Technical Field

The present disclosure relates to a computer-implemented method and system for obtaining concentration of a volatile organic compound, particularly, in an exhaled breath.

Background

Volatile organic compounds (VOCs) found in an exhaled breath can originate from within the body or from external sources, such as a diet, an environmental exposure, and prescription drugs. Therefore, the VOCs present in exhaled breaths will vary between individuals, and even from a single individual when the breath samples are taken at different times. An exemplary method for sampling VOCs is by proton transfer reaction mass spectrometry (PTR-MS). PTR-MS allows breath samples to be collected quickly and non-invasively, which can then be analyzed in real-time to give a mass spectrum showing the concentration and mass-to-charge ratio (m/z) of the molecules present in the exhaled breath.

An exhaled breath profile of every individuals can vary drastically. Existing data processing programs require manual determination and alignment of the period of the exhaled breath for determining VOCs concentrations in the exhaled breath. This leads to problems related to complex operation, low analysis efficiency and inaccurate alignment. Further, inaccurate determination of VOCs concentrations may also be caused by artificial alcohol wiping, expiration interruption, device jitter, etc. during the breath collection process.

It is therefore desirable to provide a computer-implemented method and system for obtaining concentration of a volatile organic compound in an exhaled breath which address the aforementioned problems and/or provides a useful alternative. Further, other desirable features and characteristics will become apparent from the subsequent detailed description and the appended claims, taken in conjunction with the accompanying drawings and this background of the disclosure.

Summary Aspects of the present application relate to a method and system for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath.

In accordance with a first aspect, there is provided a computer implemented method for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath, the method comprising: (i) receiving a mass spectrum of an exhaled breath sample; (ii) extracting an expiratory curve based on the mass spectrum, the expiratory curve being associated with a known VOC found in the exhaled breath sample; (iii) determining an expiratory interval using the expiratory curve, the expiratory interval being associated with a time interval of the exhaled breath of the exhaled breath sample; and (iv) obtaining an expiratory VOC concentration for the VOC measured in the mass spectrum, the expiratory VOC concentration being an average VOC concentration measured during the expiratory interval of the exhaled breath sample.

By extracting an expiratory curve associated with a known VOC and determining an expiratory interval using the expiratory curve, an expiratory VOC concentration in relation to the expiratory interval can be obtained which accurately reflects the concentration of the VOC contained within the exhaled breath. The aforementioned method can be readily implemented in real-time for improving an accuracy and reliability of spectral data processing, and has significant advantages such as simple operation, high analysis efficiency, accurate concentration calculation, etc. The expiratory interval or ‘breath period’ of the exhaled breath can also be systematically and accurately determined from the exhaled breath samples, thereby improving an accuracy and reliability of mass spectrum data processing for determining a concentration of the VOC (or concentrations of a plurality of VOCs) that are present in the exhaled breath.

The method may comprise smoothing the expiratory curve using a Savitzky-Golay convolution smoothing algorithm, and wherein determining the expiratory interval may comprise determining the expiratory interval using the smoothed expiratory curve. Smoothing of the curve aids in removing “kinks” in the expiratory curve to prepare it for subsequent analysis.

Determining the expiratory interval may comprise: identifying maxima of the expiratory curve; determining a plateau interval for each respective maximum of the expiratory curve, the plateau interval being an interval between a first plateau point and a last plateau point, the first plateau point and the last plateau point being defined as immediate adjacent data points to a first data point and a last data point having a concentration value above a plateau baseline concentration value; and identifying the expiratory interval based on the determined plateau interval for each respective maximum, wherein the expiratory interval is the plateau interval having a largest interval value.

The method may comprise removing abnormal maxima from the expiratory curve based on a concentration value of each of the maxima, the abnormal maxima having a concentration value which is outside a predetermined concentration range associated with the known VOC.

The method may comprise: identifying a peak interval of each of the maxima, the peak interval being a time interval between a first peak point and a last peak point, wherein the first peak point and the last peak point are each defined in relation to a respective predetermined percentage of a magnitude of concentration of a respective maximum. The peak interval and the plateau interval of each of the maxima can be subsequently used to filter or remove peaks which are not relevant in determining an expiratory interval of an exhaled breath.

The method may comprise: determining a background interval using the expiratory curve, the background interval being a time interval of the expiratory curve recorded prior to the expiratory interval; and obtaining a background VOC concentration for the VOC measured in the mass spectrum, the background VOC concentration being an average VOC concentration measured during the background interval of the exhaled breath sample. The background VOC concentration can be used as a channel for a convolutional neural network in subsequent disease detection and prediction step where any VOC with a concentration below the background VOC concentration will be removed.

