Intelligent system for real-time detection and analysis of volatile organic compounds (VOCS) contained in exhaled breath

Abstract

Intelligent system for real-time detection and analysis of volatile organic compounds (VOCs) contained in exhaled breath of a patient, able to detect the imprint of the respiratory volaboloma, characterized in comprising : - a nasal electronic device (100) comprising a nasal mask (101) having at least at least one inlet opening (101a) for air and at least one outlet opening (101b) for exhaled breath; - at least one electrochemical molecule detector (102) comprising at least one semiconductor sensor able to detect volatile organic compounds, connected to the outlet opening (101b) of the nasal mask (101); - at least one processing unit able to analyze data detected by the at least one semiconductor sensor in real time and to control respiratory cycles while measuring; - an integrated artificial intelligence module, able to identify patterns in data obtained from the exhaled breath, and to correlate said patterns with suspected physiological conditions; wherein said nasal electronic device (100) comprises : - a conveyor tube (104) bent at an angle a connected to, or integrated with, the nasal mask (101) so that in correspondence with one end said conveyor tube (104) is connected to the outlet opening (101b), able to ease the passage of VOCs inside it.

Description

[0001] DESCRIPTION

[0002] "Intelligent system for real-time detection and analysis of volatile organic compounds (VOCs) contained in exhaled breath"

[0003] ■*: * *

[0004] The present invention relates to the technical field of real-time data analysis and intelligent electronic devices for achieving the identification of the respiratory trace .

[0005] The present invention pertains to the technical field of a new biomedical science called Volabolomics , which, together with Volatolomics engineering, constitutes Breathomics . In particular, the biomedical invention focuses on the complex of the omics of exhaled volatile metabolites , which represent the physiological state of an individual . Consequently, the field of application relates to the diagnosis of pathologies .

[0006] The history of medicine, even prior to Hippocrates , shows how the "odor" of respiration was evaluated in order to seek clues of diseases . For example, a sweet odor is characteristic of diabetes ; a fishy odor is typical of renal insufficiency; a putrid odor is indicative of hepatic and digestive disorders . In the 1700s , Lavoisier discovered the presence in human breath of one of the volatile products of metabolism, namely carbon dioxide, which is measurable and correlated with individual physiology . In 1971 , Nobel Prize laureate Linus Pauling identified in human breath hundreds of small organic molecules with low molecular weight, low boiling point and / or high vapor pressure, known as volatile organic compounds (VOCs) , which were subsequently found to originate from physiological states or to result from pathological conditions .

[0007] Nowadays , it is known that breath is composed almost entirely of a few main compounds , while the remaining 1% contains a mixture of several thousand molecules , some of which are present in all individuals . The residual presence of these molecules can be used as a biological fingerprint for pathology screening . Numerous studies have demonstrated significant alterations in the composition of VOCs in various diseases , including different types of cancer, metabolic conditions , neurodegenerative and infectious diseases , and environmentally induced disorders [ references : Mazzatenta; Broza, Y . Y . & Haick, H . Nanomedicine (Future Medicine ) 8 , 785-806 (2013) ; Nakhleh, M. , Broza, Y . Y . , Haick, H . & 2014 , Nanomedicine (Future Medicine) 9 , 1991 -2002 ] .

[0008] The new clinical practice in medical diagnosis requires a low-cost, objective and replicable system, particularly one that is available in real time . Moreover, since all pathologies are characterized by a specific profile of physiological VOCs, it is important, in order to facilitate diagnostic practices, to obtain a characteristic fingerprint of pathological states and diseases [references: Mazzatenta; Vishinkin, R.; Haick, H., Small 2015, 11, 6142-6164; Gouma, P.; Stanacevic, M., Procedia Eng. 2011, 25, 1557-1560; Gouma, P. I.; Kalyana Sundaram, K., Appl . Phys. Lett. 2008, 93, 244102; Zhou, Y . ; Yu, G.; Chang, F.; Hu, B.; Zhong, C. J. Anal. Chim. Acta 2012, 757, 56-62] .

