Electronic nose infection monitoring system

Abstract

Disclosed is an electronic nose infection monitoring system. The electronic nose infection monitoring system includes: a sensor array including a plurality of gas sensors for individually detecting gas components included in an odor generated from a microorganism; and an infection analysis unit for determining the species of the microorganism by analyzing a reaction pattern of the gas sensors using a pre-trained deep learning algorithm.

Description

Infection Monitoring Electronic Nose System

[0001] The present invention relates to an electronic nose system for monitoring infection, and more specifically, to an electronic nose system capable of monitoring whether or not a pathogen is infected.

[0002] Premature babies, those who are underweight for their gestational age, and those requiring intensive monitoring for other reasons are cared for in incubators. Incubators provide safe protection from the external environment and regulate temperature and humidity to maintain a constant body temperature.

[0003] When hospitalized babies are discharged, neonatal incubators are thoroughly cleaned and disinfected. However, the complex structure of incubators makes cleaning and decontamination difficult. If a newborn is admitted in this condition, exposure to microorganisms can lead to neonatal infection. Consequently, there has been a recent and persistent increase in neonatal infections in incubators. High-risk newborns are vulnerable to infection and can die within days of infection.

[0004] Meanwhile, intensive care unit patients are constantly on ventilators, which can lead to pneumonia if upper respiratory and gastrointestinal bacteria are inhaled into the lungs. Ventilator-associated infections can lead to higher infection rates and increased ICU stays, contributing to increased medical costs.

[0005] For this reason, real-time monitoring for infection is required for patients who are vulnerable to infection.

[0006] The present invention provides an infection monitoring electronic nose system capable of rapidly monitoring whether a patient is infected.

[0007] In addition, the present invention provides an infection monitoring electronic nose system capable of monitoring in real time whether a newborn hospitalized in an incubator and a patient wearing a ventilator are infected.

[0008] An infection monitoring electronic nose system according to an embodiment of the present invention comprises a sensor array provided with a plurality of gas sensors that individually detect gas components contained in odors generated from microorganisms; and an infection analysis unit that analyzes the response patterns of the gas sensors using a pre-trained deep learning algorithm to determine the species of the microorganism.

[0009] Additionally, at least two of the above gas sensors may detect the same gas component, but may have different detection concentration ranges.

[0010] In addition, the gas sensors can be divided into a first group consisting of gas sensors that detect at least one of Ethanol, Butane, propane, and Alcohol solvent vapors; and a second group consisting of gas sensors that detect at least one of LPG, smoke, alcohol, propane, hydrogen, methane, carbon monoxide, CH4, natural gas, carbon monoxide, hydrogen, ammonia, benzene, and alcohol.

[0011] Additionally, the microbial species may include Staphylococcus epidermidis, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Candida albicans, Candida glabrata, and Candida parapsilosis.

[0012] Additionally, the microbial species may include Acinetobacter baumannii, Staphylococcus aureus, Klebsiella pneumoniae, and Pseudomonas aeruginosa.

[0013] Additionally, the sensor array can be mounted inside an incubator for newborns.

[0014] In addition, the device further includes an exhaust pipe connected to an incubator for newborns and through which internal air of the incubator for newborns is exhausted; and a capturing unit connected to the exhaust pipe and capturing odor particles contained in the air, wherein the sensor array may be located within the capturing unit.

[0015] In addition, the capturing unit may include a housing connected to the exhaust pipe and having a space formed therein; a filter positioned within the housing and capturing the odor particles; and a heater for heating the inside of the housing.

[0016] Additionally, the sensor array may be mounted at the outlet of the ventilator.

[0017] An infection monitoring method according to an embodiment of the present invention includes a step of measuring the reactions of a plurality of gas sensors with gas components contained in an odor generated from a microorganism; and a step of analyzing the reaction patterns of the gas sensors using a pre-trained deep learning algorithm to determine the species of the microorganism.

