WO2023157017A1

WO2023157017A1 - Method of fabricating a conducting cloth based breath humidity sensor and applications thereof

Info

Publication number: WO2023157017A1
Application number: PCT/IN2023/050142
Authority: WO
Inventors: Pradeep Thalappil; Srikrishnarka Pillalamarri; Madhuri DASI RAAGA; Kanti Jana SOURAV; George Boby
Original assignee: Indian Institute of Technology Madras
Current assignee: Indian Institute of Technology Madras
Priority date: 2022-02-17
Filing date: 2023-02-13
Publication date: 2023-08-24
Anticipated expiration: 2024-08-17

Abstract

The present invention relates to a method of fabricating frugal humidity sensor that can monitor changes in the humidity of the exhaled breath. Sensor comprises of a conducting polymer in-situ grown on a cloth which acts as a substrate. Interdigitated silver electrodes are screen-printed on the conducting cloth and stitched on a mask and transforming it into a smart mask. The sensor is connected to the measurement circuit via conducting thread using silver paste. The sensor is capable of differentiating slow, normal and fast breathing patterns from nose as well as the mouth. An Android-based mobile application is designed to collect the sensor data via the Bluetooth and data analysis is performed using deep learning and time series classification algorithms. The smart mask is comfortably worn while sleeping and performing day-to-day activities.

Description

COMPLETE SPECIFICATION

TITLE OF THE INVENTION

METHOD OF FABRICATING A CONDUCTING CLOTH BASED BREATH HUMIDITY SENSOR AND APPLICATIONS THEREOF

FIELD OF THE INVENTION

The present invention relates to humidity sensor, more specifically relate to the fabrication of conducting cloth based breath humidity sensor for sensing the humidity present in the exhaled air of individuals.

BACKGROUND OF THE INVENTION

This invention relates to the sensing of humidity in the exhaled breath having fast response and recovery time. Monitoring humidity is crucial in many industries, apart from individual usage, such as, food-processing, semi-conductor, packaging, agriculture and medicine. Chronic obstructive pulmonary disorder, asthma and cystic fibrosis are some of the common respiratory disorders that affect million around the globe. A recent study revealed that, -544 million people globally are suffering with respiratory disorders which is a~39.8 % increase was observed from 1990. Human exhaled breath comprises of vast variety of components like, carbon dioxide (CO2), water vapor, nitric oxide (NO) and various volatile organic compounds (VOCs) and monitoring these components would help in understanding the health of that individual. Disease prognosis based on the analysis of the components present in the exhaled breath is promising due to its non-invasive nature. Conventionally, mass spectrometry and Raman spectroscopy were utilized due to their high sensitivity and accuracy, however due to high-cost, requirement of trained personnel and the lack in mobility are some of the main drawbacks for their large-scale deployment in health monitoring. To address some of the issues, wearable sensors have provided some panacea. In this regard, the present invention provides a wearable, fast-responsive and affordable humidity sensor for monitoring the humidity of the exhaled breath.

Humidity sensors typically use metal-oxide, graphene oxide, conducting polymer, supramolecular nanofiber assembly etc., as their sensing material which are then deposited on flexible and rigid substrates. Metal-oxide based sensors are affordable and can be scaled but their response and recovery time is poor. On the other hand, graphene oxide and supramolecular based sensors offer quick response and recover time, they are expensive and cumbersome to manufacture in large scale.

The present invention relates to the fabrication of breath humidity sensor that can monitor the changes in humidity of the exhaled breath. Conducting polymer was in-situ grown on a cloth transforming it to a flexible and wearable cloth. This cloth was later stitched on to a mask, for monitoring the humidity in the exhaled breath continuously. These changes in the humidity are synchronous to the breathing rate, this helped in monitoring and classifying the various breathing patterns observed while working and sleeping.

OBJECTS OF THE INVENTION

An object of the present invention is to provide a conducting cloth based breath humidity sensor for sensing humidity present in the exhaled air of individuals.

Another object of the present invention relates to the fabrication of conducting cloth based breath humidity sensor by in-situ polymerization and screen printed with interdigitated electrode.

Yet another object of the present invention is to provide a smart mask using conducting cloth based breath humidity sensor, wherein the voltage drop across the sensor is measured by a microcontroller to monitor the breathing rate.