The method may comprise removing one or more of the maxima if the plateau interval of the one or more maxima is less than 50% of the peak interval. This aids in reducing the number of maxima for subsequent analysis, and therefore improve an efficiency in computing and determining an expiratory interval of an exhaled breath.

The method may comprise removing one or more of the maxima if the plateau interval of the one or more maxima is equal to or less than 3 seconds. This aids in reducing the number of maxima for subsequent analysis, and therefore improve an efficiency in computing and determining an expiratory interval of an exhaled breath.

The method may comprise merging adjacent maxima if an interval between the adjacent maxima is less than or equal to 3 seconds. This aids in reducing the number of maxima for subsequent analysis, and therefore improve an efficiency in computing and determining an expiratory interval of an exhaled breath.

In accordance with a second aspect, there is provided a computer readable medium storing processor executable instructions which when executed on a processor cause the processor to carry out any of the preceding method.

In accordance with a third aspect, there is provided a computer system for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath, the system comprising a processor and a data storing computer program instructions operable to cause the processor to: receive a mass spectrum of an exhaled breath sample; extract an expiratory curve based on the mass spectrum, the expiratory curve being associated with a known VOC found in the exhaled breath sample; determine an expiratory interval using the expiratory curve, the expiratory interval being associated with a time interval of the exhaled breath of the exhaled breath sample; and obtain an expiratory VOC concentration for the VOC measured in the mass spectrum, the expiratory VOC concentration being an average VOC concentration measured during the expiratory interval of the exhaled breath sample.

The data storage may store computer program instructions operable to cause the processor to: smooth the expiratory curve using a Savitzky-Golay convolution smoothing algorithm, and wherein the computer program instructions operable to cause the processor to determine the expiratory interval may comprise computer program instructions operable to determine the expiratory interval using the smoothed expiratory curve.

The data storage may store computer program instructions operable to cause the processor to: identify maxima of the expiratory curve; determine a plateau interval for each respective maximum of the expiratory curve, the plateau interval being an interval between a first plateau point and a last plateau point, the first plateau point and the last plateau point being defined as immediate adjacent data points to a first data point and a last data point having a concentration value above a plateau baseline concentration value; and identify the expiratory interval based on the determined plateau interval for each respective maximum, wherein the expiratory interval is the plateau interval having a largest interval value. The data storage may store computer program instructions operable to cause the processor to: remove abnormal maxima from the expiratory curve based on a concentration value of each of the maxima, the abnormal maxima having a concentration value which is outside a predetermined concentration range associated with the known VOC.

The data storage may store computer program instructions operable to cause the processor to: identify a peak interval of each of the maxima, the peak interval being a time interval between a first peak point and a last peak point, wherein the first peak point and the last peak point are each defined in relation to a respective predetermined percentage of a magnitude of concentration of a respective maximum.

The predetermined percentage of the magnitude of concentration of the maxima for the first peak point and the second peak point may be 50% to 80%.

The data storage may store computer program instructions operable to cause the processor to: determine a background interval using the expiratory curve, the background interval being a time interval of the expiratory curve recorded prior to the expiratory interval; and obtain a background VOC concentration for the VOC measured in the mass spectrum, the background VOC concentration being an average VOC concentration measured during the background interval of the exhaled breath sample.

The background interval may be defined as an interval before a first of the maxima of the expiratory curve and may have a concentration value of less than 10% of a peak height of the expiratory interval, where the peak height is an absolute value of a maximum concentration of the known VOC within the expiratory interval.

The data storage may store computer program instructions operable to cause the processor to: remove one or more of the maxima if the plateau interval of the one or more maxima is less than 50% of the peak interval.

The data storage may store computer program instructions operable to cause the processor to: remove one or more of the maxima if the plateau interval of the one or more maxima is equal to or less than 3 seconds.

The data storage may store computer program instructions operable to cause the processor to: merge adjacent maxima if an interval between the adjacent maxima is less than or equal to 3 seconds. The exhaled breath sample may be provided by a human subject and the known VOC may include isoprene or acetone. Isoprene or acetone are known VOCs which exist in high concentrations in an exhaled breath of a human. They are therefore good markers for use in identifying an expiratory interval of a human exhaled breath.

The mass spectrum may be obtained using proton transfer reaction mass spectrometry (PTR-MS).

