Rapid influenza screening system and method using detection of volatile organic compounds in exhaled breath

Abstract

The present application relates to a rapid screening system and method for influenza based on detection of volatile organic compounds in exhaled breath, belonging to the field of medical detection. The system comprises a breath sampling module, a sensor array module and a data analysis module. The breath sampling module collects exhaled breath from a subject and performs drying, filtering and quantitative pretreatment. The sensor array module detects the types and concentrations of 11 characteristic volatile organic compounds such as heptanal and nonanal in the pretreated gas. The data analysis module has an artificial intelligence algorithm built-in, which determines whether the subject is infected with influenza virus based on the disease fingerprint characteristics of volatile organic compounds. The system also includes an operation guidance module with voice and light guidance, and can be in the form of a mobile trolley or a portable device. The system uses non-invasive breath detection, does not require blood sampling or nasopharyngeal swabbing, and can complete the screening process in 2-3 minutes. The system has a sensitivity of 92.7% or higher and a specificity of 94.0% or higher, and is suitable for rapid screening of large populations in schools, communities and other places, significantly reducing the risk of cross-infection, and achieving early detection, early diagnosis and early treatment of influenza.

Description

Technical Field

[0001] This invention relates to the field of medical testing technology, specifically to a rapid influenza screening system and method for detecting volatile organic compounds in exhaled breath. Background Technology

[0002] Influenza virus is a common pathogen causing acute respiratory infectious diseases, and its seasonal epidemics and sudden pandemics place a heavy burden on global public health systems. Currently, commonly used clinical methods for influenza detection mainly include virus isolation and culture, reverse transcription polymerase chain reaction (RT-PCR), enzyme-linked immunosorbent assay (ELISA), and colloidal gold immunochromatographic strips. While virus isolation and culture is the gold standard, it takes 24-48 hours, making it difficult to meet the needs of rapid screening; RT-PCR has high sensitivity but relies on expensive thermal cycling equipment and specialized personnel, making it unsuitable for on-site testing; immunochromatographic strips are simple to operate but have low sensitivity and a high risk of false negatives. Furthermore, all of the above methods require the collection of nasopharyngeal swab samples, which is an invasive procedure, causing pain and discomfort to the examinee, especially children and the elderly, who have lower acceptance rates, and also posing a risk of cross-infection to healthcare workers during sampling. Studies have shown that there is a short effective detection window period after influenza virus infection; if patients are not tested immediately after exposure, conventional methods may produce false negative results. Therefore, developing a non-invasive, rapid, accurate, and large-scale influenza detection technology is of significant clinical and public health value.

[0003] In recent years, breath testing technology based on the analysis of volatile organic compounds (VOCs) in exhaled breath has attracted widespread attention due to its non-invasiveness and rapid detection potential. After infection with the influenza virus, the respiratory tract and metabolic system produce characteristic biochemical reactions, leading to changes in the types and concentrations of specific VOCs in exhaled breath, forming a VOCs spectrum with "disease fingerprint" characteristics. Previous studies have used gas chromatography-mass spectrometry (GC-MS) to confirm the presence of characteristic VOCs changes in the exhaled breath of influenza patients and have preliminarily explored the feasibility of using electronic nose sensors combined with artificial intelligence algorithms to identify viral respiratory infections. However, existing breath testing technologies still have many limitations: while GC-MS analysis is accurate, the equipment is expensive and bulky, requiring laboratory testing and preventing on-site real-time diagnosis; existing portable chemical sensors have limited sensitivity, only able to identify specific molecules at concentrations of one part per million, making it difficult to detect trace VOCs biomarkers; although electronic nose sensor arrays can identify disease-related VOCs patterns, current research is mostly based on headspace analysis of nasopharyngeal swab samples, and a complete technical solution for directly detecting influenza-characteristic VOCs from exhaled breath has not yet been established. In addition, exhaled breath contains a large amount of water vapor and complex matrix, and existing sampling techniques are unable to effectively remove interference and achieve quantitative detection, which limits the practical application of exhaled breath testing technology in rapid influenza screening. Summary of the Invention

[0004] To address the problems of the prior art, this invention provides a rapid influenza screening system and method for detecting volatile organic compounds in exhaled breath.

