Biomarker detection method

Abstract

A computer-implemented method 300 of analysing a biomarker test having a plurality of test lines that comprises receiving 302 an image of the biomarker test; measuring 304 intensities of pixels of the

Description

Field of the invention The present invention provides a method for analysing a biomarker test. In particular, the invention provides a computer-implemented method for analysing a biomarker test to measure a level of glycated haemoglobin in a sample. Background A biomarker is typically a measurable indicator of the severity of presence of a disease or physiological state of an organism. One way of measuring biomarkers is to deposit a sample on a biomarker test, such as a lateral flow device, which is chemically reactive so that a measurement of the biomarker is visually discernible. Traditionally, such biomarker tests were visually read and interpreted by a user or medical practitioner. More recently, cameras such as those from smartphones have been used to capture an image of the biomarker test to enable recording and analysis of the measurement. European Patent Application EP4113429A1 discloses a computer-implemented method and system of image correction for a biomarker test. The biomarker test has a calibration array and a biomarker site. The calibration array has a plurality of coloured patches and the biomarker site is colour-responsive to indicate a measurement of biomarkers present. Summary of the invention Embodiments of the disclosure aim to provide a computer-implemented method of analysing a biomarker test having a plurality of test lines, the method comprising any or all of the following steps: receiving an image of the biomarker test; measuring intensities of pixels of the image; locating the plurality of test lines in the image; summing the intensities of the pixels over at least a portion of at least one test line of the plurality of test lines; and calculating a concentration of a biomarker based on summing the intensities. The method may provide the advantage of automatically quantifying biomarker test results without the need for manual interpretation by a medical practitioner. The biomarker test may be in the form of a lateral flow device. The biomarker test may comprise a strip along which a sample (such as blood, saliva, etc.) is configured to travel, for example by capillary action, along a flow direction. The strip may have a light colour, such as white, before any sample has been introduced. Upon introduction of a sample containing a target substance (such as a target molecule), detection regions (typically lines) of the strip are configured to turn a darker colour (such as red), such that the contrast between these darker detection regions and the lighter remainder of the strip indicates the presence of the target substance. The biomarker test may be for measuring the amount of glycosylated haemoglobin (HbA1 c) in a sample. The biomarker test may be for measuring the amount of haemoglobin (Hb) in a sample. The biomarker test may be configured to measure the proportion of haemoglobin in the glycosylated form. Locating the plurality of test lines in the image may comprise analysing the measured intensities of the pixels. The image may be captured by a camera, such as a camera of a smartphone. Measuring intensities of pixels of the image may comprise measuring the lightness or luminosity of the pixels. The pixels may be classified according to their L*a*b* values. Measuring the intensities may comprise measuring the L* values of the pixels in the image. Locating the plurality of test lines in the image may comprise identifying peaks in the measured intensities of the pixels. The peaks may represent regions of the biomarker test that contrast with the lighter, background colour. In this respect, the peaks may represent dark regions of the biomarker test. When the intensities of pixels are measured in some unit of lightness, such as the L* value, a darker pixel may have a lower intensity. As such, the intensity peak may also be described as a trough. Identifying peaks may comprise identifying characteristic features of the biomarker test in the image. For example, the biomarker test may comprise an identifier, such as a QR code and / or one or more calibration barcodes. Locating such characteristic features may be used to identify and locate peaks. For example, the biomarker test may comprise a calibration barcode adjacent to a test strip that contains the test lines. As such, by identifying the location and the orientation of the calibration barcode, the location of the test strip and the test lines may be identified, for example based on a stored spatial relationship between the identifier and the test lines. Identifying peaks may comprise analysing the intensity of the pixels. For example, the intensity of the pixels along a portion of the biomarker test, such as a one-dimensional portion which spans across at least one test line, may be analysed to identify regions of intensity changes. This may comprise calculating a rate of change of pixel intensity along the portion to identify minima and maxima of pixel intensity. Locating the plurality of test lines may comprise using a neural network. In this respect, the neural network may be configured to receive information from the image, such as pixel intensities, and output information corresponding to the location of the test lines in the