It should be appreciated that features relating to one aspect may be applicable to the other aspects. Embodiments therefore provide a method and system for obtaining concentration of a VOC in an exhaled breath. By extracting an expiratory curve associated with the known VOC and determining an expiratory interval of the exhaled breath using the expiratory curve, an expiratory VOC concentration in relation to the expiratory interval can be obtained which accurately reflects the concentration of the VOC contained within the exhaled breath. The aforementioned method can be readily implemented in real-time for improving an accuracy and reliability of spectral data processing, and has significant advantages such as simple operation, high analysis efficiency, accurate concentration calculation, etc. The expiratory interval or ‘breath period’ of the exhaled breath can also be systematically and accurately determined from the exhaled breath samples, thereby improving an accuracy and reliability of mass spectrum data processing for determining a concentration of the VOC (or concentrations of a plurality of VOCs) that are present in the exhaled breath.

Brief description of the drawings

Embodiments will now be described, by way of example only, with reference to the following drawings, in which:

Figure 1 shows a block diagram of a computer system for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath in accordance with an embodiment;

Figure 2 is a flowchart showing steps of a method for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath in accordance with an embodiment;

Figure 3 is a flowchart showing steps of a method for determining an expiratory interval of the expiratory curve of Figure 2 in accordance with an embodiment; Figure 4 shows a diagram of an expiratory curve of isoprene as extracted from a sample mass spectrum in accordance with an embodiment;

Figure 5 shows a diagram of the expiratory curve of isoprene of Figure 4 after smoothing in accordance with an embodiment;

Figure 6 shows a diagram of the smoothed expiratory curve of Figure 5 to illustrate merging of adjacent maxima in accordance with an embodiment;

Figure 7 shows a diagram of the smoothed expiratory curve of Figure 5 to illustrate filtering of sharp peaks using determined plateau interval of each of the identified maxima in accordance with an embodiment;

Figure 8 shows a diagram of the smoothed expiratory curve of Figure 5 to illustrate identification of an expiratory interval of the exhaled breath where the expiratory interval is defined as the plateau interval having a largest interval value in accordance with an embodiment; and

Figure 9 shows a diagram of the smoothed expiratory curve of Figure 5 to illustrate identification of a background interval of the exhaled breath sample in accordance with an embodiment.

Detailed description

Exemplary embodiments relate to a method and system for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath.

Figure 1 shows a block diagram of a computer system 100 for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath in accordance with an embodiment.

As shown in Figure 1 , the computer system 100 has memory that stores computer program modules which implement computer-implemented method for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath. The computer system 100 comprises a processor 102, a working memory 104, an input module 106, an output module 108, a user interface 110, program storage 112 and data storage 114. The processor 102 may be implemented as one or more central processing unit (CPU) chips. The program storage 112 is a non-volatile storage device such as a hard disk drive which stores computer program modules such as a VOC concentration determination module 116 and a mass spectrometry module 118. These computer program modules are loaded into the working memory 104 for execution by the processor 102. The input module 106 is an interface which allows data, for example data in relation to an exhaled breath sample, such as a mass spectrum obtained etc., to be received by the computer system 100. The output module 108 is an output device which allows data and results generated by the computer system 100 to be output. The output module 108 may be coupled to a display device or a printer. The user interface 110 allows a user of the computer system 100 to input selections and commands and may be implemented as a graphical user interface.

In the present embodiment, the program storage 112 stores the VOC concentration determination module 116 and the mass spectrometry module 118. The VOC concentration determination module 116 includes a smoothing module 120, an expiratory interval module 122, a background interval module 124 and a VOC concentration calculation module 126. These computer program modules cause the processor 102 to execute various analytical processes which are described in more detail below. For example, the smoothing module 120 can be executed by the processor 102 to smooth an expiratory curve extracted from a mass spectrum. The expiratory interval module 122, in the present embodiment, can be executed by the processor 102 to determine an expiratory interval of an expiratory curve or a smoothed expiratory curve. The background interval module 124 can be executed by the processor 102 to determine a background interval of an expiratory curve or a smoothed expiratory curve, while the VOC concentration calculation module 126 can be executed by the processor 102 to obtain at least an expiratory VOC concentration for a VOC measured in a mass spectrum using a determined expiratory interval. In an embodiment, the VOC concentration calculation module 126 can be executed by the processor 102 to obtain a background VOC concentration for a desired VOC, or background VOC concentrations for each of the VOCs measured, during the expiratory interval. The program storage 112 may be referred to in some contexts as computer readable storage media and/or non-transitory computer readable media. In an embodiment, the computer program modules such as the VOC concentration determination module 116 and the mass spectrometry module 118 can be stored in a cloud storage. As depicted in Figure 1 , the computer program modules 116, 118, 120, 122, 124, 126 are distinct modules which perform respective functions implemented by the computer system 100. It will be appreciated that the computer program modules 116, 118, 120, 122, 124, 126 as presented in Figure 1 and their boundaries are exemplary only, and that alternative embodiments may merge modules or impose an alternative decomposition of functionality of modules. For example, the modules discussed herein may be decomposed into sub-modules to be executed as multiple computer processes, and, optionally, on multiple computers. Moreover, alternative embodiments may combine multiple instances of a particular module or submodule. It will also be appreciated that, while a software implementation of the computer program modules is described herein, these may alternatively be implemented as one or more hardware modules (such as field-programmable gate array(s) or applicationspecific integrated circuit(s)) comprising circuitry which implements equivalent functionality to that implemented in software.