[0005] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution: Firstly, a rapid influenza screening system for detecting volatile organic compounds in exhaled breath, comprising: A respiratory sampling module is used to collect the exhaled gas of the subject and preprocess the gas. A sensor array module, connected to the respiratory sampling module, is used to detect the types and concentrations of specific volatile organic compounds in pretreated exhaled gas and generate detection signals; and The data analysis module is connected to the sensor array module and has a built-in artificial intelligence recognition algorithm. The data analysis module receives the detection signal and determines whether the subject is infected with the influenza virus based on the disease fingerprint characteristics of volatile organic compounds.

[0006] In one specific embodiment of the first aspect, the specific volatile organic compound includes at least one selected from heptanal, nonanal, octanal, hexanal, propanal, decanal, acetone, benzaldehyde, cyclohexanone, formaldehyde, and acetaldehyde.

[0007] In one specific embodiment of the first aspect, the respiratory sampling module includes: A disposable sterile respiratory sampler is used to collect the exhaled gases of the subject. A drying and filtering device, connected to the disposable sterile respiratory sampler, is used to remove moisture and particulate matter from exhaled air; and A quantitative injection device, connected to the drying and filtering device, is used to control the volume of gas entering the sensor array module.

[0008] In one specific embodiment of the first aspect, an operation guidance module is also included, which includes a voice prompt unit and / or a light indicator unit, for guiding the user to complete the detection operation according to the three-step process of power-on calibration, breath sampling, and automatic analysis.

[0009] In one specific embodiment of the first aspect, the system has multiple device forms, including mobile cart-type devices used in medical institutions and lightweight battery-powered portable devices suitable for public health scenarios such as campuses, communities, large events, airports, and ports.

[0010] Secondly, a rapid influenza screening method based on the detection of volatile organic compounds in exhaled breath includes the following steps: The respiratory sampling module collects the subject's exhaled gas and preprocesses the gas. The sensor array module detects the types and concentrations of specific volatile organic compounds in the pre-treated exhaled gas and generates a detection signal. The detection signal is analyzed using the built-in artificial intelligence recognition algorithm of the data analysis module, and the subject's infection status is determined based on the fingerprint characteristics of volatile organic compounds; and Output the detection results.

[0011] In one specific embodiment of the second aspect, the step of pre-treating the gas includes: removing water vapor and particulate matter from the exhaled gas using a drying filter device, and controlling the gas volume entering the sensor array module using a quantitative injection device. In one specific implementation of the second aspect, the method further includes a guidance step: guiding the user to complete the testing operation through a three-step process of power-on calibration, breath sampling, and automatic analysis via a voice prompt unit and / or a light indicator unit.

[0012] In one specific implementation of the second aspect, the method is a non-invasive detection method that does not require blood drawing or nasopharyngeal swab sampling of the subject.

[0013] In one specific implementation of the second aspect, the method is applicable to large-scale population screening scenarios, including mobile screening or rapid channel testing in schools, communities, large events, airports, and ports.

