is the number of handwashing action categories. It is a sample The true label, The model predicts the sample. Category The probability of [something]. During training, validation is performed every 50 training batches (epochs), with the validation set accounting for [a certain percentage] of the total dataset. The total number of batches for model training is set to 300.

[0046] Specifically, several key technical points are crucial when training a handwashing action recognition model using a convolutional neural network. Stochastic Gradient Descent (SGD) optimization algorithm is a powerful training tool. The learning rate is set to 0.001, controlling the size of the model's "steps" when updating parameters. A smaller learning rate leads to more stable convergence but slower training; 0.001 strikes a balance between stability and efficiency. The momentum parameter is set to 0.9, adding inertia to the model's training, allowing it to move more smoothly in the optimization direction, avoiding lingering in local optima and accelerating convergence to a globally optimal solution. The loss function used is the cross-entropy loss function, which measures the difference between the model's prediction and the true label. In the handwashing action recognition scenario, the number of samples is the number of collected handwashing action data instances, and the number of handwashing action categories is a predefined set of different action categories, such as wetting hands, applying soap, and rubbing. The true label represents the actual handwashing action category, and the model's prediction probability is the likelihood that the model classifies a particular action. By minimizing this loss function, the model continuously adjusts its parameters, improving the accuracy of recognizing handwashing actions. During training, a validation set is performed every 50 training batches, which is like giving the model a regular "check-up" to understand its performance on new data. The validation set comprises 20% of the total dataset; this data is not used in training but is specifically used to evaluate the model's generalization ability. The total number of training batches is set to 300. Through so many iterations, the model can continuously optimize, enabling it to more accurately determine whether handwashing actions are correct in real-world handwashing monitoring scenarios such as hospitals and schools, thus helping to improve hygiene and health standards.

[0047] Preferably, the RFID tag uses the ultra-high frequency (UHF) band, with a frequency range of 860-960MHz, and its transmission power does not exceed 2W. The RFID tag has a unique identification code with a length of 64 bits, which can store the medical staff's employee number, name, department, and other identification information. It also has an anti-collision mechanism to ensure accurate identification of each tag's information in scenarios where multiple tags are read simultaneously.

[0048] Specifically, RFID (Radio Frequency Identification) technology is widely used in modern management. The RFID tags mentioned here use the Ultra High Frequency (UHF) band, with a frequency range of 860-960MHz. This band has unique advantages: it can achieve a longer reading distance and has relatively good signal penetration, making it suitable for places like hospitals with high personnel flow and complex spaces. Its transmission power does not exceed 2W, ensuring effective signal transmission while also considering energy consumption and interference with surrounding equipment. Each RFID tag has a unique 64-bit identification code, just like a person's ID number. It can store crucial identity information for medical staff, such as employee ID numbers that accurately correspond to individual work numbers, facilitating internal hospital work arrangements and management; names for daily communication; and department information that clearly identifies the work affiliation of medical staff, enabling rapid determination of personnel location and responsibilities in scenarios such as resource allocation and emergency rescue. Anti-collision mechanisms are a key guarantee for the practical application of RFID tags. In a hospital, multiple medical staff may pass through a certain identification area simultaneously, such as entering the operating room or pharmacy. Without an anti-collision mechanism, when multiple tags are read simultaneously, the signals may interfere with each other, leading to incorrect or unreadable information. With this mechanism, even in complex scenarios where multiple tags appear simultaneously, the system can accurately and systematically identify the information of each tag, ensuring the accuracy and efficiency of personnel identification. This, in turn, improves the hospital's management efficiency and operational smoothness, playing a crucial role in areas such as daily attendance, access control, and medical supply requisition.

[0049] Preferably, the hand sanitizer uses an infrared sensing sensor with a detection distance range of 5-15 cm and a response time of no more than 0.1 seconds. When the liquid level in the hand sanitizer bottle is detected to be lower than a preset threshold, the sensor sends a low liquid level alarm signal to the data acquisition and transmission module. The preset threshold can be flexibly configured according to the capacity of the hand sanitizer bottle and the normal usage cycle, with a configuration range of 10%-20% of the total capacity of the bottle.