The data storage 114 stores various model data and model parameters. As shown in Figure 1 , the data storage 114 has storage for VOC concentration determination data 128 and mass spectrometry data 130 for use with their corresponding modules 116, 118. In the present embodiment, the VOC concentration determination data 128 comprises smoothing data 132, expiratory interval data 134, background interval data 136 and VOC concentration calculation data 138. The smoothing data 132 stores data in relation to a smoothing process performed on an expiratory curve of a known VOC extracted from the mass spectrum, such as one or more smoothing algorithms which can be used for the smoothing process. The expiratory interval data 134 stores data in relation to determination of the expiratory interval, for example the identified maxima of the expiratory curve, determined expiratory interval associated with the expiratory curve of the known VOC, identified peak interval and determined plateau interval for each maxima of the expiratory curve etc. The background interval data 136 stores e.g. data in relation to the background interval obtained using the expiratory curve, and the VOC concentration calculation data 138 stores data in relation to determined expiratory VOC concentration for a desired VOC or determined expiratory VOC concentrations for each of the VOCs measured during the expiratory interval. The VOC concentration calculation data 138 may also store data in relation to determined background VOC concentration for a desired VOC or determined background VOC concentrations for each of the VOCs measured during the expiratory interval. The mass spectrometry data 130 stores mass spectrum data in relation to measured mass spectrum data obtained from exhaled breath samples by mass spectrometry (e.g. PTR-MS) etc. Similar to the computer program modules as discussed above, it will be appreciated that the boundaries between the data 128, 130, 132, 134, 136, 138 of the data storage 114 as presented in Figure 1 are exemplary only, and that alternative embodiments may impose an alternative decomposition.

Although the technical architecture is described with reference to a computer system 100, it should be appreciated that the technical architecture may be formed by two or more computers in communication with each other that collaborate to perform a task. For example, but not by way of limitation, an application may be partitioned in such a way as to permit concurrent and/or parallel processing of the instructions of the application. Alternatively, the data processed by a computer program module may be partitioned in such a way as to permit concurrent and/or parallel processing of different portions of a data set by the two or more computers. In an embodiment, virtualization software may be employed by the technical architecture to provide the functionality of a number of servers that is not directly bound to the number of computers in the technical architecture. In an embodiment, the functionality disclosed above may be provided by executing a computer program module or computer program modules in a cloud computing environment. Cloud computing may comprise providing computing services via a system connection using dynamically scalable computing resources. A cloud computing environment may be established by an enterprise and/or may be hired on an as-needed basis from a third-party provider.

Figure 2 is a flowchart showing steps of a method 200 for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath in accordance with an embodiment. The method 200 is carried out using the computer system 100.

In a step 202, the mass spectrometry module 118 is executed by the processor 102 to receive a mass spectrum of an exhaled breath sample. Data in relation to the received mass spectrum can be stored as mass spectrometry data 130. The received mass spectrum is used by the VOC concentration determination module 116 in subsequent steps for determining a concentration of VOC in the exhaled breath of the exhaled breath sample.

In a step 204, the VOC concentration determination module 116 is executed by the processor 102 to extract an expiratory curve based on the mass spectrum received in the step 202. The expiratory curve is associated with a known VOC found in the exhaled breath sample. The known VOC is therefore dependent on a subject who produces the exhaled breath sample. In an embodiment where the exhaled breath sample is from a human subject, either isoprene or acetone can be identified as the known VOC for extracting the relevant expiratory curve. In an embodiment, the expiratory curve is extracted directly from the mass spectrum.

In a step 206, the VOC concentration determination module 116 is executed by the processor 102 to smooth the expiratory curve using a smoothing algorithm. In the present embodiment, the Savitzky-Golay convolution smoothing algorithm can be used.