[0014] The beneficial effects of this invention are as follows: 1. By employing a combination of sensor arrays and artificial intelligence algorithms, this technology enables the joint detection of 11 characteristic volatile organic compounds (VOCs) in exhaled breath, including heptanal, nonanal, octanal, hexanal, propional, decanal, acetone, benzaldehyde, cyclohexanone, formaldehyde, and acetaldehyde. Utilizing the characteristic VOC fingerprint profiles generated by human metabolism after infection with different influenza viruses, it achieves rapid screening for influenza viruses with high sensitivity and specificity. The sensor array integrates multiple types of sensitive elements, including molecularly imprinted QCM sensors, MOS sensors, SAW sensors, and electrochemical sensors. Through redundant detection and cross-sensitivity mechanisms, it effectively eliminates cross-interference and drift errors that may exist with single sensors, significantly improving the stability and reliability of the detection signal. Combined with a drying and filtration device to remove water vapor and particulate matter from exhaled breath, and a quantitative injection device to strictly control the injection volume, it eliminates interference from external environmental factors on the detection results, ensuring the repeatability and comparability of the detection data. The artificial intelligence algorithm, based on a deep learning model, automatically extracts multidimensional features from the sensor response curve through a convolutional neural network and compares and analyzes them with a pre-set influenza VOCs feature fingerprint database. It can accurately identify influenza virus infection in the early stage of infection (within 24 hours of onset), with a detection sensitivity of over 92.7% and a specificity of over 94.0%. It provides reliable technical support for early clinical intervention and truly achieves the prevention and control goal of "early detection, early diagnosis, and early treatment".

[0015] This invention employs a non-invasive breath test method, eliminating the need for blood draws or nasopharyngeal swabs, thus avoiding the pain and cross-infection risks associated with invasive procedures. It is particularly suitable for influenza screening of children over 3 years old, the elderly, and individuals contraindicated for invasive examinations. The operation guidance module uses a combination of voice prompts and light indicators to guide users through the three-step process of power-on calibration, breath sampling, and automatic analysis, significantly reducing operator training needs and human error. The system adopts a modular design, configurable as a mobile cart-style device or a lightweight portable device depending on the application scenario. Both the drying filter and sensor array feature automatic calibration and one-click replacement, enabling maintenance-free operation. By uploading test data to a remote monitoring platform in real time, dynamic monitoring and epidemiological analysis of large-scale population screening results can be achieved, providing data support for medical institutions to quickly triage patients and optimize medical resource allocation. Simultaneously, it reduces the frequency of contact between medical staff and patients, effectively minimizing the risk of nosocomial infections. This invention is easy to operate, quick to detect (the whole process takes about 2-3 minutes), and requires no professional laboratory conditions. It can be widely deployed in public health scenarios such as campuses, communities, airports, and ports, providing an efficient technical means for early warning and rapid response to influenza outbreaks. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system framework of the present invention.

[0017] Figure 2 This is a schematic diagram of the respiratory sampling module of the present invention.

[0018] Figure 3 This is a schematic diagram of the sensor array module of the present invention.

[0019] Figure 4 This is a schematic diagram of the detection workflow of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] like Figures 1 to 4 The present invention describes a rapid influenza screening system and method for detecting volatile organic compounds in exhaled breath.

[0022] I. System Overall Architecture like Figure 1 As shown, the rapid influenza screening system of the present invention includes five core modules: a respiratory sampling module, a sensor array module, a data analysis module, a result output module, and an operation guidance module. The respiratory sampling module and the sensor array module are connected via an airway; the sensor array module is signal-connected to the data analysis module; and the data analysis module is electrically connected to both the result output module and the operation guidance module.

[0023] The system adopts a modular design and can be configured as a mobile cart-type device or a lightweight portable device depending on the application scenario. The cart-type device is suitable for scenarios such as hospital fever clinics and emergency departments, and is equipped with a large-capacity gas source and a continuous power supply system; the portable device is battery-powered, weighs less than 2kg, and is suitable for mobile screening scenarios such as campuses, communities, and airports.

[0024] II. Specific Implementation of the Respiratory Sampling Module The respiratory sampling module is used to collect and preprocess the exhaled air of the subject, including a disposable sterile respiratory sampler, a drying and filtering device, and a quantitative injection device.

[0025] 2.1 Disposable sterile respiratory sampler This disposable sterile respiratory sampler is made of medical-grade polypropylene and includes a mouthpiece, a one-way valve, and a sampling bag. The mouthpiece is ergonomically designed with a built-in backflow preventer to prevent gas contamination. The one-way valve ensures that exhaled gas enters the sampling bag in one direction. The sampling bag has a capacity of 500mL-1000mL and is made of inert Tedlar® or aluminum foil composite membrane, exhibiting low VOC adsorption (-5).