[0050] Specifically, sensors used in hand sanitizer dispensers play a crucial role in daily hygiene. These sensors employ infrared sensing technology, which utilizes infrared light to detect objects and is characterized by its stability and reliability. The detection distance is set between 5-15 centimeters, a range that effectively detects hand movements near the dispenser, avoiding both overly short detection distances and false alarms. The response time is less than 0.1 seconds, meaning that when a hand approaches, the dispenser reacts quickly, dispensing hand sanitizer rapidly, greatly improving convenience, especially in high-traffic areas where waiting time is reduced. The sensor's role is even more critical when it comes to liquid level monitoring. It constantly monitors the liquid level in the dispenser, sending a low-level alarm signal to the data acquisition and transmission module once the level falls below a preset threshold. The preset threshold can be flexibly configured based on the dispenser's capacity and normal usage cycle, ranging from 10% to 20% of the total volume. For example, in high-traffic public areas such as hospitals, schools, and shopping malls, hand sanitizer is used frequently. If a reasonable threshold is set, an alert can be issued when the remaining sanitizer is low, allowing management to replenish it promptly and preventing situations where hand sanitizer is unavailable. This ensures that people in the area have the opportunity to wash their hands and maintain a good hygienic environment. The application of this sensor not only improves the convenience of hand sanitizer use but also optimizes resource management through intelligent level monitoring and early warning functions, making the maintenance of public health facilities more efficient and orderly.

[0051] Preferably, the data storage module adopts a distributed storage architecture, consisting of multiple storage nodes connected via a high-speed network. A consistent hashing algorithm is used for data sharding, distributing data evenly across the storage nodes according to their hash values, thus improving the scalability and fault tolerance of the storage system. When a storage node fails, the system can automatically migrate data to other healthy nodes, ensuring data integrity and availability.

[0052] Specifically, the distributed storage architecture adopted by the data storage module is an effective solution for handling large-scale data storage and management needs. It consists of multiple storage nodes, which act like small data "warehouses," tightly connected by a high-speed network to ensure fast and smooth data transmission and interaction between nodes. The consistent hashing algorithm plays a crucial role. It acts like an intelligent "data allocator," distributing data evenly across the storage nodes according to hash values. In this way, data is not concentrated on a few nodes but is stored in a relatively even distribution. This greatly improves the scalability of the storage system. When more data needs to be stored, new storage nodes can be easily added, and the system can easily cope without being overwhelmed by a surge in data volume. At the same time, fault tolerance is also enhanced because the impact of a single node failure is significantly reduced due to the distributed storage of data. In practical applications, such as handwashing monitoring systems in large hospitals, a large amount of data related to medical staff's handwashing actions and hand sanitizer usage is continuously generated. If traditional centralized storage is used, it may face storage capacity bottlenecks and the risk of single points of failure. In a distributed storage architecture, even if a storage node fails unexpectedly, the system can quickly transfer the data from the failed node to other normal nodes through an automatic data migration mechanism, ensuring that the data is intact and not lost, and can still be accessed and used normally. This guarantees the stability and reliability of the entire handwashing monitoring system's data storage and management, allowing subsequent work such as analysis and management based on this data to proceed smoothly.

[0053] Preferably, the audible and visual alarm in the reminder feedback module uses a combination of flashing lights and sounds at different frequencies for reminders. When improper handwashing procedures are detected, the light flashes three times per second, and the sound frequency is 1000Hz; when the hand sanitizer level is low, the light flashes twice per second, and the sound frequency is 800Hz. The volume of the audible and visual alarm can be adaptively adjusted according to the noise level in the hospital environment, with an adjustment range of 40-80 decibels.

[0054] Preferably, the hand movement sensor in the handwashing action monitoring module is an inertial measurement unit (IMU) integrating an accelerometer and a gyroscope. The accelerometer measurement range is... The gyroscope's measurement range is The sampling frequency is 100Hz. By fusing accelerometer and gyroscope data, a complementary filtering algorithm is used to calculate the attitude angle of hand movements for monitoring hand rubbing and rinsing actions. The formula for the complementary filtering algorithm is: ,in, It is the first Attitude angle at any moment These are the filter coefficients, and their values ​​range from [value range missing]. It is the attitude angle measured by the accelerometer. It is the attitude angle measured by the gyroscope.

[0055] Specifically, the technical details of the reminder feedback module and the sound and hand movement monitoring module are crucial in the handwashing monitoring system. The reminder feedback module's sound and light alarm uses a combination of flashing lights and sounds at different frequencies to provide alerts. When improper handwashing procedures are detected, the light flashes three times per second, accompanied by a 1000Hz sound. This high-frequency flashing and sound quickly attracts attention, indicating incorrect handwashing techniques and prompting timely correction. When the hand sanitizer level is low, the light flashes twice per second, accompanied by a softer 800Hz sound, indicating the need for more sanitizer. Furthermore, the alarm volume can adaptively adjust between 40-80 decibels based on the ambient noise level in the hospital, avoiding both excessively low volume (unheard) and excessively high volume (causing noise interference), ensuring effective alerts without disrupting normal operations.