image. The plurality of test lines may comprise a first test line, a second test line and a third test line. Locating the third test line may be based on locating the first test line and / or the second test line. The second test line may be adjacent to and between the first test line and the third test line. Locating the third test line in the image may be based on a predetermined separation between the first test line and the second test line. The biomarker test may be configured such that the presence of the target substance is detected particularly strongly by the first test line and the second test line, for example such that a relatively strong test line is produced, resulting in a relatively strong peak. Locating the plurality of test lines may comprise identifying the first test line and the second test line by identifying the relatively strong peaks. Once the first test line and the second test line are identified, a separation therebetween may be calculated. The separation, for example measured in pixels, may be calculated from peak-to-peak of the first test line and the second test line. This may be performed by identifying the highest pixel intensity associated with the first test line and the highest pixel intensity associated with the second line, and measuring the number of pixels therebetween. Locating the plurality of test lines may be based on a stored predefined separation between a pair of the plurality of test lines. The biomarker test may have a predefined separation between the first test line and the second test line, and between the first test line and the third test line, and so on. The biomarker test may have a predefined relative separation between, for example, the first test line and the second test line and between the second test line and the third test line. The plurality of test lines may be separated by the same distance, such that they are evenly spaced. Summing the intensities may be performed over a range, for example a line, of pixels which spans the test line. The range of pixels need not include all pixels associated with a given test line. Rather, the range may be provided perpendicularly to the test line, for example as a slice through the test line from a position in front of the test line to a position behind the test line. The range over which the intensities are summed may be part of the portion of the biomarker test. Summing the intensities may be performed for multiple test lines such that a summed intensity can be calculated for each of the plurality of test lines. Summing the intensities may comprise calculating the area associated with a peak from a test line on a plot of intensity against distance. Summing the intensities may comprise integrating the intensities over the at least one test line. Summing the intensities of the pixels over the at least one test line may comprise identifying a start and an end of the at least one test line. Such identification may be performed by analysing a rate of change of pixel intensity across the at least one test line. Summing the intensities of the pixels over the at least one test line may comprise identifying a start and an end of the at least one test line using a neural network. Summing the intensities may be performed over a central portion of the at least one test line. Summing the intensities may be performed over a first portion of the at least one test line and a second portion of the at least one test line. Summing the intensities may be performed over a plurality of portions of the at least one test line. The plurality may comprise two portions, or at least three portions, or at least four portions, or at least five portions. The portions may be parallel to one another. Each portion may be perpendicular to the at least one test line. Each portion may be parallel to the flow direction. The portions may be distributed across the biomarker test, for example across its width. Calculating the concentration may be based on summing the intensities of the first portion and the second portion. The summed intensities for a given test line may be averaged across the portions. The summed intensities from the first test line taken from the first portion and the second portion may be averaged, for example by calculating a mean summed intensity. Calculating the concentration may comprise applying a multiplier to the summed intensities. The biomarker test may comprise a first test strip and a second test strip. The method further may comprise analysing the intensities of the pixels to locate a first set of test lines of the first test strip and a second set of test lines of the second test strip. Locating the first set of test lines and / or locating the second set of test lines may be performed using any of the methods described above. Summing the intensities may be performed over at least a portion of at least one test line of the first set of test lines and at least a portion of at least one test line of the second set of test lines. Summing the intensities over the first test line and / or over the second test line may be performed using any of the methods described above. Calculating the concentration may be performed based on a comparison of a first summed intensity of the first set of test lines and a second summed intensity of the second set of test lines. Calculating the concentration of the biomarker may be based on a ratio of the first summed intensity to