[0018] According to the present invention, the infection monitoring electronic nose system can learn the reaction patterns of odor and gas sensors caused by infection using an artificial intelligence algorithm, thereby determining the type of microorganism from the odor and confirming the presence or absence of infection.

[0019] In addition, according to the present invention, by analyzing the odor generated from the breath of a newborn admitted to an incubator and the reaction pattern of gas sensors, it is possible to determine whether the newborn is infected.

[0020] In addition, according to the present invention, by analyzing the odor generated from the breath of a patient wearing an artificial respirator and the reaction pattern of a gas sensor, it is possible to determine whether the patient is infected.

[0021] FIG. 1 is a drawing showing an infection monitoring electronic nose system according to an embodiment of the present invention.

[0022] Figure 2 is a heat map showing the reactions of gas sensors according to an embodiment of the present invention to a total of nine gas components included in the odor generated from microorganisms.

[0023] Figure 3 is a graph showing the results of classifying gas components by a combination of gas sensors according to Table 1 and the results of classifying gas components by a combination of gas sensors according to Table 2.

[0024] Figure 4 is a confusion matrix showing the accuracy of gas component analysis learned by the three artificial intelligence algorithms described above.

[0025] Figure 5 is a graph measuring the reaction between gas sensors according to Table 2 and odors generated from microorganisms.

[0026] Figures 6 to 12 are graphs showing the measurement of odors generated by various microorganisms and the reactions of gas sensors.

[0027] FIG. 13 is a diagram showing an infection monitoring electronic nose system according to one embodiment of the present invention.

[0028] FIG. 14 is a diagram showing an infection monitoring electronic nose system according to another embodiment of the present invention.

[0029] Figure 15 is a drawing showing the detailed configuration of an infection monitoring electronic nose system.

[0030] FIG. 16 is a diagram showing an infection monitoring electronic nose system according to another embodiment of the present invention.

[0031] Figure 17 is a drawing showing an infection monitoring method according to an embodiment of the present invention.

[0032] An infection monitoring electronic nose system according to an embodiment of the present invention comprises a sensor array provided with a plurality of gas sensors that individually detect gas components contained in odors generated from microorganisms; and an infection analysis unit that analyzes the response patterns of the gas sensors using a pre-trained deep learning algorithm to determine the species of the microorganism.

[0033] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. However, the technical concept of the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided to ensure that the disclosed content is thorough and complete and to sufficiently convey the spirit of the present invention to those skilled in the art.

[0034] In this specification, when a component is referred to as being on another component, it means that it can be formed directly on the other component, or a third component may be interposed between them. In addition, in the drawings, the thicknesses of films and regions are exaggerated for the purpose of effectively explaining the technical contents.

[0035] Also, although terms such as first, second, and third have been used to describe various components in various embodiments of this specification, these components should not be limited by these terms. These terms are only used to distinguish one component from another. Thus, what is referred to as a first component in one embodiment may be referred to as a second component in another embodiment. Each embodiment described and illustrated herein also includes its complementary embodiments. Also, the term "and / or" has been used herein to mean including at least one of the components listed before and after.

[0036] In the specification, singular expressions include plural expressions unless the context clearly dictates otherwise. In addition, terms such as "comprise" or "have" are intended to specify the presence of a feature, number, step, component, or combination thereof described in the specification, and should not be construed as excluding the presence or addition of one or more other features, numbers, steps, components, or combinations thereof. In addition, the term "connection" is used in the present specification to mean both indirectly connecting multiple components and directly connecting them.

[0037] In addition, when describing the present invention below, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

[0038]

[0039] FIG. 1 is a drawing showing an infection monitoring electronic nose system according to an embodiment of the present invention.

[0040] Referring to Fig. 1, the infection monitoring electronic nose system (10) detects odors generated by microorganisms and learns the odor patterns observed by each microorganism using an artificial intelligence algorithm to determine the species of the microorganism from the odor. Microorganisms emit microbial VOCs (mVOCs) during their metabolic processes, and mVOCs are emitted as organic compounds with different components depending on the microorganism. Microorganisms include various pathogens. The infection monitoring electronic nose system (10) detects the components of mVOCs to determine the species of the microorganism.