[0009] At present, there exists the technical problem of identifying the entire "spectrum" of VOCs in the presence of interfering or background gaseous species. In fact, despite advancements in instruments for the detection of VOCs through highly selective detection methods, including those based on nanotechnology, it remains highly uncertain to achieve an effective system. Even though specific identification may be rendered possible under limited circumstances, most diseases cannot currently be reliably identified by means of individual and specific VOCs.

[0010] Another critical limitation is the problem of synthesizing highly selective nano-selective materials for each VOC [references: Mazzatenta; Haick, H.; Broza, Y. Y.; Mochalski, P.; Ruzsanyi, V.; Amann, A., Chem. Soc. Rev. 2014, 43, 1423-1449; Konvalina, G.; Haick, H., Acc. Chem. Res. 2014, 47, 66-76] . The use of the so-called "electronic nose" technology, referring to devices capable of detecting and differentiating the VOCs contained in air by means of dedicated sensors, enables real-time determination of the exhaled physiological fingerprint of a subj ect . The respiratory fingerprint may represent a true biopsy characteristic of the health state or pathological condition of an individual , when adequately detected . Consequently, the electronic nose has proven to be the most effective diagnostic aid for identifying the disease fingerprint .

[0011] A known device for breath data analysis is described in the patent text WO 03 / 064994 , wherein the problem of detecting low-concentration components in breath is highlighted, and a solution is proposed in the form of an optimized system for preserving the sampled breath that outperformed existing technologies in detecting VOCs .

[0012] A universal system for breath analysis capable of performing differential diagnostic analysis among diseases, rather than with respect to a single pathology or class of pathologies, is discussed in US2019 / 0271685A1 .

[0013] According to such inventions, a portion of the exhaled breath, prior to exposure to the VOC detection sensors , is sampled, condensed and / or reduced in order to capture the compounds to be measured in a separate analysis device by means of gas chromatography (GC) or gas chromatography-mass spectrometry (GC-MS) .

[0014] The invention according to WO 2021 / 236653 relates to a portable real-time breath analysis device, for detecting, through one or more sensors , a series of VOCs depending on the required diagnosis .

[0015] In the publication "Chemical Sensor and Electronic Nose Based on 1-D Metal Oxide Nanostructures , " IEEE Transactions on Nanotechnology, IEEE Service Center, Piscataway, NJ, US, vol . 7 no . 6, pp . 668 - 682 , a device for detecting compounds in human breath, suitable for use exclusively at laboratory scale, is described.

[0016] The currently available instruments and methods , including those cited above, are essentially based on three archaic concepts : the collection of a breath sample , its subsequent analysis , and the search for a biomarker . Unfortunately, breath sample collection presents obvious limitations , including sample modification due to instantaneous oxidation of many molecules , or reaction of the molecules with collection bags , and contamination of the sample with molecules derived from the manufacturing of the same bags . Collection columns are somewhat better, but even these, owing to their chemical nature, present strong limitations due to the partiality of the chemical families that they can bind and retain . Consequently, the operator is compelled to make a priori choices regarding which molecules to collect, thereby limiting the measurement of VOCs .

[0017] Furthermore, approaches based on detecting a single molecule or a few molecules characteristic of a disease, as in WO 2021 / 236653, cannot address the problem of preventive diagnosis of disease, except in a partial manner . That is , while the characterization of a disease might naively suggest one or a few dominant metabolites , as in the case of diabetes producing acetone, in reality this is already too late for biomedical purposes , since the physiology already denotes a markedly pathological state .

[0018] The patent application US2021219867A1 describes a method and a system for the real-time analysis and detection of analytes exhaled by a patient . The method and system detect one or more analytes in the breath and provide alerts with reference to discrepancies in the administered drugs , verify the administered drugs , and manage the pharmacological treatment of a patient .

[0019] The patent application US2015105683A1 describes a portable device for the analysis of breath samples comprising a mouthpiece, an ambient air filter for filtering the inhaled air, and an analysis compartment provided with a chemical trace detector . However, the known systems do not allow the acquisition of the fingerprint, i . e . , the complete and unique imprint of the respiratory volaboloma .

[0020] Furthermore, the known systems do not allow the elimination or strong limitation of contaminations and errors associated with the interaction with the air present in the environment .