[0026] The sampler is sterilized with ethylene oxide before leaving the factory and is individually aseptically packaged to ensure single use and prevent cross-contamination. The sampler and the drying and filtering device are connected by a quick connector for easy replacement.

[0027] 2.2 Drying and Filtration Device The drying filter removes moisture and particulate matter from exhaled breath, protecting the sensor array from interference. The device employs a two-stage filtration structure: The first stage is a hydrophobic membrane filter, using a polytetrafluoroethylene (PTFE) microporous membrane with a pore size of 0.45 μm, which can effectively remove water droplets and particulate matter with a diameter greater than 0.45 μm from the gas. The filter membrane has hydrophobic properties, with a water vapor contact angle greater than 120°, preventing water vapor condensation and clogging of the filter pores.

[0028] The second stage is a drying tube containing molecular sieve desiccant (3A or 4A type molecular sieve), which further adsorbs residual moisture and reduces the relative humidity of the gas to below 30%. The drying tube is made of transparent polycarbonate, allowing observation of the desiccant's color change (e.g., blue silica gel turns pink after absorbing water), indicating when it needs to be replaced.

[0029] The drying and filtration device adopts a modular design and can be replaced as a whole. It is recommended to replace it every 100 tests or when the desiccant changes color by more than 50%.

[0030] 2.3 Quantitative injection device The quantitative injection device is used to control the gas volume entering the sensor array module, ensuring consistent gas injection volume for each detection and guaranteeing detection repeatability. The device includes a miniature gas pump, a solenoid valve, a flow sensor, and a quantitative loop.

[0031] The miniature air pump is a diaphragm vacuum pump with a maximum flow rate of 300 mL / min and selectable operating voltage of 5V / 12V. The solenoid valve is a two-position three-way miniature solenoid valve with a response time of less than 10ms. The flow sensor is a MEMS thermal flow sensor with a range of 0-500 mL / min and an accuracy of ±2%FS. The metering loop uses a stainless steel capillary tube, with a volume precisely controlled to 50 mL ± 0.1 mL.

[0032] Workflow: Initially, the solenoid valve is closed and the gas pump is not operating. After detection is initiated, the gas pump operates, and the gas, after being dried and filtered, enters the metering loop. When the gas pressure within the metering loop reaches the set value (usually atmospheric pressure), the solenoid valve switches, and the gas within the metering loop is propelled into the sensor array module by the carrier gas (clean air). The injection volume is precisely controlled by the metering loop volume and is unaffected by the blowing force or blowing time.

[0033] III. Specific Implementation of the Sensor Array Module The sensor array module is used to detect the types and concentrations of specific volatile organic compounds in exhaled breath, and includes multiple high-sensitivity gas sensors, signal conditioning circuitry, and analog-to-digital converters.

[0034] 3.1 Sensor Array Design The sensor array consists of 8-16 gas sensors made of different sensitive materials, optimized for influenza-specific VOCs. The target VOCs include 11 compounds such as heptanal, nonanal, octanal, hexanal, propanal, decanal, acetone, benzaldehyde, cyclohexanone, formaldehyde, and acetaldehyde.

[0035] Sensor types include: For the detection of aldehydes, a quartz crystal microbalance (QCM) sensor-2 prepared using molecular imprinting technology is preferred. A molecularly imprinted thin film is prepared on the surface of the QCM gold electrode using hexanoic acid and octanoic acid as template molecules and polyacrylic acid as the polymer matrix. This sensor has a response time of less than 5 seconds and a recovery time of less than 12 seconds for hexanal, heptanal, and nonanal, achieving a sensitivity at the ng / mL level.