[0056] The handwashing motion monitoring module uses an inertial measurement unit (IMU) to integrate an accelerometer and a gyroscope. The accelerometer has a measurement range of ±8g, capturing the acceleration and deceleration of hand movements; the gyroscope has a measurement range of ±2000%, sensing the rotational movements of the hand. Both sensors have a sampling frequency of 100Hz, enabling rapid and accurate data acquisition. By fusing data from these two sensors, a complementary filtering algorithm is used to calculate the hand movement attitude angle. The filter coefficient is set between 0.1 and 0.3, a range that effectively balances the weighting of the accelerometer and gyroscope data, accurately calculating the attitude angle. This effectively monitors hand rubbing and rinsing movements, determining whether handwashing actions are standardized, and helping hospitals and other facilities improve the handwashing standards of medical staff and related personnel, ensuring hygiene and safety.

[0057] Preferably, the system further includes a device status monitoring module, which is connected to the identity recognition module, handwashing action monitoring module, data acquisition and transmission module, data storage module, data analysis module, and alert feedback module. The device status monitoring module monitors the operating status of each hardware device in real time, including power level, network connection status, and sensor operating status. When a device malfunction is detected, it sends a device malfunction alarm to the alert feedback module and records the malfunction time, malfunction type, and other information to the data storage module.

[0058] Preferably, all hardware components of the system are waterproof and dustproof, with a protection level reaching IP65. The equipment casing is made of antibacterial material with an antibacterial rate of not less than 99%, effectively preventing bacterial growth, adapting to the complex operating environment of hospitals, and ensuring long-term stable operation of the system.

[0059] Specifically, in handwashing monitoring systems in places such as hospitals, the characteristics of equipment status monitoring modules and hardware components are crucial.

[0060] The device status monitoring module acts like a "health guardian" for the entire system, closely connected to various modules such as identity recognition and handwashing action monitoring. It monitors the real-time operating status of each hardware device, such as power supply, to anticipate potential power shortages and prevent data loss or system failures due to sudden power outages. Network connectivity is also monitored to ensure smooth data transmission; any network interruptions can be detected and addressed promptly. Sensor status is also tracked to ensure accurate data collection, including handwashing actions. Upon detecting a device malfunction, it quickly sends an alarm to the alert feedback module and records crucial information such as the time and type of malfunction in the data storage module, facilitating rapid troubleshooting and resolution by technicians, ensuring the system's continued operation.

[0061] The system's hardware components are designed with the complex operating environment of hospitals in mind. They are waterproof and dustproof, achieving an IP65 protection rating. This means the equipment can prevent significant dust intrusion and withstand a certain level of water spray, ensuring stable operation even in the daily cleaning and humid conditions of a hospital. The equipment casing is made of antibacterial materials with an antibacterial rate of no less than 99%, effectively inhibiting bacterial growth on the equipment surface and reducing the risk of cross-infection. This not only protects the equipment itself but also meets hospital hygiene and safety requirements, ensuring long-term stable system operation and providing reliable handwashing monitoring for medical staff and patients, thus supporting the hospital's infection control efforts.

[0062] like Figure 2 As shown, the operation steps of the IoT-based hand hygiene monitoring and management system include:

[0063] Step S1: The identity verification module uses RFID tags and facial recognition camera technology to confirm the identity of medical personnel entering the handwashing area. The RFID tag reading unit scans the RFID tags worn by medical personnel to read their identity information, such as employee ID, name, and department. Simultaneously, the facial recognition camera captures facial images of the medical personnel to further ensure the accuracy of each individual's identity. The main purpose of this process is to associate each handwashing action with the specific identity of the medical personnel, ensuring that each handwashing record has a clearly defined responsible person, and providing accurate data support for subsequent data analysis and management.

[0064] Step S2: Multiple sensors work together to monitor healthcare workers' handwashing behavior. A sensor detects whether hand sanitizer has been dispensed, while a faucet switch sensor monitors the faucet's on / off status to ensure proper water usage. Hand motion sensors, combined with accelerometers and gyroscopes, monitor the rubbing and rinsing movements of healthcare workers' hands in real time. Data from these sensors is collected in real time and transmitted to a data acquisition and transmission module, providing raw data for compliance assessment of handwashing behavior.