Yet another object of the present invention relates to an Android application created to collect the data from the microcontroller and analyze the patterns in breathing using machine learning and deep learning.

SUMMARY OF THE INVENTION

The present invention relates to a conducting cloth based breath humidity sensor for sensing humidity present in the exhaled air of individuals.

In one embodiment, the present invention discloses a method for fabricating the conducting cloth based breath humidity sensor. For the fabrication of a flexible, wearable sensor, cloth was chosen as the substrate. Since, it not only possessed structural integrity against multiple bending, it is highly flexible and if offered least resistance for the passage of air. The method comprises typically, about 6 mL of aniline and 12 g of APS were added to two different beakers containing 80 mL of 1 M aqueous HC1 solution. The non-woven PP cloth of dimension 12 x 12 cm was soaked in water for 12 h to ensure complete wetting. This wet cloth was later immersed in the aniline solution for 2 min, transferred to another beaker containing clean water to remove any unbound aniline molecules. After which, the cloth is soaked in the beaker containing APS for 2 min and finally transferred into another beaker containing clean water. This cycle was repeated 17 times for growth of the poly aniline (PANI) on the mat. The cloth was later dried at 75 °C for 12 h for further use. Due to the growth of polyaniline on the cloth, the blue colored cloth transformed into green color. Interdigitated electrodes were screen printed on the sensor via screen printing with commercial grade Ag paste. Upon drying of the ink, the cloth was cut into individual sensor elements for further analysis.

In other embodiment, the present invention illustrates the sensor is stitched onto a face mask and conducting thread was utilized to connect the sensor to a measurement and filtering circuit. The voltage drop across the sensor is measured by a microcontroller. This system was then used to monitor the breathing rate.

In another embodiment, the present invention relates to an Android application to collect and visualize the data from the microcontroller, and also upload it to cloud for further processing. Using machine learning and deep learning, the collected data was analyzed to detect and classify the patterns in breathing.

Other aspects of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learnt by the practice of the invention

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 Schematic representation of the sensor fabrication processing steps, (a) Polypropylene (PP) mat roll, (b) SILAR procedure where a small piece of PP was immersed in beaker 1 containing 6 mL aniline in 1 M aqueous HC1 solution. From this beaker, the mat was later transferred to beaker (2) containing DI water. From this beaker its further immersed in beaker (3) where, it contains 12 g of APS in 1 M aqueous HC1 solution. Finally, the mat rinsed in beaker (4) containing DI water, the mat color changed from blue to green, (c) The dried conducting cloth was dried and stretched on the frame, the interdigitated electrode template was placed on top of the conducting cloth. Upon which, Ag silver paste was spread as shown in step-3. Finally, the dried sensor array cloth is dried from which one sensor was trimmed. The individual sensor is magnified as shown in the inset, (d) The sensor was then stitched on to a surgical mask, a ruler was placed on the mask intended to depict the overall size of the sensor.

Figure 2 (a) FT-IR and (b) Raman spectrum of untreated PP and PANI@PP cloth.

Figure 3 Contact angle measurements performed on (a) untreated PP mat (b) PANI@PP mat.

Figure 4 Scanning electron micrographs of (a) bare PP, (b,c) PANi coated PP. (Inset scale bar is 10 pm and for (c) scale bar is 1pm). (d) Time-dependent Raman spectra for understanding water desorption, (e) Variation of intensity for the peak at 3002 cm’¹, (f) Nyquist plot to determine the R_s and RCT of the sensor at room temperature, dry air, nasal and oral breath exhalation.

Figure 5 Chronoamperometric analyses for (a) nasal breath, (b) oral breath and inset shows a magnified region for understanding the response and recovery time, (c) Nasal breathing of the sensor which was stitched on a surgical mask.

Figure 6 Chronoamperometric responses while performing the droplet proximity test, (a) Different concentrations of ethanol-water mixture, (c) different concentrations of acetone-water mixture. In (b) and (d), a scatter plot depicting concentration of solution with respect to their drop in current response.

Figure 7 (a) Schematic of the sensor calibration system and (b) sensor calibration at different voltage.

Figure 8 Photographs of (a) volunteer wearing the smart mask and (b) microcontroller set-up. (c) Circuit diagram depicting the components needed while measuring the humidity in the exhaled breath.