[0041] The infection monitoring electronic nose system (10) includes a sensor array (100) and an infection analysis unit (200).

[0042] The sensor array (100) detects gas components contained in an odor generated from microorganisms. The sensor array (100) is provided with a plurality of gas sensors (110, 120). The gas sensors (110, 120) individually detect gas components contained in the odor. Each of the gas sensors (110, 120) may exhibit different responses depending on the gas component and gas concentration. According to an embodiment, at least two or more of the gas sensors (110, 120) detect the same type of gas component, but gas sensors having different detection concentration ranges may be used.

[0043] Gas sensors (110, 120) can be divided into at least two groups according to detectable gas components and gas concentrations. According to an embodiment, the gas sensors can be divided into two groups (G1, G2).

[0044] Gas sensors (110) belonging to the first group (G1) can detect at least one gas component among Ethanol, Butane, Propane, Alcohol solvent vapors, hydrogen, and carbon monoxide.

[0045] According to an embodiment, the first group (G1) includes six gas sensors (110), and the first gas sensor can detect hydrogen and ethanol among odor particles in the air, and can detect hydrogen at a concentration of 1-300 ppm. The second gas sensor detects odor particles in the air, and has high response sensitivity mainly to ethanol gas. The second gas sensor can detect ethanol gas at a concentration of 1-30 ppm. The third gas sensor detects odor particles in the air, and has high response sensitivity to ethanol gas. The third detection sensor can detect ethanol gas at a concentration of 1-10 ppm. The fourth gas sensor has high response sensitivity to butane and propane, and can detect butane and propane gases at a concentration of 1-25% LEL. The fifth gas sensor can detect alcohol solvent vapors, and can detect alcohol solvent vapors at a concentration of 50-5,000 ppm. The sixth gas sensor can detect alcohol and carbon monoxide, and can detect alcohol and carbon monoxide at concentrations of 50-5000 ppm.

[0046] Gas sensors (120) belonging to the second group (G2) can detect at least one gas component among LPG, smoke, alcohol, propane, hydrogen, methane, carbon monoxide, CH4, natural gas, carbon monoxide, hydrogen, ammonia, benzene, and alcohol.

[0047] According to an embodiment, the second group (G2) includes six gas sensors (120), and the first gas sensor can detect LPG, smoke, alcohol, propane, hydrogen, methane, and carbon monoxide gases, and can detect gases in a concentration range of 200 to 10,000 ppm. The second gas sensor can detect CH4 and natural gas gases, and can detect gases in a concentration range of 200 to 10,000 ppm. The third gas sensor can detect carbon monoxide gas in a concentration range of 20 to 2,000 ppm. The fourth gas sensor can detect hydrogen in a concentration range of 100 to 10,000 ppm. The fifth gas sensor can detect carbon monoxide and methane, and can detect carbon monoxide in a concentration range of 10 to 500 ppm, and methane in a concentration range of 300 to 10,000 ppm. The sixth gas sensor can detect ammonia, benzene, and alcohol, and can detect ammonia and alcohol at concentrations of 10 to 300 ppm, and benzene at concentrations of 10 to 1,000 ppm.

[0048] The gas components and concentration ranges that can be detected by gas sensors (110, 120) belonging to the first and second groups are as shown in Table 1 below.

[0049] GroupGas sensorGas compositionGas concentration rangeⅠ1Air contaminants (hydrogen, ethanol)1-300ppm (H2)2Air pollutantsEthanol 1-30ppm3Air pollutantsEthanol 1-10ppm4Alcohol solvent vapors50-5000ppm5Butane, propane1-25% LEL6Alcohol, carbon monoxide50-5000ppmⅡ1LPG, smoke, alcohol, propane, hydrogen, methane, carbon monoxide200-10000ppm2CH4, natural gas200-10000ppm3carbon monoxide20-2000ppm4hydrogen100-10000ppm5Carbon monoxide, methane10-500ppm (CO), 300-10000ppm (CH4)6ammonia, benzene, alcohol10-300ppm (NH3, alcohol), 10-1000ppm(benzene)

[0050] By combining the above-described gas sensors (110, 120), the reaction with gas components generated from microorganisms was measured.