[0065] Step S3: The main task of the data acquisition and transmission module is to integrate the data collected by the identity recognition module and the handwashing action monitoring module, and to package and transmit the data to the data storage module and the data analysis module via IoT communication methods (such as Wi-Fi, Bluetooth, or ZigBee). This module ensures the accurate transmission of data from field devices to the storage and analysis system, guaranteeing the integrity and timeliness of hand hygiene monitoring data.

[0066] Step S4: The collected data is stored and categorized using the data storage module. This module uses database storage technologies (such as MySQL and Oracle) to categorize and store medical staff's identity information, handwashing time, and handwashing action data, forming a structured data archive. This data not only supports subsequent data analysis but also provides managers with long-term data records, facilitating traceability and auditing.

[0067] Step S5: The data analysis module processes and analyzes the stored handwashing data, using pre-set analysis algorithms to assess the frequency and compliance of handwashing procedures among healthcare workers. Through this data analysis, the system can identify the compliance of handwashing behaviors, determine which healthcare workers are not following standard handwashing procedures, and provide feedback. The analysis results are transmitted to the alert and feedback module to help managers address non-compliant behaviors promptly, thereby improving the standardization and effectiveness of hand hygiene implementation.

[0068] Step S6: When the data analysis module determines that a medical staff member's handwashing behavior is non-compliant, the reminder feedback module will trigger an audible and visual alarm, issuing a reminder signal to prompt the medical staff member to correct the non-compliant handwashing actions in a timely manner. Simultaneously, the handwashing records and analysis results will be presented to administrators in a visual interface, facilitating the tracking and handling of non-compliant behaviors. Furthermore, the equipment status monitoring module monitors the operating status of each hardware device in the system in real time, including power consumption, network connectivity, and sensor operating status. When the system detects a device malfunction, the equipment status monitoring module will immediately send an alarm signal to the reminder feedback module and record the fault information to ensure the normal operation of the system.

[0069] Through the above six steps, the IoT-based hand hygiene monitoring and management system can accurately monitor and analyze the handwashing behavior of medical staff, effectively improve hand hygiene compliance in hospitals and other places, reduce the risk of cross-infection, and ensure public health safety.

[0070] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A hand hygiene monitoring and management system based on the Internet of Things, characterized in that, The system includes: an identity recognition module, a handwashing action monitoring module, a data acquisition and transmission module, a data storage module, a data analysis module, and an alert and feedback module. The identity recognition module is connected to the handwashing action monitoring module and the data acquisition and transmission module, respectively. It includes an RFID tag reading unit and a face recognition camera unit. The RFID tag reading unit is used to read the RFID tag information worn by medical staff to obtain the medical staff's identity. The face recognition camera unit captures facial images of medical staff entering the handwashing area and transmits the captured facial image information to the data acquisition and transmission module. The handwashing action monitoring module is connected to the identity recognition module and the data acquisition and transmission module, and the internal sensors work together. The module is equipped with a hand sanitizer usage sensor, a faucet switch sensor, and a hand motion sensor. The hand sanitizer usage sensor is used to monitor whether the hand sanitizer has been squeezed out; the faucet switch sensor detects the open and closed state of the faucet; and the hand motion sensor monitors the actions of medical staff rubbing their hands and rinsing them. Each sensor transmits the monitoring data to the data acquisition and transmission module in real time.

2. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The data acquisition and transmission module is connected to the identity recognition module, handwashing action monitoring module, data storage module, and data analysis module. It is responsible for collecting personnel identity information obtained by the identity recognition module and handwashing process data collected by the handwashing action monitoring module. The data is packaged and transmitted to the data storage module and data analysis module through Internet of Things (IoT) communication methods, including Wi-Fi, Bluetooth, or ZigBee. The data storage module is connected to the data acquisition and transmission module and the data analysis module. It adopts a database storage method to classify and store the collected personnel identity information, handwashing time and handwashing action data, providing data support for the data analysis module. The data analysis module is connected to the data acquisition and transmission module, the data storage module, and the reminder feedback module, respectively. It uses a preset analysis algorithm to process the stored data, including counting handwashing frequency, analyzing the compliance of handwashing steps, and transmitting the analysis results to the reminder feedback module. The reminder feedback module is connected to the data analysis module. When the data analysis module determines that the medical staff's handwashing behavior is not compliant, it will issue a reminder signal through an audible and visual alarm. At the same time, the handwashing records and analysis results will be presented to the management personnel in the form of a visual interface for review and processing. The data analysis module uses a convolutional neural network deep learning model to analyze handwashing action data. The model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. The input layer receives hand motion sensor data from the handwashing action monitoring module. The data dimension is [batch_size, sequence_length, feature_dim], where batch_size represents the number of samples in each input data, sequence_length represents the length of the action sequence, and feature_dim represents the dimension of each action feature. Convolutional layers use A convolution kernel of size 1, stride 1, padding 1, is calculated using the following formula: ,in, It is the first Convolutional layers at position The output, It is the first The convolutional kernel weights of the layer, It is the first Layer in position Input, It is the first Layer bias; Pooling layer adopts The maximum pooling size, with a step size of 2, is calculated using the following formula: For the l-th pooling layer at position The output; The weight matrix of the fully connected layer is , bias is The calculation formula is: It is the output of the fully connected layer. It is the feature vector after processing by the pooling layer; The output layer uses the Softmax function to convert the output of the fully connected layer into the probability of handwashing action classification. The formula for the Softmax function is: , It is the output vector of the fully connected layer. It represents the number of handwashing action categories.