In a step 208, the VOC concentration determination module 116 is executed by the processor 102 to determine an expiratory interval using the smoothed expiratory curve obtained in the step 206. The expiratory interval is associated with a time interval of the exhaled breath of the exhaled breath sample. Detail for determining the expiratory interval is further described in relation to Figure 3 below.

In a step 210, the VOC concentration determination module 116 is executed by the processor 102 to determine a background interval using the smoothed expiratory curve. The background interval may be associated with a time interval of the expiratory curve recorded prior to the expiratory interval. In the present embodiment, the background interval is defined as a time interval before a first of the maxima of the expiratory curve, where a concentration value of the background interval is less than 10% of a peak height of the expiratory interval. The peak height of the expiratory interval is defined as an absolute value of a maximum concentration of the known VOC within the expiratory interval.

In a step 212, the VOC concentration determination module 116 is executed by the processor 102 to obtain an expiratory VOC concentration and a background VOC concentration for the VOC measured in the mass spectrum. The expiratory VOC concentration relates to an average VOC concentration measured during the expiratory interval in the exhaled breath sample, while the background VOC concentration relates to an average VOC concentration measured during the background interval.

Figure 3 is a flowchart showing steps of a method 300 for determining an expiratory interval using the expiratory curve in the step 208 of Figure 2 in accordance with an embodiment. The method 300 is carried out using the computer system 100.

In a step 302, the VOC concentration determination module 116 is executed by the processor 102 to identify maxima of the expiratory curve. In the present embodiment, the maxima are searched using the local_maxima_1d function in Python with peak search algorithm for searching local maxima. In a step 304, the VOC concentration determination module 116 is executed by the processor 102 to merge adjacent maxima to form a merged maximum if an interval between the adjacent maxima is less than or equal to a predetermined interval. In the present embodiment, the predetermined interval is 3 seconds but it should be appreciated that any other suitable predetermined interval can be used, for example, 1 second, 2 second, 4 seconds etc.

In a step 306, the VOC concentration determination module 116 is executed by the processor 102 to identify a peak interval of each of the maxima (and/or merged maxima if applicable), the peak interval being a time interval between a first peak point, Li, and a last peak point, Ri. In the present embodiment, the first peak point Li and the last peak point Ri are each defined in relation to a respective predetermined percentage of a magnitude of concentration of a respective maximum of the expiratory curve. The predetermined percentage of the magnitude of concentration of the maxima for the first peak point and the second peak point is 60% in the present embodiment, but it should be appreciated that this can vary, for example within a range of 50% to 80%, depending on applications.

In a step 308, the VOC concentration determination module 116 is executed by the processor 102 to determine a plateau interval for each respective maximum of the expiratory curve. In the present embodiment, the plateau interval is defined as an interval between a first plateau point and a last plateau point, where the first plateau point and the last plateau point are defined as immediate adjacent data points to a first data point and a last data point having a concentration value above a plateau baseline concentration value. The plateau baseline concentration value can be a predetermined value in relation to a peak height of each of the respective maximum. In the present embodiment, the plateau baseline is calculated using Xi - AHi x 0.3, where Xi is a concentration value of the respective maximum, and AHi is defined as a difference between Xi and a larger value of the first peak point, Li and the last peak point, Ri (i.e. larger value of (Li, Ri)). The first plateau point and the last plateau point are being defined as the immediate adjacent data points to a first data point and a last data point having a concentration value above a plateau baseline concentration value. In other words, the first plateau point is the immediate adjacent data point to the first data point which is above the plateau baseline concentration value and the last plateau point is the immediate adjacent data point to the last data point which is above the plateau baseline concentration for that respective maximum. In a step 310, the VOC concentration determination module 116 is executed by the processor 102 to remove one or more of the maxima if the plateau interval of the one or more maxima is less than a predetermined peak interval value. In the present embodiment, the VOC concentration determination module 116 is executed by the processor 102 to remove one or more of the maxima if the plateau interval of the one or more maxima is less than 50% of the peak interval.

In a step 312, the VOC concentration determination module 116 is executed by the processor 102 to remove one or more of the maxima if the plateau interval of the one or more maxima is equal to or less than a predetermined plateau interval. In the present embodiment, the predetermined plateau interval is 3 seconds.

In a step 314, the VOC concentration determination module 116 is executed by the processor 102 to remove abnormal maxima from the expiratory curve based on a concentration value of each of the maxima, the abnormal maxima having a concentration value which is outside a predetermined concentration range associated with the known VOC. In the present embodiment, the predetermined concentration range for isoprene (which was used as the known VOC for a human exhaled breath sample) is between 15 and 500 parts per billion (ppb). Therefore, any maximum of the expiratory curve which has a concentration value outside this range will be removed.