[0051]

[0052] Figure 2 is a heat map showing the reactions of gas sensors according to an embodiment of the present invention to a total of nine gas components included in the odor generated from microorganisms.

[0053] Referring to Figure 2, it can be seen that the degree of reaction with gas components contained in odors emitted from microorganisms varies depending on the gas sensor (110, 120). This reaction pattern varies depending on the microorganism. These reaction patterns can be used to identify the microbial species generating the odor.

[0054] In order to find a combination of gas sensors (110, 120) that can effectively distinguish microbial species, the research team configured the 12 types of gas sensors (110, 120) in various combinations and analyzed the data measured by each gas sensor (110, 120).

[0055] Principle component analysis (PCA) was utilized for data analysis. Because the types of gas components contained in microbial odors and the types of gas components measured by multiple gas sensors are diverse, PCA was utilized to reduce high-dimensional data to low-dimensional data while retaining as much information as possible from the data measured by gas sensors (110, 120), and visualize the data.

[0056] Through these experiments, we derived a combination of nine gas sensors optimized for analyzing odors generated from microorganisms, as shown in Table 2 below.

[0057] GroupGas sensorGas compositionGas concentration rangeⅠ1Air pollutantsEthanol 1-30ppm2Air pollutantsEthanol 1-10ppm3Alcohol solvent vapors50-5000ppm4Butane, propane1-25% LELⅡ1LPG, smoke, alcohol, propane, hydrogen, methane, carbon monoxide200-10000ppm2CH4, natural gas200-10000ppm3carbon monoxide20-2000ppm4hydrogen100-10000ppm5ammonia, benzene, alcohol10-300ppm (NH3, alcohol), 10-1000ppm(benzene)

[0058] Figure 3 is a graph showing the results of classifying gas components by a combination of gas sensors according to Table 1 and the results of classifying gas components by a combination of gas sensors according to Table 2.

[0059] Referring to Figure 3, it can be confirmed that the result of principal component analysis, the result (B) of classifying gas components by a combination of gas sensors according to Table 2, shows a higher classification score than the result (A) of classifying gas components by a combination of gas sensors according to Table 1.

[0060] The combination of gas sensors according to Table 2 can effectively detect odors generated by microorganisms. Specifically, the combination can effectively detect gas components contained in odors generated by any one of the microorganisms Staphylococcus epidermidis, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Candida albicans, Candida glabrata, and Candida parapsilosis. The above microorganisms are bacteria that can easily infect newborns in a neonatal incubator.

[0061] Additionally, the combination of gas sensors according to Table 2 can effectively detect gas components contained in odors generated by any one of the microorganisms Acinetobacter baumannii, Staphylococcus aureus, Klebsiella pneumoniae, and Pseudomonas aeruginosa. The above microorganisms are causative agents of pneumonia, which patients wearing ventilators can easily become infected with.

[0062] The infection analysis unit (200) learns the response pattern of the gas sensor (110, 120) to the gas component using an artificial intelligence algorithm, and determines the type of microorganism from the response pattern of the gas sensor (110, 120). Any one of XGBoost (Extreme Gradient Boosting), SVM (Support Vector Machine), and Decision Tree may be used as the artificial intelligence algorithm.

[0063] XGBoost is an ensemble technique that combines multiple decision trees, sequentially correcting model errors to build a model. XGBoost supports both regression and classification problems and boasts superior performance.