Figure 9 Chronovoltametric analyses highlighting the various breathing patterns (a) nasal slow, (b) nasal normal and (c) nasal fast breathing, (d) Oral slow, (e) oral normal and (f) oral fast breathing.

Figure 10 Screen grabs from the graphic user interface (GUI) of the application needed to transfer the data from the microcontroller to the mobile phone and visualize it. Referring to the drawings, the embodiments of the present invention are further described. The figures are not necessarily drawn to scale, and in some instances the drawings have been exaggerated or simplified for illustrative purposes only. One of ordinary skill in the art may appreciate the many possible applications and variations of the present invention based on the following examples of possible embodiments of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The following description is presented to enable any person skilled in the art to make and use the embodiments, and is provided in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the scope of the present invention. Thus, the present invention is not limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

In one embodiment, the invention discloses the in-situ polymerization of conducting polymers such as polyaniline (PANi) or polyprrole on cloths such as polypropylene, cotton, silk, nylon, leather and polyester. Interdigitated electrode is screen printed on this conducting cloth.

In other embodiment, the present invention illustrates the sensor is stitched onto a face mask and conducting thread was utilized to connect the sensor to a Wheatstone bridge differential amplifier to capture its response. The bridge is balanced with resistors of 330 kQ since the resistance of the sensor varies in the range of 330 kQ and 1.3 MQ. A unity feedback Sallen-Key low-pass filter is then used to reduce the effect of noise signals such as ADC noise, contact noise and motion noise on the response of the sensor and to improve the signal-to-noise ratio (SNR). A cut-off frequency of 48.2 Hz was chosen for the low pass filter by observing the peak frequencies present in the Fast Fourier Transform (FFT) spectrum of the raw sensor response. The output of the filter is smoothed using the exponentially weighted moving average (EWMA) function with a smoothing factor a = 0.18 and connected to a microcontroller for subsequent analysis. This system was then used to monitor the breathing rate. In another embodiment, the present invention relates to an Android application, to collect the data from the microcontroller via Bluetooth and visualize the breath cycles. Using deep learning and time series classification algorithms, the collected data was analyzed to detect and classify the patterns in breathing. The data was also uploaded to cloud for building a dataset used to train the classification algorithms.

For the fabrication of a flexible, wearable sensor, cloth was chosen as the substrate. Since, it not only possessed structural integrity against multiple bending, it is highly flexible and if offered least resistance for the passage of air. A schematic representation of the fabrication process is depicted in Figure 1. Typically, about 6 mL of aniline and 12 g of APS were added to two different beakers containing 80 mL of 1 M aqueous HC1 solution. The non-woven PP cloth of dimension 12 x 12 cm was soaked in water for 12 h to ensure complete wetting. This wet cloth was later immersed in the aniline solution for 2 min, transferred to another beaker containing clean water to remove any unbound aniline molecules. After which, the cloth is soaked in the beaker containing APS for 2 min and finally transferred into another beaker containing clean water. This cycle was repeated 17 times for growth of the polyaniline (PANI) on the mat. This methodology of growth on substrates is termed as successive ionic layer adsorption and reaction (SILAR), which has been extensively used for the fabrication of quantum dot sensitized solar cells. The cloth was later dried at 75 °C for 12 h for further use. Due to the growth of polyaniline on the cloth, the blue colored cloth transformed into green color. Interdigitated electrodes were screen printed on the sensor via screen printing with commercial grade Ag paste. Upon drying of the ink, the cloth was cut into individual sensor elements for further analysis.

To confirm the presence of PANi on PP, FTIR and Raman spectroscopic analysis were performed and the corresponding results are shown in Figure 2. FTIR spectrum of untreated and PANi coated PP is shown in Figure 2a, peaks at 2870, 2920 and 2950 cm’¹ correspond to the stretching, asymmetrical stretching and asymmetrical stretching of CH3, respectively. The symmetric CH3 bending has peaks at 1376 and 1456 cm’¹. A small peak at 1167 cm’¹ correspond to the C-H wagging and rocking. These peaks are present in both the bare PP and PANi coated PP as shown in the black and red trace, respectively. The peak at 1299 cm’¹ correspond to the v(C-N) stretching, is a characteristic feature of PANi. The peak at 1577 cm’¹ correspond to the Quinoid ring stretching. Peak at 1302 cm’¹ was assigned to the secondary aromatic amine v(C-N) stretching. Finally a small hump at 3233 cm’¹ was assigned to N-H stretching confirming the presence of PANi.