In a step 316, the VOC concentration determination module 116 is executed by the processor 102 to identify the expiratory interval based on the determined plateau interval for each respective maximum, wherein the expiratory interval is the plateau interval which has a largest interval value.

The method 300 therefore provides a method for determining the expiratory interval of the exhaled breath accurately, which can be used in the method 200 for determining the expiratory VOC concentration in the exhaled breath.

An exemplary embodiment for obtaining concentration of a volatile organic compound (VOC) in an exhaled breath is described below, in conjunction with Figures 4 to Figure 9. In the present embodiment, proton transfer reaction mass spectrometry (PTR-MS) was used to obtain the exhaled breath sample and a brief description of it is provided below. Although PT-RMS was used, it should be appreciated that other mass spectrometry may be used in other embodiments. It should be appreciated that the terms “breath collection” or “breath sampling” can used interchangeably herein to refer to any breath sampling process. The concentration value mentioned below are all in ppb (parts per billion) as labelled in Figures 4 to 9.

Proton-Transfer-Reaction Mass Spectrometry (PTR-MS)

An online breath sampler is connected to the PTR-MS to transfer the breath sample to it in real time for analysis. This method effectively avoided sample absorption, storage and transportation, thereby minimizing sample loss and contamination. The method also enables fast on-spot detection which allows point-of-care diagnosis. The PTR-MS measures the concentration of a few hundred VOCs, some of which are disease biomarkers which can be singled out for data analysis. The peaks were identified using the library of PTR-MS, and further validated by external standards.

The instrument comprises an ionization section and a detection section. During the ionization process, the instrument forms protonated water ions (H ₃ O ⁺ ) by a hollow cathode discharge in the ion source. These H ₃ O ⁺ ions are then introduced into the drift tube by an electric drift field, where they chemically ionize the volatile organic compounds (VOC) in breath samples via proton-transfer reaction (PTR). Only the VOCs with higher proton affinity (PA) value than that of H ₂ O molecules will be ionized by H ₃ O ⁺ and proceed to the detection section. These ionized VOCs are extracted by the electric field towards the time-of-flight mass spectrometer (TOF MS) to be differentiated and detected with respect to their mass-to-charge ratio (m/z). In the present embodiment, the drift tube voltage used was 600 V and the drift tube pressure was 2.3 mbar. The E/N ratio was 139 Td. The sampling line and buffer tube were kept at 70 °C.

In the present embodiment, the exhaled breath sample was obtained from a human subject and the known VOC used was isoprene.

Figure 4 shows a diagram of an expiratory curve 400 of isoprene (e.g. m069 molecules) in accordance with an embodiment. The expiratory curve 400 of isoprene was extracted from the mass spectrum of the exhaled human breath sample obtained using PTR-MS. The expiratory curve 400 of isoprene was used as it is a known VOC which is present in a human’s breath. It therefore serves as a good marker for determining an expiratory interval (or a breath period) of an exhaled breath. The expiratory curve 400 of isoprene as obtained (i.e. in the form of raw data) is recorded as M(x) in the present embodiment.

Figure 5 shows a diagram of the expiratory curve 400 of Figure 4 after smoothing in accordance with an embodiment. In the present embodiment, the expiratory curve 400 of the isoprene molecule was smoothed with Savitzky-Golay convolution smoothing algorithm. The smoothed expiratory curve data 500 is recorded as S(x).

The detailed smoothing method used in the present embodiment is as follows: quadratic polynomial fitting was performed for each data point Xi, the 2 data points before and the 2 data points after Xi, a total of 5 data points. The data on both sides of the first point and the last point was then set to 0. After substituting the values of the five data points into quadratic polynomials, the coefficients form a three-dimensional linear system of equations, which was solved by the Gauss Jordan elimination method. The fitted value of point Xi can be obtained after the coefficient was solved. The smoothed data obtained by fitting all points of M(x) on the curve was recorded as S(x).

In the present embodiment, the expiratory interval was then determined using the smoothed expiratory curve. First, local maxima of the expiratory were searched. This can be performed as follows: search from front to back along the smooth curve S(x) 500. If a data point Xi is higher than the data points Xi-1 and Xi+1 on both its sides at the same time, it is considered to be a local maximum.

Once the local maxima of the smoothed expiratory curve S(x) 500 were located, a number of processes were performed for determining an expiratory interval of the exhaled breath of the exhaled breath sample.