[0064] SVM is a supervised learning model that finds a decision boundary that is as far away from two classes as possible. It is a classifier that determines which side the data belongs to through the decision boundary. The data classification uses the Radial Basis Function (RBF) kernel to specify a nonlinear boundary, and the model can be designed with the optimal parameters C = 50 and gamma = 1 by performing a grid search. Here, C is the influence that each data point has on the model. The smaller the influence, the more restricted the model becomes, such as a linear model. Gamma is the sensitivity to each data point. The smaller the gamma value, the slower the decision boundary changes, reducing the complexity of the model.

[0065] Decision Tree is a supervised learning model that classifies data by dividing it into two variable areas according to specific criteria (nodes). After the division, learning is performed in a direction that increases the purity of each area and minimizes impurity / uncertainty.

[0066] Figure 4 is a confusion matrix showing the accuracy of gas component analysis learned by the three artificial intelligence algorithms described above.

[0067] Referring to Figure 4, when the predicted labels were designated as Ethyl Alcohol, Cyclohexanone, 3-octanone, Indole, Isobutyl acetate, Formaldehyde, 2-Undecanond, Acetic acid, and 3-methyl-1-butanol, and as a result of training, XGBoost showed the highest accuracy compared to the true label. XGBoost showed an accuracy of 97%, SVM showed an accuracy of 96%, and Decision Tree showed an accuracy of 93%.

[0068]

[0069] Figure 5 is a graph measuring the reaction between gas sensors and odors emitted from microorganisms according to Table 2. The horizontal axis represents the measurement time, and the vertical axis represents the resistance value of the gas sensor. The experiment measured the odors emitted from the microorganisms after culturing the microorganisms for a certain period of time. In this experiment, the microorganisms were cultured for 1 hour and 20 minutes, and the reaction between the odors emitted from the microorganisms and the gas sensor was measured for 200 seconds.

[0070] Referring to Figure 5, we can see that each gas sensor reacts differently to microbial odors. For the first gas sensor, a significant change occurs within a short period of time, followed by a rapid decrease in response. For the second gas sensor, the initial response is rapid, followed by a slow decrease. Thus, each gas sensor exhibits a different response to microbial odors, and these response patterns can be used to identify microbial species.

[0071]

[0072] Figures 6 to 12 are graphs showing the measurement of odors generated by various microorganisms and the reactions of gas sensors.

[0073] In the graphs of FIGS. 6 to 12, (A) is a graph measuring the response of a gas sensor for 200 seconds after culturing microorganisms for 1 hour and 20 minutes, (B) is a graph measuring the response of a gas sensor for 200 seconds after culturing microorganisms for 5 hours and 20 minutes, (C) is a graph measuring the response of a gas sensor for 200 seconds after culturing for 9 hours and 20 minutes, and (D) is a graph measuring the response of a gas sensor for 200 seconds after culturing for 14 hours and 40 minutes.

[0074] Figure 6 shows the odor generated from C. albicans and the response of gas sensors, Figure 7 shows the odor generated from C. glabrata and the response of gas sensors, Figure 8 shows the odor generated from C. parapsilosis and the response of gas sensors, Figure 9 shows the odor generated from E. coli and the response of gas sensors, Figure 10 shows the odor generated from K. pneumoniae and the response of gas sensors, Figure 11 shows the odor generated from S. aureus and the response of gas sensors, and Figure 12 shows the odor generated from S. epidermidis and the response of gas sensors.

[0075] Referring to Figures 6 to 12, it can be seen that the odors emitted by microorganisms contain different types of gas components for each microorganism, and thus the response patterns of the gas sensors differ for each microorganism. The infection analysis unit can identify the species of microorganism through these changes in the response patterns of the gas sensors.

[0076]

[0077] FIG. 13 is a diagram showing an infection monitoring electronic nose system according to one embodiment of the present invention.

[0078] Referring to Fig. 13, the infection monitoring electronic nose system (10) monitors whether a newborn is infected with microorganisms in an incubator for newborns (20). The infection monitoring electronic nose system (10) monitors whether the newborn is infected with Staphylococcus epidermidis, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Candida albicans, Candida glabrata, and Candida parapsilosis, which are major infectious pathogens in newborns.