[0021] The purpose of the present invention to provide a system capable of identifying and detecting a breath imprint and processing it in such a way that it results not only largely anticipatory of a state of disease, but also as focused on detecting a physiological alteration that is as complete and well defined as possible .

[0022] In particular, a purpose of the present invention is to provide a system capable of investigating, by means of non- invasive real-time collection and analysis of exhaled breath, through an electronic nose and a specific system and which allows, thanks to artificial intelligence, to highlight patterns of suspected physiological anomalies such as to suggest the specific diagnosis of a future disease .

[0023] In this sense, it is to be considered patterns anything that can be used to provide a diagnosis , through the classification of the detected physiological alteration . For the evaluation of the physiological alteration, reference is made both to the concentration and type of substances detected, and to other physiological , biological and related to the medical record of the patient data .

[0024] The purpose of the present patent is to provide a portable diagnostic system capable of solving the technical problem of detecting and analyzing in real time , in a non- invasive manner, without contamination or loss of the sample, the respiratory imprints emitted in the form of complex mixtures of volatile organic compounds (VOCs ) , and of interpreting their characteristic compositions indicative of the presence of pathologies and, according to the analysis of the characteristic variables of the detected respiratory imprint, providing a preventive diagnosis of the health state of a subj ect .

[0025] A further purpose of the present invention is to provide an efficient system and a diagnostic tool capable of anticipating a physiological alteration related to a more advanced pathological state .

[0026] According to the present invention, an intelligent system for real-time detection and analysis of volatile organic compounds (VOCs) contained in breath is realized, as def ined in claim 1 .

[0027] For a better understanding of the present invention, a preferred embodiment is now described, purely by way of nonlimiting example, with reference to the attached drawings , in which : - figure 1 shows a schematic view of an intelligent system for the real-time detection and analysis of volatile organic compounds contained in exhaled breath, according to the invention;

[0028] - figure 2 shows a detailed view of the system in which the positioning of sensors in correspondence with the outlet of a conveying tube is highlighted, according to the invention;

[0029] - figure 3 shows a schematic side view of the system according to the invention .

[0030] With particular reference to Figure 1 , the intelligent system 100 for the real -time detection and analysis of volatile organic compounds (VOCs ) contained in exhaled breath is shown, according to the invention .

[0031] The system according to the present invention comprises a nasal electronic device 100 , belonging to the field of "electronic noses" as described above, comprising :

[0032] - a nasal mask 101 provided with at least one inlet opening 101a and at least one outlet opening 101b, for allowing a person to inhale and exhale through the nasal mask 101 ;

[0033] - at least one molecule detector 102 , comprising at least one semiconductor sensor, or an array of electrochemical sensors , for detecting volatile organic compounds (VOCs ) , connected to the outlet opening 101b of the nasal mask 101 .

[0034] In particular, the molecule detector 102 operates with one or more sensors that detect the presence of VOCs in the exhaled breath in real time .

[0035] According to one aspect of the invention, the molecule detector 102 comprises :

[0036] - at least one semiconductor sensor;

[0037] - at least one heating element able to heat the at least one semiconductor sensor;

[0038] - at least one processor 103 , or electronic board, configured to control the heating element; and

[0039] - a detection circuit configured to detect the variation in resistance of the at least one semiconductor sensor, said variation being determined at least in part by the presence of at least one chemical signal reacting with the at least one semiconductor sensor .

[0040] The processor 103 is configured to receive signals from chemical or biological sensors (for example, MOS sensors able to detect VOCs ) and to convert such signals into digital electrical signals able to be processed .

[0041] A processing unit, which may include a further microprocessor or a microcontroller, receives the digital electrical signals and executes preliminary analysis algorithms , filters the data, and prepares these data for further processing through artificial intelligence (Al ) models .

[0042] The electronic board 103 is further configured to regulate and distribute energy to the various components of the device, ensuring that the system operates correctly and ef ficiently .

[0043] According to one aspect of the invention, the electronic board or processor 103 is connected with a user interface, which may be an LCD display, a mobile application, or a web interface . The electronic board 103 thus transmits the processed data and the results of the analyses to the user interface .