Raman spectra comparing both bare PP and PANi coated PP mat is shown in Figure 2b, where the black trace represent the bare PP and the red trace represent PANi coated mat. Peak at 834 cm’¹ correspond to the r(CH3) and is present in both bare PP and PANi coated PP. The presence of a peak at 1167 cm’¹ is assigned to v(C-C) stretching. A graphitic disorder peak (D band) was observed for both untreated and treated PP mat at 1345 cm’¹. A peak at 1350 cm’¹ in the PANi treated PP corresponds to the v(C-N⁺) stretching and the peak at 1397 cm’¹ confirms the presence of phenoxazine like units. Presence of a peak at 1457 cm’¹ was assigned to the 5(CH3) asymmetric stretching, this peak is present in both bare and PANi treated sample. A small peak at 1489 cm’¹ was assigned to the quinoid C=N stretching and a peak at 1500 cm’¹ present in the PANi coated sample was assigned to 5(NH) stretching. Peak at 1567 cm’¹ corresponds to the C-C stretching, and a peak at 1606 cm’¹ was assigned to the C=C stretching. Peaks at 2898 and 2966 cm’¹ present in the untreated PP corresponds to the symmetric and asymmetric v(CH3) stretching, respectively.

An ideal sensor must possess affinity towards water molecules and also must be fastdesorbing in nature, for quick response and recovery time. A complex system having both hydrophilic and hydrophobic nature would be necessary. To understand this complex behaviour, water contact angle before and PANi coating was measured and shown in Figure 3. PP mat is hydrophobic in nature as shown in Figure 3a, the contact angle was found to be 143°. This mat upon treatment did slightly reduce its hydrophobic nature as the contact angle has to reduced to 120° as shown in Figure 3b. This decrease is still not that drastic as the system is still hydrophobic in nature. Due to this complex behaviour a quick response and recovery time-scale is observed.

To visualize the morphologic changes occurred by SILAR treatment, both the bare PP and PANi coated PP mats were subjected to Scanning electron microscopy. The resulting micrographs are shown in Figure 4(a-c). Fibers of PP mat were randomly arranged and their surface appeared smooth, the diameter of these PP fibers was ~ 20 pm as observed in Figure 4a. Upon polymerization of PANi on these fibers, the surface of these fibers was coated as PANi wrapped around the fiber. Apart from wrapping the fiber, at vicinity between fiber junction, PANi appeared to have formed bridge-like structure connecting the two PP fiber as seen in Figure 2c. These structures are highly porous in nature which could assist in enhancing the sensitivity of the sensor. To understand the adsorption-desorption dynamics of water molecules from the sensor, a time-dependent Raman spectral analysis on the mat was performed. A mat which was saturated with humidity time-lapse Raman spectra were collected for 18 s at an interval of 1 s as shown in Figure 4c. A broad peak at ~ 3000 cm’¹ was attributed to grouped frequencies of C-H and N-H as previously observed from our previous work. There could be a hydrogen bond between -(NH)⁺ bond of PANi and H2O which assists in proton hopping. The intensity of this peak reduces as the water molecule desorb from the surface of PANi. The intensity of the peak at 3002 cm’¹ as a function of time is presented in Figure 3d. Observed the decrease is instantaneous, within 1 s the intensity was reduced by ~ 27.7 % within 1 s and this further reduced as the time progressed. The intensity was stabilized at the end of ~ 12s, we suspect complete desorption of water molecules from the surface.

Impedance spectroscopy is performed to elucidate the conducting mechanism of the sensing element under various humidity environments. The Nyquist plot obtained is shown in Figure 4f. Dry-air condition was created by purging pure N2 gas over the sensor the corresponding R_ct was 58 kQ and this value decreased to 26 kQ under room temperature. Upon nasal exhalation on the sensor this resistance decreased to 24 kQ and further reduced to -21.9 kQ upon oral exhalation. The surface resistance to Rs and resistance to proton hopping as Rct, as the water molecules adsorb on the surface the overall surface resistance decreases.

Due to this complex behaviour a quick response and recovery time-scale is observed. To study the performance of the sensor under oral and nasal exhaled breath, the sensor was connected to a Palmstat electrochemical workstation and chronoamperometric study was performed. An external voltage of 3 V was given the sensor and its current was measured under different breathing conditions as shown in Figure 5.