Figure 6 shows a diagram 600 of the smoothed expiratory curve S(x) 500 of Figure 5 to illustrate merging of adjacent maxima in accordance with an embodiment. First, local maxima which were close to one another were merged and this is called “peak merging”. In the present embodiment, if the distance 601 between two local maxima 602, 604 is less than or equal to 3 seconds, the two maxima 602, 604 are merged. The larger local maximum is taken as the merged peak.

To perform further filtering of the maxima for identifying the expiratory interval of the exhaled breath, a peak interval and a plateau interval were determined.

To determine the peak interval of a peak formed by each respective local maximum or a merged maximum, the following was performed. An example is described herewith in relation to a local maximum Xi. First, extend from both sides of the local maximum Xi. If a point on the left is less than 60% of the maximum value (i.e. the value of Xi), the previous point will be recorded as Li (i.e. a first peak point) and the search will be terminated. Similarly, if a point on the right is less than 60% of the maximum value, the point that exceeds the previous value is recorded as Ri (i.e. a last peak point). The interval between Li and Ri is defined as the peak interval of the local maximum Xi, and the peak width of the peak interval is recorded as A Wi 702. This is shown in relation to Figure 7. The difference between the maximum value Xi and the larger value of (Li, Ri) is recorded as A Hi 704. In an embodiment where the local maximum concerned is a merged peak, then the maximum value will be the maximum concentration of the merged peak.

To identify plateau interval, the following was performed. First, determine a plateau baseline. The plateau baseline can be calculated using: Xi- A Hi X 0.3 and will extend from both sides of the local maximum Xi. When a data point on the left or right is lower than the plateau baseline, the extension will stop. The two ends (i.e. the first plateau point and the last plateau point, respectively) after the extension stops are regarded as a peak plateau and a width of the peak plateau (i.e. the plateau interval) is recorded as A Pi 706.

Figure 7 shows a diagram 700 of the smoothed expiratory curve of Figure 5 to illustrate filtering of sharp peaks using the determined plateau interval 706 and/or the determined peak interval 702 of each of the identified maxima in accordance with an embodiment.

A stable peak plateau is typically expected for an exhaled breath in the expiratory curve of the known VOC. A plateau interval can therefore be a good parameter for use in determining an expiratory interval of the exhaled breath. In the present embodiment, any plateau interval which is less than 3 seconds, i.e., A Pi is less than 3 seconds, were removed. Further, in the present embodiment, the plateau interval was also required to be greater than or equal to 50% of the peak width, i.e., A Pi/ A Wi>=0.5. As shown in Figure 7, the peak 708 on the left which has a ratio A Pi/ A Wi < 0.5 was therefore filtered out or removed from consideration.

In addition, peaks or local maxima with abnormal concentration values (e.g. exceptionally high or exceptionally low) were filtered. In the present embodiment, only peaks or maxima with a concentration value between 15 and 500 were selected as the normal concentration range of isoprene molecules (i.e. the selected known VOC in this case) exhaled from an individual is between 15 and 500. Figure 8 shows a diagram 800 of the smoothed expiratory curve 500 of Figure 5 to illustrate identifying of an expiratory interval of the exhaled breath. Using the above processes or steps as described, most of the local maxima were filtered out. If there are other peaks or maxima left, the one with the widest plateau interval (i.e. the plateau interval 802) was selected as the expiratory interval. Therefore, in the present embodiment, the expiratory interval can be determined as the plateau interval having a largest interval value. For example, as shown in Figure 8, the right-hand local maximum 804 was filtered out in this case.

Figure 9 shows a diagram 900 of the smoothed expiratory curve 500 of Figure 5 to illustrate identifying of a background interval of the exhaled breath sample in accordance with an embodiment. As described above, the expiratory interval 802 of the expiratory curve for the isoprene molecules was obtained. However, the background interval can also be determined from the expiratory curve for subsequent processing. The background interval provides a means to determine background VOC concentration in the environment. To determine the background interval, the VOC concentration determination module 116 is executed by the processor 102 to search in front of the peak where the first maximum value occurs, set the concentration not higher than 10% of the peak height of the expiratory interval 802 (i.e. the widest peak plateau) as the threshold value of the background, and find the background interval. The background interval 902 is shown as circle data points prior to the 10 s mark on the x-axis of the plot of Figure 9.

Once the expiratory interval 802 and the background interval 902 of the expiratory curve 500 were obtained, in the present embodiment, the average values of all VOC molecules measured during the expiratory interval were recorded as expiratory VOC concentrations for the exhaled breath, and the average values of all the VOC molecules measured during the background interval were recorded as background VOC concentrations. If the concentration of a VOC molecule is lower than the ambient background concentration of the VOC molecule, its concentration was set to 0 in the present embodiment.