[0044] According to another aspect of the invention, the nasal mask 101 of the nasal electronic device 100 allows a person wearing the device itself to breathe normally without the need to blow into the molecule detector 102 , thereby recording the pulmonary gases exhaled through normal respiratory activity, which is representative of the metabolic activity under physiological conditions .

[0045] Advantageously, the analysis of the exhaled air is immediate and in situ, eliminating the need to collect, preserve, and subsequently examine the breath sample, thereby reducing the risk of degradation, reactions , or contamination of the sample during storage . As shown in Figure 1 , the nasal mask 101 comprises a wearable portion in correspondence with the nose of a user or patient, into which air enters through the inlet opening 101a .

[0046] The system further comprises a conveying tube 104 that i s bent, or folded, at an angle 'ex' , preferably comprised between 110 ° and 90 ° . A first end of the conveying tube 104 is connected to the wearable portion of the nasal mask 101 . In correspondence with a second end of the conveying tube 104 is present a fixing of a sensor array, molecule detector 102 or at least one sensor, or of the electronic board 103 .

[0047] According to one aspect of the invention, the conveyor tube 104 is connected to, or integrated with, the nasal mask 101 so that in correspondence with one end, the conveyor tube 104 itself is connected to the outlet opening 101b, able to facilitate the passage of VOCs inside the same outlet opening 101b .

[0048] The electronic board or processor 103 is preferably arranged in correspondence with the sensors 102 and is connectable, by means of a USB connection, to a user interface .

[0049] According to one aspect of the invention, the system further comprises an integrated artificial intelligence module configured to identify patterns present in the data of the exhaled breath that may indicate suspected physiological conditions .

[0050] According to another aspect of the invention, the system is configured to monitor, during measurement, the respiratory cycles performed by a patient through the nasal mask 101 .

[0051] According to one aspect of the invention, the system further comprises a user interface configured to display the results of the analysis of the exhaled breath in real time .

[0052] In use, by means of a database assembled with patterns characterizing specific pathologies or diseases, the response patterns obtained from the sensor outputs are compiled, and corresponding outputs are generated . These outputs can be compared with the patterns characterizing known pathologies, which may be identified and differentiated within a plurality of diseases starting from a single exhalation of breath by the patient .

[0053] According to one embodiment of the present invention, by using the artificial intelligence module, the composition of the ambient air in which the diagnosis is performed is measured and subtracted from the data related to the exhaled breath of the patient .

[0054] In this way, advantageously, the system avoids the need to purify the air or to use cylinders containing clean air . Advantageously, the nasal mask 101 allows a person to breathe normally without mechanical effort, thereby avoiding physiological bias associated with forced breathing into a tube .

[0055] Advantageously, the system, and in particular the molecule detector 102 , has a compact and portable form, is easy to wear and non-invasive, thus allowing an accurate and reproducible real-time analysis of breath .

[0056] Said detector 102 determines the presence of different volatile organic compounds (VOCs ) produced in the body in the exhaled air . The detector 102 is capable of detecting VOCs and carbon dioxide at least through a redox reaction at high temperature, resulting in a characteristic variation, detectable by the sensor .

[0057] Preferably, the temperature at which the redox reaction occurs is comprised between 40 °C and 100 °C .

[0058] Such a feature proves technically advantageous since the high temperature of the sensors ensures that the water vapor contained in the exhaled breath does not fog the sensors , while the VOCs present in the exhaled breath can be directly measured .

[0059] Indeed, due to the high temperature at which the at least one sensor operates, the water vapor present in the exhaled air does not affect the measurements , and no additional dehumidification or condensation operations are required, so that, therefore, the nasal electronic device 100 does not require additional supports such as pre concentrators or analysis chambers .

[0060] The redox reaction takes place through the following steps :

[0061] 1 . Adsorption :

[0062] - The VOCs and CO2are adsorbed onto the surface of the sensor, which is made of a semiconductor material , or of metal oxides such as tin oxide (SnO2) and zinc oxide ( ZnO) .