The sensor was initially allowed to stabilize to the room temperature and upon which, first nasal exhalation was done. The current rose from 246 to 256 p A as shown in Figure 5a and the nasal inhalation and exhalation was continued for 200 s after which the sensor allowed to stabilize to room temperature condition. It took ~ 13 s for humidity to desorb completely from the sensor. But from the inset in Figure 5 a it clearly evident that, the sensor is versatile in determining inhalation and exhalation. A response time of ~ 1.4 s and recovery time of ~ 2.5 sis observed. This sensor was further subjected to oral exhalation and the chronoamperometric analysis is shown in Figure 5b. Upon first oral exhalation, there are a rise of 17 p A in the current, oral inhalation and exhalation was continued for 200 s. Approximately 65 s were necessary for the water molecules to desorb from the sensor. However, as previously witnessed as in case of the nasal exhalation, the sensor was capable of distinguishing both with a response time of ~ 2 s and a recovery time of ~ 3 s as seen in the inset of Figure 5b.

Exhaled breath comprises many different compounds, analyzing compounds would assist in understanding the well-being of that individual. Acetone and ethanol are some of the commonly known species present in the exhaled breath, especially in diabetics 15 and alcoholics, 16 respectively. A droplet containing different concentration of ethanol-water mixture was brought close to the sensor and the response from the sensor is shown in Figure 6a. As the concentration of ethanol increased, the drop in current was evident and a plot of concentration vs current drop is shown in Figure 6b. Similarly, acetone-water mixture was brought near the sensor and the response is shown in Figure 6c. A plot depicting the current-drop vs concentration of acetone is shown in Figure 6d. Current drop is highest for pure ethanol compared to that of acetone, however, at 5% concentration, there was gain in current in case of ethanol unlike acetone. Unlike sensor’s response towards moisture, where resistivity decreases, under the presence of VOCs, the resistivity increases.

Conducting thread was adhered on the contact junctions of sensor using conducting silver paste and chronoamperometric study was performed to determine any interference from this conducting thread. The response of the sensor is shown in Figure 5c, its evident due to the presence of the conducting thread there has been a significant drop in the current. However, the sensor was capable in discerning various breathing patterns as marked. Upon first exhalation a rise of 3 pA was there and as breathing continued the peaks are clearly present. Initially, the volunteer was asked breath normally for a period of 7 min, time period for 1 inhalation and exhalation was ~ 2.4 ± 0.5 s. For a period of ~ 2 min (from 15 - 17 min), a volunteer was asked to breath fast and response was noted, during this period, the total time for inhalation-exhalation was ~ 1.2 s and a breath rate of ~ 82 breaths/min. Subsequently, the volunteer was asked to take deep inhalation and exhale slowly, the time -period for one breath cycle was ~ 4.4 ± 0.3 s. Finally, at the end of 23 min, the volunteer upon his last exhalation was asked to hold his breath for a period of ~ 1 min after which he resumed his breathing. A slight increase in the overall current as time progressed is observed. This could be attributed to the increased adsorption of moisture over the course of time.

Sensor’s response under different controlled humidity was measured as shown in Figure 7a. Supersaturated salt solutions were used to control the humidity inside the bottle, a calibration curve depicting the voltage and sensitivity at different humidity is shown in Figure 7b.

A volunteer wearing the smart mask in shown in Figure 8a, the conducting threads extend from the mask which are further connected to the measurement circuit. Photograph of the microcontroller along with the measurement and filtering circuit used is shown in Figure 8b. A schematic of the circuit structure necessary for sensing the humidity is shown in Figure 8c.

The sensor unit after calibration was then stitched on a surgical mask, the conducting thread was then used to connect the mask to the prototype circuit as shown in Figure 8. The volunteer was then asked exhale through their mouth and nose. The response obtained from the microcontroller in the presence of different breathing patterns are plotted in Figure 9. Unlike in the case of chronoamperometric study, wherein upon exhalation there was a rise in current. However, from Figure 9a, upon exhalation, observed a stark decrease in the output voltage. From Figure 9a, the time taken for one exhalation and inhalation was ~ 10.3 s, as the volunteer was asked to breathe slowly. This decreased to 5.1 s for normal breathing and decreased further to ~ 2.2 s when asked to breathe fast. From Figure 9(a, b and c), 6, 13 and 28 breaths per min were observed for slow, normal and fast breathing. The oral breathing patterns are show in Figure 9(d, e and f), where in similar to the nasal breathing the time period for one inhalation and exhalation was -12.3 s and this reduced to ~ 4.7 s for normal breathing as shown in Figure 4e. For fast breathing the time period was ~ 2.4 s as shown in Figure 9f. For both the nasal and oral breathing cycles, similar number of breaths per min was noted.