the second summed intensity. Calculating the concentration may comprise applying multipliers to the first summed intensity and the second summed intensity. The first set of test lines may comprise five test lines and the second set of test lines may comprise five test lines. Summing the intensities may be performed over all of the plurality of test lines. Embodiments of the disclosure also aim to provide a computer program product comprising program instructions configured to program a programmable processor to perform the method described hereinabove. Embodiments of the disclosure also aim to provide programmable logic configured to perform the method described hereinabove. Brief description of the drawings Embodiments of the disclosure will now be described, by way of example only, with reference to the accompanying drawings, in which: Fig. 1 is a schematic illustration of a biomarker test according to embodiments of the disclosure; Fig. 2 is a flowchart illustrating a method of analysing a biomarker test according to embodiments of the disclosure; Fig. 3A is a schematic illustration of an image of a biomarker test; Fig. 3B is a plot of intensity against distance; Fig. 4 is the plot of Fig. 3B illustrating areas of intensity; Fig. 5 is a schematic illustration of part of the image of Fig. 3A; Fig. 6 is a schematic illustration of a computer system for implementing the method of Fig. 2. Detailed description of the drawings Embodiments of the disclosure may relate to a computer-implemented method of analysing a biomarker test. The biomarker test being analysed by the method is a lateral flow test. As such, the biomarker test can be in the form of a handheld cassette which has a sample port, onto which a sample is deposited, and a test strip which chemically reacts with the deposited sample to indicate the presence of a target substance. The test strip of a lateral flow test typically has a test line to indicate the presence of the target substance, and a control line to indicate whether the test has been completed successfully. The biomarker test of the present disclosure has multiple test lines. The test lines may be configured to provide different reaction intensities with the sample. The biomarker test has two such test strips, each having multiple test lines. This can allow a relative quantification of the presence of a biomarker. The method includes receiving an image of the biomarker test. As such, after a test has been carried out and the results have been displayed on the test strips, a user can capture an image of the test, for example using a smartphone. Image processing is then performed to allow quantification of the measured biomarker. Initially, the test lines in the image are located. The test lines have a predictable shape and position with respect to the cassette, so this can be achieved with the assistance of computer vision and / or by measuring and analysing the intensities of pixels in the image. Once the test lines have been identified, the strength of each line is measured to give an indication of the strength of the chemical reaction. This can be done by considering at least one slice through each line and summing the intensities of the pixels of the slice. In this way, a concentration of the biomarker can be calculated. Figure 1 is a biomarker test 100 according to the disclosure. The biomarker test 100 comprises a housing in the form of a cassette 101 which contains various components of the biomarker test 100. The cassette 101 can be in the form of a generally rectangular housing, for example made of plastic. The biomarker test 100 comprises an identifier 104, a sample port 102, a first test strip 110 and a first calibration barcode 119. The identifier 104 is configured to provide a unique identification of the biomarker test 100, for example using a QR code that encodes information the biomarker test. The identifier 104 can be scanned by a smartphone, for example, in order to link the particular test to a particular user. The sample port 102 is configured to permit the introduction of a sample (such as blood, urine, saliva, etc.) into the cassette 101. In the arrangement shown, the sample port 102 comprises an opening in the cassette 102. The first test strip 110 is provided within the cassette 101. In the arrangement shown, a portion of the test strip 110 can be viewed through a first window 117 in the cassette 101. The first test strip 110 extends within the cassette from a first end, in proximity to the sample port 102, to a second end. As such, sample deposited at the sample port 102 can be transferred along the first test strip 110 from the first end to the second end along a flow direction. The first test strip 110 is configured to measure the level of a biomarker. In this respect, the first test strip 110 comprises reactive molecules, such as antigens, which are configured to react with a target analyte in the sample to produce a visually detectable signal. The first test strip 110 comprises detection regions at which the reactive molecules are disposed such that the detectable signals are produced at the detection regions. In the arrangement shown, the detection regions are in the form of a plurality of test lines. The biomarker test 100 may comprise five test lines, such as a first test line 111, a second test line 112, a third test line 113, a fourth test line 114 and a fifth test line 115. Each line may be arranged