[0063] 2 . Oxidation-Reduction Reaction :

[0064] - The VOCs undergo an oxidation process on the surface of the sensor . In the presence of oxygen, the organic compounds are oxidized, releasing electrons .

[0065] 3 . Change in Conductivity :

[0066] - The oxidation process of the VOCs on the sensor surface causes the release of electrons , which alters the electrical conductivity of the semiconductor .

[0067] - Under normal conditions, the oxygen adsorbed on the sensor surface captures electrons from the semiconductor material, thereby increasing its resistance . When the VOCs are oxidi zed release electrons that reduce the resistance of the sensor .

[0068] 4 . Detection of the Change :

[0069] - The sensor detects the variations in conductivity caused by the interaction with the VOCs or other gases present . These changes are directly correlated to the concentration of the detected chemical compounds .

[0070] The sensor response is then converted into an electrical signal proportional to the concentration of the VOCs or CO2 in the exhaled air .

[0071] In one embodiment, the nasal electronic device 100 comprises a disposable neutral / mechanical filter configured to protect the sensors from particulate matter, including bacteria .

[0072] Preferably, the filter is located at the outlet section of the conveyor tube 104 , i . e . , in correspondence with the point where the air flows into the conveyor tube 104 and enters into the molecule detector 102 .

[0073] According to one aspect of the invention, the nasal electronic device 100 is portable and designed to be fastened to the head of the user undergoing the test .

[0074] According to one aspect of the invention, the nasal electronic device 100 is connected to a computer via cable, Bluetooth, or Wi-Fi .

[0075] According to one aspect of the invention, the nasal mask 101 may further comprise a support made of soft fabric with Velcro, configured to be worn on the patient' s head, thereby supporting the nasal electronic device 100 thanks to its low weight . In one embodiment of the system according to the present invention, a pyramid-shaped container, not shown in the figures, is provided . Said container houses the nasal mask 101 , made of neutral and hypoallergenic material , connected to a conveyor tube 104 that conveys the exhaled air to at least one semiconductor sensor, together with the other components previously described .

[0076] The normal alternation of inhalation and exhalation cycles allows the simultaneous acquisition of the breath imprint and physiological information, such as breath amplitude and frequency, which are essential for monitoring volatile compounds and controlling the exhaled air, thereby avoiding erroneous acquisitions and verifying the accuracy of real-time measurements .

[0077] According to one aspect of the invention, the artificial intelligence module employs machine learning algorithms , including at least deep neural networks (DNNs ) , convolutional neural networks (CNNs ) , support vector machines (SVMs ) , and tree-based algorithms such as Random Forest and XGBoost, in order to improve accuracy in the classification of suspected physiological conditions .

[0078] According to one aspect of the invention, the artificial intelligence module is programmed to analyze data related to VOCs , in order to detect patterns that may indicate the onset of future diseases . The system according to the present invention is suitable for preventive diagnosis of diseases including : neoplastic diseases ; neurovegetative diseases ; genetic diseases ; dental diseases ; endocrine and metabolic diseases ; otolaryngological diseases ; respiratory diseases ; and environmentally induced diseases .

[0079] According to one aspect of the invention, the processing unit is a microcontroller compatible with TinyML, allowing the local execution of machine learning algorithms, thereby reducing latency and energy consumption, while improving patient data privacy .

[0080] According to one aspect of the invention, the system comprises an additional module configured to measure and analyze ambient air and to subtract environmental influences from the exhaled breath data, in order to improve diagnostic accuracy .

[0081] According to one aspect of the invention, the artificial intelligence module is configured to continuously monitor inhalation and exhalation cycles , acquiring physiological information such as breath amplitude and frequency, so as to avoid false acquisitions and to verify the accuracy of the real-time measurements .

[0082] According to one aspect of the invention, the system comprises a pyramid-shaped container configured to house the nasal mask 101 , which is made of neutral and hypoallergenic material and connected to the conveyor tube 104 .

[0083] According to one aspect of the invention, the nasal electronic device 100 comprises a disposable neutral / mechanical filter configured to protect the sensors from particulate matter, including bacteria .

[0084] According to one aspect of the invention, the nasal electronic device 100 is portable and comprises a fastening system for securing to the head of a patient, preferably a support made of fabric with Velcro .