[0036] 3.2 Sensor Array Layout The sensor array adopts a circular layout, with a reference sensor (coated with a non-imprinted polymer) at the center and 7-15 sensitive sensors evenly distributed around it. The array is placed in a constant-temperature chamber, with the temperature controlled at 37℃±0.1℃ to simulate the human body temperature environment and reduce the impact of temperature fluctuations on the sensor response. The chamber is made of stainless steel with an inert inner wall to reduce VOCs adsorption.

[0037] 3.3 Signal Processing Circuit The frequency signal (QCM / SAW sensor) or resistance signal (MOS sensor) output by the sensor is amplified and filtered by the signal conditioning circuit. The QCM sensor uses an oscillation circuit to convert frequency changes into a voltage signal, while the MOS sensor uses a voltage divider circuit to detect resistance changes. The signal-conditioned analog signal is converted into a digital signal by a 24-bit high-precision analog-to-digital converter and transmitted to the data analysis module via the SPI / I2C interface.

[0038] 3.4 Automatic calibration function The sensor array module features automatic calibration. Upon each power-on, the system automatically introduces a standard calibration gas (nitrogen containing known concentrations of VOCs) and records the sensor response value as a baseline. If the response value deviates from the preset range (±10%), the system automatically adjusts the sensor operating parameters (such as heating voltage and bias voltage) or prompts for sensor replacement. The calibration process takes approximately 30 seconds, with the user receiving voice prompts from the operation guidance module to wait.

[0039] IV. Specific Implementation of the Data Analysis Module The data analysis module incorporates an artificial intelligence recognition algorithm, receives detection signals from the sensor array module, and determines whether the subject is infected with the influenza virus based on the disease fingerprint characteristics of volatile organic compounds.

[0040] 4.1 Hardware Platform The data analysis module employs an embedded processor solution, including a main control chip, memory, and communication interfaces. The main control chip is an ARM Cortex-M7 series microcontroller (such as the STM32F767) or a low-power edge computing chip (such as the LingSi CSK6), with a main frequency of 400MHz or higher, and built-in 2MB Flash and 1MB RAM-7. For cart-type devices, an NPU acceleration module (such as a neural network processing unit) can be added to improve the efficiency of AI algorithm operation.

[0041] The memory includes external Flash (16MB-64MB) for storing algorithm models and historical data, and a MicroSD card slot for data export. Communication interfaces include UART (for connecting to the sensor array), USB (for connecting to the results output module), and Wi-Fi / Bluetooth (for connecting to a remote monitoring platform).

[0042] 4.2 Algorithm Model The artificial intelligence recognition algorithm uses a deep learning model, including a feature extraction layer and a classification layer. The algorithm development process is as follows: Step 1: Data Acquisition and Preprocessing. Exhaled gas samples were collected from influenza-positive patients diagnosed by RT-PCR and healthy volunteers. Gas chromatography-mass spectrometry (GC-MS) was used to determine the types and concentrations of VOCs in the samples. The sensor array response data was correlated with the GC-MS data to create a training dataset.

[0043] Step 2: Feature Extraction. A Convolutional Neural Network (CNN) or ResNet is used to automatically extract features from the sensor response curves. The multidimensional time-series data (8-16 channels × 30-second sampling) of the sensor array is converted into feature vectors. The network structure includes 3 convolutional layers, 2 pooling layers, and 1 fully connected layer, using ReLU activation and Dropout (ratio 0.5) to prevent overfitting.

[0044] Step 3: Classification and Recognition. A Softmax classifier is used to map the feature vectors to three categories: influenza positive, influenza negative, and invalid test. The classification layer outputs three probability values, and the category corresponding to the maximum value is taken as the final result. The algorithm model is evaluated on the training set (70% of the data), validation set (15%), and test set (15%), and the accuracy can reach over 90%.

[0045] 4.3 Model Deployment The trained algorithm model is converted into C code or TensorFlow Lite format and stored in the Flash memory of the data analysis module. During runtime, the system inputs real-time response data from the sensor array into the model and calculates the classification result through forward propagation. The single inference time is less than 100ms, meeting real-time detection requirements.