Each of the expiratory VOC concentration and the background VOC concentration are then output, preferably on the user interface 110. The expiratory curve 400 and/or the smoothed expiratory curve 500 of the isoprene molecules, and the labels in relation to the expiratory interval 802 and the background interval 902 can also be shown. This aids the operators/users of the mass spectrometry, e.g. PTR-MS in this case, to view the stability and reliability of the data acquisition. In the above description, the terms “subject”, “individual”, “patient” or “host” are used interchangeably herein, refer to any subject. The term "subject" includes any human or non-human animal. In one embodiment, the subject is a human. The term "non-human animal" includes all vertebrates, e.g., mammals and non-mammals, such as non-human primates, sheep, dog, cow, chickens, amphibians, reptiles, etc.

The embodiment as described above was performed using online breath sampling where a breath sample was collected in real-time from the subject using a PTR-MS system. Although the breath sample was collected in real-time, the breath sample was collected when it was exhaled or external from the subject. The breath sample had therefore been extracted from the subject.

In some embodiments, a breath sample can be collected using the PTR-MS system but performance of the aforementioned described method can happen at a later time. In other embodiments, the aforementioned methods can be performed using offline breath sampling where a breath sample was collected using a breath collection device (e.g. BIO VOC or breath collection bag), and the breath sample was subsequently analyzed by the PTR-MS system. In this case, similar to the aforementioned method, an expiratory curve can be determined based on the mass spectrum as the curve that relates to a known VOC (e.g. isoprene or acetone) found in an exhaled breath sample. The exhaled breath in this case will be related to a withdrawal rate of the exhaled breath sample from the breath collection device and defined using the expiratory curve. The expiratory interval of the exhaled breath as defined in the aforementioned method can be determined in a similar manner using the expiratory curve, and therefore the expiratory VOC concentration of the exhaled breath sample can be obtained in a similar manner as afore- described.

It should be appreciated that certain steps or processes as described in the exemplary embodiment are optional. For example, the smoothing step may be considered optional. Also, it may not be necessary to determine the background interval and/or the background VOC concentrations in some embodiments. Further, it should be appreciated that one or more of the filtering steps as described above (e.g. in relation to the steps 304, 310, 312, 314) are also optional in some embodiments. For example, an expiratory interval of the expiratory curve can be determined, e.g. based on a widest peak or maxima of the expiratory curve, without one or more of these filtering steps. Also, identification of the peak interval may also be optional if the relevant or associated filtering/removal step is not used.

Other alternative embodiments include: (1) use of other form of mass spectrometry (e.g. gas chromatography mass spectrometry and selected ion flow tube mass spectrometry) for obtaining a mass spectrum of an exhaled breath sample; (2) use of other appropriate expiratory curves (e.g. acetone curve, and other molecules also depending on a nature of the subject) for determining the expiratory interval; (3) other smoothing algorithm for smoothing the expiratory curve; (4) use of other algorithms or methods for locating the local maxima of the expiratory curve (e.g. the local_maxima_1d function in MATLAB etc.); (5) the predetermined percentage of the magnitude of concentration of the maxima for the first peak point and the second peak point for identifying a peak interval being selected from between 50% to 90%; (6) other suitable definitions for defining the plateau interval and/or the peak interval for subsequent filtering processes; (7) a plateau baseline concentration value which is defined differently, e.g. being Xi-AHix0.4 or Xi-AHix0.2 or other suitable definitions; (8) removing one or more of the maxima if the plateau interval of the one or more maxima is less than 70% or 60% or 40% or 30% of the peak interval or any other suitable percentage; (9) removing one or more the maxima if the plateau interval of the one or more maxima is equal to or less than 5 seconds, or equal to less than 4 seconds, or equal to less than 2 seconds, or equal to less than 1 second; (10) merging adjacent maxima if an interval between the adjacent maxima is less than or equal to 5 seconds, less than or equal to 4 seconds, less than or equal to 2 seconds or less than or equal to 1 second; (11) the background interval having a concentration value of less than a percentage selected from between 20% to 2% of a peak height of the expiratory interval; and (12) other methods or definitions used to determine the expiratory interval, for example, the expiratory interval may be determined as the full-width at half maximum of the widest maxima or peak of the expiratory curve.

Although only certain embodiments of the present invention have been described in detail, many variations are possible in accordance with the appended claims. For example, features described in relation to one embodiment may be incorporated into one or more other embodiments and vice versa.