Examples

Use of conducting cloth for transforming mask to a smart one:

The invention proposed to use the conducting cloth stitched onto a conventional mask and transform it into a smart mask for monitoring the humidity of the exhaled breath. Use of conducting cloth for detecting ethanol and acetone:

The predominant use of this invention is to measure humidity present in the exhaled breath, and tested its capability in detecting various concentrations of acetone and ethanol. This can be further extended to other volatile organic compounds.

Use of mask for monitoring breathing rate:

The present invention monitors the changes in the humidity of exhaled breath, this correlates to the breath rate.

Use of Android application for analyzing sensor data:

The present invention classifies the different patterns present in the sensor’s response and detects the nature of breathing.

Use of mask in varied situations:

The smart mask can be worn comfortably during usual day-to-day activities, while sleeping and while exercising for monitoring the breath rate.

Use of other substrates:

The present smart mask has non-woven polypropylene cloth as the substrate, by this method, conducting polymer can be further grown on cloth material such as cotton, silk, nylon and polyester. Other porous media that is not water soluble is also an ideal substrate.

Use of other conducting polymer:

The present invention can be further extended to other conducting polymer like polyyrole for sensing humidity of the exhaled breath.

It may be appreciated by those skilled in the art that the foregoing drawings, examples and experimental evidences are merely illustrative and are not to be taken as limitations upon the scope of the invention. Various changes and modifications to the disclosed embodiments will be apparent to those skilled in the art. Such changes and modifications may be made without departing from the scope of the invention.

Claims

We Claim:

1. A method of fabricating a conducting cloth based breath humidity sensor for sensing humidity of the exhaled breath, the method comprises a. in-situ polymerization of conducting polymers on clothes; b. soaking the polymerized clothes from step (a) in water for 12 h for complete wetting and uniform polymerization; c. immersing the soaked clothes from step (b) in 1 M aqueous HC1 solution consisting of 6 mL of aniline for 2 min; d. immersing the soaked clothes from step (c) in 1 M aqueous HC1 solution with 12 g of ammonium persulfate for 2 min after removing unbound aniline molecules; e. cleaning the cloth from step (d) in clean water for 2 min and drying at 75 °C for 12 h; f. screen printing the interdigitated sensor array on the conducting cloth obtained from step (e) which in turn is connected to the measurement circuit via conducting thread using silver paste; characterized in that, the conducting cloth with interdigitated sensor array is stitched on to the face mask and transformed into smart mask, when the user wearing the smart mask that differentiates the breathing patterns from nose as well as the mouth and thereby measuring the humidity in the exhaled breath.

2. The method as claimed in claim 1, wherein the conducting polymers is selected from polyaniline and polypyrrole.

3. The method as claimed in claim 1, wherein the cloth is selected from cotton, silk, nylon, leather, polyester and polypropylene.

4. The method as claimed in claim 1, wherein voltage drops across the sensor is measured using a microcontroller.

5. The method as claimed in claim 1, wherein the measured data is filtered and smoothed to remove noise and enhance pattern recognition.

6. The method as claimed in claim 1 , wherein the sensor is connected to an Android mobile application to collect the data via Bluetooth, to store and to compute the respiration rate.

7. The method as claimed in claim 6, wherein the data collected from the sensor readings are used to create database of labelled data. The method as claimed in claim 1, wherein the collected data is used to train classification algorithms including ID Convolutional Neural Network, k-Nearest Neighbours, Symbolic Aggregate approximation in Vector Space Model, Bag-of-SFA Symbols in Vector Space and Logistic Regression, and the trained model is used to detect patterns in breathing. The method as claimed in claim 1, wherein the sensor detects the concentration of ethanol, acetone and volatile organic compounds that are present in the exhaled breath. The method as claimed in claim 1, wherein the sensor detects the signs of sleep apnoea and irregular breathing.