[0046] 4.4 Feature Fingerprint Database The data analysis module incorporates a built-in influenza VOCs characteristic fingerprint database, containing characteristic spectra of different types of influenza viruses (H1N1, H3N2, and B, etc.). As clinical data accumulates, the fingerprint database can be remotely updated via OTA (Over-The-Air) to improve detection accuracy and coverage.

[0047] V. Specific Implementation of the Operation Guidance Module The operation guidance module includes a voice prompt unit and a light indicator unit, which guides users to complete the testing operation according to the three-step process of power-on calibration, exhalation sampling, and automatic analysis.

[0048] 5.1 Voice Prompt Unit The voice prompt unit uses a speech synthesis chip (such as SYN6288) or an embedded MP3 playback module, paired with a small speaker (1W-3W) to achieve voice broadcasting. The voice content is pre-recorded in Mandarin Chinese, but other languages ​​can also be customized upon request.

[0049] Typical voice prompts: After powering on: "System self-test in progress, please wait..." Calibration in progress: "Sensor calibration in progress, approximately 30 seconds..." Calibration complete: "Calibration complete, please begin blowing air." Sampling Stage: "Please hold the disposable mouthpiece in your mouth, breathe normally, then take a deep breath and exhale forcefully... Sampling complete, thank you for your cooperation." Analysis phase: "Analyzing, please wait...approximately 1 minute" Output: "Test complete. Result is negative / positive / invalid. Please check the screen." 5.2 Lighting indicator unit The lighting indicator unit uses RGB LED light strips or indicator lights to intuitively indicate the system status through color changes -4-7: The light indicators and voice prompts work in sync, making it convenient for users with hearing impairments or in noisy environments.

[0050] VI. Specific Implementation of Equipment Form 6.1 Mobile cart-type equipment Suitable for medical facilities, the device is integrated into a medical trolley with an adjustable height (80cm-120cm) and equipped with omnidirectional silent casters. Main components include: Main unit: houses the respiratory sampling module, sensor array module, and data analysis module; Touchscreen display: 10-15 inch industrial-grade touchscreen, displaying the user interface and test results; Printer: Thermal printer, prints test reports in real time; Power system: Built-in lithium battery (4-6 hours of battery life) and external power adapter; Consumables storage area: Stores disposable respiratory samplers, drying and filtering devices, and other consumables.

[0051] 6.2 Portable devices Suitable for public health settings, the overall dimensions are less than 200mm × 150mm × 80mm, and the weight is less than 2kg. Main components include: Main unit: Integrated design, incorporating all modules; Battery: Rechargeable lithium battery, with a battery life of 8-10 hours (after more than 50 continuous tests); Display: 5-7 inch touchscreen; Handheld sampler: Connects to the main unit via a flexible tube for easy handheld operation by the subject; Carrying case: Waterproof and shockproof ABS material, with built-in consumables storage compartment.

[0052] VII. Detailed Workflow Description 7.1 Power-on Preparation Press the power button to initiate a system power-on self-test. The operation guidance module will display a blue breathing light and a voice prompt stating "System self-test in progress." The self-test includes: sensor array connectivity, air pump operating status, memory read / write operations, and communication interface checks. After the self-test passes, the system will automatically enter the calibration procedure.

[0053] 7.2 Sensor Calibration The system automatically introduces standard calibration gas, and the sensor array module records the baseline response value. The calibration process takes approximately 30 seconds, during which a voice prompt will say, "Sensor calibration is in progress, please wait," and the indicator light will remain blue and breathing. If the calibration is successful, a voice prompt will say, "Calibration complete, please begin blowing gas"; if the calibration fails, a voice prompt will say, "Calibration failed, please check the sensor or contact maintenance," and the indicator light will flash red.

[0054] 7.3 Exhalation sampling The operator unpacks the disposable sterile respiratory sampler and installs it into the drying and filtering device. A voice prompt appears: "Please hold the disposable mouthpiece in your mouth, take a deep breath after normal breathing, and then exhale forcefully." The light flashes green.