3. The hand hygiene monitoring and management system based on the Internet of Things according to claim 2, characterized in that, The convolutional neural network employs a stochastic gradient descent optimization algorithm during training, with a learning rate of 0.001, a momentum parameter of 0.9, and a cross-tab loss function, as shown in the formula: ,in, It is the sample size. It is the number of handwashing action categories. It is a sample The true label, The model predicts the sample. Category The probability is determined by validation, which is performed every 50 training batches during training. The validation set is a proportion of the total dataset. The total number of batches for model training is set to 300.

4. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The RFID tag uses the ultra-high frequency band, with a frequency range of 860-960MHz and a transmission power of no more than 2W. The RFID tag has a unique identification code with a code length of 64 bits, which stores the medical staff's employee number, name, and department identity information.

5. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The hand sanitizer uses an infrared sensing sensor with a detection distance of 5-15 cm and a response time of no more than 0.1 seconds. When the liquid level in the hand sanitizer bottle is detected to be lower than a preset threshold, the sensor sends a low liquid level alarm signal to the data acquisition and transmission module. The preset threshold can be flexibly configured according to the capacity of the hand sanitizer bottle and the normal usage cycle, and the configuration range is 10%-20% of the total capacity in the bottle.

6. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The data storage module adopts a distributed storage architecture, consisting of multiple storage nodes. The storage nodes are connected through a high-speed network and use a consistent hash algorithm for data sharding and storage. The data is evenly distributed across the storage nodes according to the hash value to improve the scalability and fault tolerance of the storage system. When a storage node fails, the system automatically migrates the data to other normal nodes.

7. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The audible and visual alarm in the reminder feedback module uses a combination of flashing lights and sounds of different frequencies to provide reminders. When improper handwashing procedures are detected, the light flashes 3 times per second and the sound frequency is 1000Hz. When the hand sanitizer level is low, the light flashes 2 times per second and the sound frequency is 800Hz. The volume of the audible and visual alarm is adaptively adjusted according to the noise level in the hospital environment, with an adjustment range of 40-80 decibels.

8. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The handwashing motion monitoring module uses an inertial measurement unit (IMU) to integrate an accelerometer and a gyroscope. The accelerometer's measurement range is [missing information]. The gyroscope's measurement range is The sampling frequency is 100Hz. By fusing accelerometer and gyroscope data, a complementary filtering algorithm is used to calculate the posture angle of hand movements for monitoring hand rubbing and rinsing actions. The formula for the complementary filtering algorithm is: ,in, It is the first Attitude angle at any moment These are the filter coefficients, and their values ​​range from [value range missing]. It is the attitude angle measured by the accelerometer. It is the attitude angle measured by the gyroscope.

9. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The system also includes a device status monitoring module, which is connected to the identity recognition module, handwashing action monitoring module, data acquisition and transmission module, data storage module, data analysis module, and reminder feedback module. The device status monitoring module monitors the operating status of each hardware device in real time, including power level, network connection status, and sensor working status. When a device malfunction is detected, it sends a device malfunction alarm information to the reminder feedback module and records the malfunction time and malfunction type information to the data storage module.

10. The hand hygiene monitoring and management system based on the Internet of Things according to claim 1, characterized in that, The hardware components of the system are all designed to be waterproof and dustproof, with an IP65 protection rating. The equipment casing is made of antibacterial material with an antibacterial rate of no less than 99%, which is used to prevent bacterial growth.

Images