[0079] A sensor array (100) is provided within an incubator (20) for newborns, and a plurality of gas sensors (110, 120) react with gas components emitted through the newborn's breathing.

[0080] The infection analysis unit (200) analyzes the response patterns of gas sensors (110, 120) to determine whether the newborn is infected, the type of infected bacteria, and the degree of infection.

[0081]

[0082] FIG. 14 is a drawing showing an infection monitoring electronic nose system according to another embodiment of the present invention, and FIG. 15 is a drawing showing a detailed configuration of the infection monitoring electronic nose system.

[0083] Referring to FIGS. 14 and 15, the infection monitoring electronic nose system (10) further includes an exhaust pipe (300) and a capture unit (400).

[0084] The exhaust pipe (300) is connected to the neonatal incubator (20) and provides a passage through which the internal air of the neonatal incubator (20) is exhausted to the outside by driving the circulation pump.

[0085] The capture unit (400) is connected to the exhaust pipe (300) and captures odor particles contained in the air exhausted to the outside through the exhaust pipe (300). The capture unit (400) includes a housing (410), a filter (420), and a heater (430).

[0086] The housing (410) is connected to the exhaust pipe (300), and a space is formed inside.

[0087] The filter (420) is located within the housing (410) and captures odor particles contained in the air introduced into the housing (410). An activated carbon filter may be used as the filter (420).

[0088] The heater (430) heats the inside of the housing (410) to evaporate odor particles captured in the filter (420).

[0089] The sensor array (100) is arranged within the housing, and a plurality of gas sensors (110, 120) react with odor particles evaporated from the filter (420).

[0090] The infection analysis unit (200) analyzes the response patterns of gas sensors (110, 120) to determine whether the newborn is infected, the type of infected bacteria, and the degree of infection.

[0091] The amount of odor particles contained in the air emitted during a newborn's breathing process is extremely small compared to the size of the newborn incubator (20). Therefore, when the sensor array (100) is positioned within the newborn incubator (20), it is not easy for the gas sensors (110, 120) to quickly react to the odor particles. If the odor particles are not quickly monitored, the problem of the newborn's infection being discovered after a considerable amount of time has passed may occur.

[0092] In an embodiment of the present invention, the internal air of a neonatal incubator (20) is exhausted through an exhaust pipe (300), and odor particles contained in the air during the exhaust process are captured by a filter (420). Since a sufficient amount of odor particles are captured by the filter (420), the odor particles can react with the gas sensors (110, 120) in sufficient amounts when evaporated from the filter (420). As a result, the response pattern of the gas sensors (110, 120) can be quickly and accurately obtained.

[0093]

[0094] FIG. 16 is a diagram showing an infection monitoring electronic nose system according to another embodiment of the present invention.

[0095] Referring to Fig. 16, the sensor array (100) can be mounted on the exhaust port of the artificial respirator (30). The gas sensors (110, 120) react with odor particles emitted during the breathing process of a patient wearing the artificial respirator.

[0096] The infection analysis unit (200) analyzes the response patterns of the gas sensors (110, 120) to determine whether the patient has pneumonia, the type of infected bacteria, and the degree of infection. The infection analysis unit (200) can determine whether the patient is infected with Acinetobacter baumannii, Staphylococcus aureus, Klebsiella pneumoniae, or Pseudomonas aeruginosa. Depending on whether the respirator (30) is infected, the medical staff can determine when to replace the disposable tube, expiratory valve, and attached humidifier.

[0097]

[0098] Figure 17 is a drawing showing an infection monitoring method according to an embodiment of the present invention.

[0099] Referring to Fig. 17, the infection monitoring method includes a step of measuring odor using gas sensors (S100) and a step of identifying microbial species through reaction pattern analysis (S200).