[0085] According to one aspect of the invention, the system comprises a connection system to a computer, via cable, Bluetooth, or Wi-Fi .

[0086] Advantageously, the system according to the present invention provides a portable diagnostic tool capable of detecting and analyzing, in real time , in a non-invasive manner , without contamination or loss of the sample, breath imprints emitted in the form of complex mixtures of volatile organic compounds (VOCs ) , and of interpreting their compositions as characteristic of the presence of pathologies , according to the analysis of the characteristic variables of the detected breath imprint, as well as of providing a preventive diagnosis of the health status of a subj ect . It is finally clear the system, described and illustrated herein, may be subj ect to modifications and variations without departing from the protective scope of the present invention, as defined in the attached claims .

Claims

CLAIMS1. Intelligent system for real-time detection and analysis of volatile organic compounds (VOCs) contained in exhaled breath of a patient, able to detect the imprint of the respiratory volaboloma, characterized in comprising:- a nasal electronic device (100) comprising a nasal mask (101) having at least at least one inlet opening (101a) for air and at least one outlet opening (101b) for exhaled breath;- at least one electrochemical molecule detector (102) comprising at least one semiconductor sensor able to detect volatile organic compounds, connected to the outlet opening (101b) of the nasal mask (101) ;- at least one processing unit able to analyze data detected by the at least one semiconductor sensor in real time and to control respiratory cycles while measuring;- an integrated artificial intelligence module, able to identify patterns in data obtained from the exhaled breath, and to correlate said patterns with suspected physiological conditions; wherein said nasal electronic device (100) comprises:- a conveyor tube (104) bent at an angle a connected to, or integrated with, the nasal mask (101) so that in correspondence with one end said conveyor tube (104) isconnected to the outlet opening ( 101b) , able to ease the passage of VOCs inside said outlet opening ( 101b) .2 . System according to claim 1 , characterized in that said electrochemical molecule detector (102 ) comprises : at least one semiconductor sensor; at least one heating element , configured to heat the at least one semiconductor sensor; at least one processor ( 103 ) configured to control the heating element; a detection circuit configured to detect changes in the resistance of the semiconductor sensor, determined at least in part by the presence of at least one chemical signal reacting with the semiconductor sensor; with said processor ( 103 ) able to receive signals from the at least one semiconductor sensor, of electrochemical type, and to convert said signals into digital electrical signals able to be processed .3 . System according to claim 1 , characterized in that the electrochemical molecule detector ( 102 ) is configured to determine the presence of various VOCs and carbon dioxide in the exhaled breath exiting from the outlet opening ( 101b) through a high temperature oxidation-reduction reaction, resulting in a characteristic change able to detected by the sensor .4 . System according to claim 1, characterized in comprising a pyramid-shaped case, able to contain the nasal mask (101) made of neutral and hypoallergenic material, connected to the conveyor tube (104) .

5. System according to claim 1, characterized in that said nasal electronic device (100) comprises a disposable neutral / mechanical filter able to protect the sensors from particulate matter, including bacteria.

6. System according to claim 1, characterized in that said nasal electronic device (100) is portable and comprises an attachment system to a head of a patient.

7. System according to claim 1, characterized in comprising a system for connecting to a computer, through cable, Bluetooth or Wi-Fi.

8. System according to claim 1, characterized in that the processing unit is a TinyML-compliant microcontroller, allowing a local execution of machine learning algorithms, reducing latency, power consumption and improving the privacy of the data concerning the patient.9 . System according to claim 1 , characterized in that the processing unit receives the digital electrical signals, converted by said processor ( 103) , and executes preliminary analysis algorithms, filters the data and prepares said data for further processing by means of artificial intelligence models .10 . System according to claim 1 , characterized in comprising a further module configured to measure and analyze ambient air and subtract environmental influences from data concerning exhaled breath, in order to improve the accuracy of the diagnosis .11 . System according to claim 1 , characterized in comprising a user interface, consisting of an LCD display, a mobile app or a web interface, the processor ( 103 ) being connected to the user interface and being able to transmit the processed data and results of the analysis to said user interface .

Images