[0055] Subjects follow the prompts to exhale. The breath sampling module collects exhaled air, which is then dried, filtered, and controlled by a quantitative injection device before entering the sensor array module. The sampling process takes approximately 15-30 seconds. Upon completion, a voice prompt will say "Sampling complete, thank you for your cooperation," and the indicator light will turn solid yellow.

[0056] 7.4 Automatic Analysis The sensor array module begins detection, recording the response curves of each sensor in real time. The sampling time is 60 seconds, recording the sensor's dynamic response (rise time, steady-state value) and recovery characteristics. The data analysis module acquires sensor data in real time, once every 100ms, forming 8-16 channel × 600 point time-series data.

[0057] The data analysis module inputs the collected sensor response data into an artificial intelligence algorithm model, compares it with a feature fingerprint database, and calculates the probability of a positive influenza test. The analysis process takes approximately 60 seconds, during which a voice prompt will say "Analyzing, please wait," and the lights will remain constantly yellow.

[0058] 7.5 Results Output After the analysis is complete, the data analysis module sends the results to the results output module. The results output module includes a touch screen and voice announcement. The screen displays: The system simultaneously announces "Test complete, result is negative / positive / invalid" and the lights change to solid green (negative), solid red (positive), or flashing red (invalid) depending on the result.

[0059] Test results can be uploaded to a remote monitoring platform (such as a hospital information system or a public health monitoring platform) via Wi-Fi / Bluetooth, supporting data storage, statistical analysis, and epidemiological early warning.

[0060] 7.6 Consumables Replacement After each test, the system prompts you to replace the disposable breath sampler. The drying filter should be replaced every 100 tests or when more than 50% of it changes color. The sensor array module has a lifespan indicator, allowing for 500-1000 tests; the system will prompt you to replace it after this period.

[0061] The consumable replacement adopts a one-button design. Press the release button to remove the old consumable, insert the new consumable and it will lock automatically. The voice prompt "Consumable replacement successful, you can proceed to the next test"-7.

[0062] VIII. Application Scenarios and Examples Example 1: Campus Influenza Screening During an influenza outbreak at a primary school, the portable device of this invention was used to conduct rapid screening of all teachers and students. The device was deployed in the school clinic and operated by the school doctor. Each student's test took approximately 2-3 minutes (including preparation time), and a single device could screen 160-240 people in 8 hours. The screening identified 12 students with fever and influenza symptoms, of whom 10 tested positive, a positive rate of 83.3%. Positive students were advised to isolate at home and seek medical confirmation, effectively controlling the spread of the epidemic.

[0063] Example 2: Airport Port Quarantine Rapid, mobile testing equipment has been deployed at the arrival halls of an international airport. Two devices are used in parallel at each hall. Arriving passengers complete breathalyzer sampling while waiting for their baggage, and test results are available before baggage claim. Flights from influenza-endemic areas undergo full screening, with a capacity of 60-80 people per hour. Positive passengers are guided by quarantine personnel to a medical point for further examination.

[0064] Example 3: Influenza surveillance among the elderly in the community A community health service center provided free influenza screening services for people aged 65 and above. Portable devices were used, operated by community nurses, and test results were printed immediately. After three months of continuous monitoring, a total of 1200 people were screened, detecting 86 positive cases (7.2% positivity rate). All positive cases were identified and referred for treatment within 24 hours of symptom onset. The average course of illness was shortened by 3-4 days compared to previous years, and the hospitalization rate decreased by 45%.

[0065] IX. Verification of Technical Effects 9.1 Detection Performance Verification The system of this invention was used to test 300 clinical samples (150 positive for influenza by RT-PCR and 150 negative), and the results are as follows: 9.2 Repeatability Validation Ten repeated tests on the same influenza-positive sample all yielded positive results, with a coefficient of variation (CV) of less than 5%. Ten repeated tests on the same negative sample all yielded negative results, with no false positives.