[0100] The odor measurement step (S100) using gas sensors measures the reactions of multiple gas sensors with gas components contained in odors emitted from microorganisms. Specifically, reaction data with gas components contained in odors emitted from microorganisms is generated through gas sensors provided in a sensor array. The gas sensors may be provided as a combination of the gas sensors described in Table 1 or Table 2.

[0101] In an embodiment, the gas sensors may generate response data by reflecting gas components included in an odor generated by any one of the microorganisms Staphylococcus epidermidis, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Candida albicans, Candida glabrata, and Candida parapsilosis.

[0102] In another embodiment, the gas sensors may generate response data by reflecting gas components contained in an odor generated by any one of the microorganisms Acinetobacter baumannii, Staphylococcus aureus, Klebsiella pneumoniae, and Pseudomonas aeruginosa.

[0103] The step (S200) of identifying microbial species through response pattern analysis analyzes the response patterns of the gas sensors using a pre-trained deep learning algorithm to identify the species of the microbial species. The pre-trained deep learning algorithm may be any one of XGBoost (Extreme Gradient Boosting), SVM (Support Vector Machine), and Decision Tree. Since the odor generated by microorganisms contains different types of gas components for each microorganism, the response patterns of the gas sensors are different for each microorganism. The pre-trained deep learning algorithm can identify the species of microorganisms through these response patterns.

[0104]

[0105] While the present invention has been described in detail using preferred embodiments, the scope of the present invention is not limited to the specific embodiments described above, and should be construed in accordance with the appended claims. Furthermore, those skilled in the art will appreciate that numerous modifications and variations are possible without departing from the scope of the present invention.

[0106]

[0107] An electronic nose system according to an embodiment of the present invention can be used to monitor for pathogen infection.

Claims

1. A sensor array provided with multiple gas sensors that individually detect gas components contained in odors generated from microorganisms; and An infection monitoring electronic nose system including an infection analysis unit that analyzes the response patterns of the gas sensors using a pre-trained deep learning algorithm to determine the species of the microorganism.

2. In paragraph 1, An infection monitoring electronic nose system in which at least two of the above gas sensors detect the same gas component but have different detection concentration ranges.

3. In paragraph 1, The above gas sensors A first group of gas sensors detecting at least one of ethanol, butane, propane, and alcohol solvent vapors; and An infection monitoring electronic nose system, classified into a second group, comprising gas sensors that detect at least one of LPG, smoke, alcohol, propane, hydrogen, methane, carbon monoxide, CH4, natural gas, carbon monoxide, hydrogen, ammonia, benzene, and alcohol.

4. In paragraph 1, The above microbial species include Staphylococcus epidermidis, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Candida albicans, Candida glabrata, and Candida parapsilosis.

5. In paragraph 1, The above microbial species include Acinetobacter baumannii, Staphylococcus aureus, Klebsiella pneumoniae, and Pseudomonas aeruginosa, an infection monitoring electronic nose system.

6. In paragraph 1, The above sensor array is an infection monitoring electronic nose system installed in an incubator for newborns.

7. In paragraph 1, An exhaust pipe connected to a neonatal incubator and through which the internal air of the neonatal incubator is exhausted; and It is connected to the exhaust pipe and further includes a capturing unit that captures odor particles contained in the air. The above sensor array is an infection monitoring electronic nose system located within the capture unit.

8. In paragraph 7, The above capturing part A housing connected to the above exhaust pipe and having a space formed inside; a filter located within the housing and capturing the odor particles; and An infection monitoring electronic nose system comprising a heater for heating the interior of the housing.

9. In paragraph 1, The above sensor array is an infection monitoring electronic nose system mounted on the outlet of a ventilator.

10. A step of measuring the gas components contained in the odor generated from microorganisms and the reactions of multiple gas sensors; A method for monitoring microorganisms using an electronic nose system, comprising a step of analyzing the response patterns of the gas sensors using a pre-learned deep learning algorithm to determine the species of the microorganisms.

Images