[0066] 9.3 Stability Verification After 8 hours of continuous operation (testing 50 people), the equipment was retested for standard gas. The sensor response value drift was less than ±3%, and it returned to normal after calibration. The equipment operates normally in environments with temperatures ranging from 10℃ to 35℃ and humidity from 20% to 80%, and its performance indicators meet the requirements.

[0067] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various 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 rapid influenza screening system for detecting volatile organic compounds in exhaled breath, characterized in that, include: A respiratory sampling module is used to collect the exhaled gas of the subject and preprocess the gas. The sensor array module, connected to the breathing sampling module, is used to detect the types and concentrations of specific volatile organic compounds in the pretreated exhaled gas and generate detection signals. as well as The data analysis module is connected to the sensor array module and has a built-in artificial intelligence recognition algorithm. The data analysis module receives the detection signal and determines whether the subject is infected with the influenza virus based on the disease fingerprint characteristics of volatile organic compounds.

2. The rapid influenza screening system for detecting volatile organic compounds in exhaled breath according to claim 1, characterized in that: The specific volatile organic compounds include at least one of heptanal, nonanal, octanal, hexanal, propanal, decanal, acetone, benzaldehyde, cyclohexanone, formaldehyde, and acetaldehyde.

3. The rapid influenza screening system for detecting volatile organic compounds in exhaled breath according to claim 1, characterized in that: The respiratory sampling module includes: A disposable sterile respiratory sampler is used to collect the exhaled gases of the subject. A drying and filtering device, connected to the disposable sterile respiratory sampler, is used to remove moisture and particulate matter from exhaled air; and A quantitative injection device, connected to the drying and filtering device, is used to control the volume of gas entering the sensor array module.

4. The rapid influenza screening system for detecting volatile organic compounds in exhaled breath according to claim 1, characterized in that: It also includes an operation guidance module, which includes a voice prompt unit and / or a light indicator unit, used to guide the user to complete the testing operation according to the three-step process of power-on calibration, exhalation sampling, and automatic analysis.

5. The rapid influenza screening system for detecting volatile organic compounds in exhaled breath according to claim 1, characterized in that: The system has various device forms, including mobile cart-type devices used in medical institutions and lightweight battery-powered portable devices suitable for public health scenarios such as campuses, communities, large events, airports, and ports.

6. A rapid influenza screening method based on the detection of volatile organic compounds in exhaled breath, characterized in that: Includes the following steps: The respiratory sampling module collects the subject's exhaled gas and preprocesses the gas. The sensor array module detects the types and concentrations of specific volatile organic compounds in the pre-treated exhaled gas and generates a detection signal. The detection signal is analyzed by the built-in artificial intelligence recognition algorithm of the data analysis module, and the subject is determined to be infected with influenza virus based on the fingerprint characteristics of volatile organic compounds; and the detection results are output.

7. The rapid influenza screening method for detecting volatile organic compounds in exhaled breath according to claim 6, characterized in that: The gas pretreatment steps include: removing water vapor and particulate matter from exhaled gas using a drying and filtration device, and controlling the gas volume entering the sensor array module using a quantitative injection device.

8. The rapid influenza screening method for detecting volatile organic compounds in exhaled breath according to claim 6, characterized in that: The method also includes a guidance step: guiding the user to complete the testing operation through a three-step process of power-on calibration, exhalation sampling, and automatic analysis via a voice prompt unit and / or light indicator unit.

9. The rapid influenza screening method for detecting volatile organic compounds in exhaled breath according to claim 6, characterized in that: The method described is a non-invasive detection method that does not require blood or nasopharyngeal swab sampling from the subject.

10. The rapid influenza screening method for detecting volatile organic compounds in exhaled breath according to claim 6, characterized in that: The method is applicable to large-scale population screening scenarios, including mobile screening or rapid testing at campuses, communities, large events, airports, and ports.

Images