such that the sample flows across each line as the sample moves along the flow direction. As such, each line may be arranged perpendicular to the flow direction. The plurality of test lines are arranged sequentially along the flow direction. In this way, the test lines are arranged such that, as the sample flows along the first test strip 110, the sample first encounters the first test line 111, then the second test line 112, and so on. The first strip 110 also comprises a control region, which may be in the form of a control line 116. The control line 116 is configured to provide an indication as to whether the biomarker test 100 is functioning correctly, for example by indicating whether the sample has been transferred along a sufficient length of the first test strip 110. In this respect, the control line 116 may be arranged similarly to the plurality of test lines but positioned after the plurality of control lines in the flow direction. In the illustrated example, the control line 116 is positioned towards the second end of the first test strip 110, after the fifth test line 115. The first calibration barcode 119 is configured to provide a coloured reference for the digital analysis of the biomarker test. The first calibration barcode 119 may be arranged on the cassette 101 adjacent to the first test strip 110, in particular to the first window 117, and may extend parallel to the flow direction. The first calibration barcode 119 may comprise a plurality of regions, for example rectangles, which each comprise a different colour. In the arrangement shown, different colours are represented by different patterns of the first calibration barcode 119. To perform a test, a user deposits a sample on the sample port 102 of the cassette 101. By capillary action or otherwise, the sample is transferred within the cassette 101 from the sample port 102 along the first test strip 110 in order to interact with the plurality of test lines and the control line. A positive result of the target analyte is indicated by a colour change of the first test strip 110 in the region of each test line. If no colour change is observed at the control line 116 after a predetermined time, this indicates an invalid test, for example because the volume of sample deposited was not sufficient to extend along the length of the first test strip 110. In view of the above, the biomarker test 100 may provide a lateral flow device. In the illustrated arrangement, the biomarker test 100 also comprises a second test strip 120 and a second calibration barcode 129. These may be configured similarly to the corresponding first test strip 110 and first calibration barcode 119 described above. At least a portion of the second test strip 120 may be visible through a second window 127 of the cassette 101. The second test strip 120 comprises a plurality of test lines, for example a first test line 121, a second test line 122, a third test line 123, a fourth test line 124 and a fifth test line 125, arranged in sequence along the flow direction. The second test strip 120 may be configured to detect a different biomarker compared to the first test strip 110. The second calibration barcode 129 may comprise a different arrangement of colours compared to the first calibration barcode 119, for example to visually distinguish the two barcodes from one another. The sample port 102 is configured to receive a sample such that it can be transferred, preferably in equal volumes, to the first test trip 110 and the second test strip 110. The operation of the second test strip 120 is similar to that of the first test strip 110. As such, the biomarker test 100 can be formed as a lateral flow device having two test strips which are served by the same sample port 102. In some examples, the first test strip 110 is configured to detect glycated haemoglobin (HbA1c) and the second test strip 120 is configured to detect haemoglobin (Hb). In this respect, the test lines of each test strip are configured to detect the corresponding target by including appropriate reactive molecules, such as antigens, along the test lines. By measuring two different biomarkers, the biomarker test 100 can be configured to measure the proportion of haemoglobin in a sample that is in the glycated form. Figure 2 shows a computer-implemented method 300 of analysing a biomarker test. The method 300 may be performed on an image of a biomarker test having a plurality of test lines, such as the biomarker test 100 described in relation to Figure 1. The method 300 comprises receiving 302 an image of the biomarker test; measuring 304 intensities of pixels of the image; locating 306 the plurality of test lines in the image; summing 308 the intensities of the pixels over at least one test line; and calculating 310 a concentration of a biomarker based on summing the intensities. The step of receiving 302 an image of the biomarker test may comprise receiving an image from a camera. The camera may be a camera of a smartphone, for example. In this respect, Figure 3A shows an example of such an image 103 of a biomarker test, such as a biomarker test 100 described in relation to Figure 1. The image 103 shows the first test strip 110 and the second test strip 120, together with the calibration barcodes associated with each test strip. The step of measuring 304 intensities of pixels of the image may comprise measuring at least one characteristic of the pixels related to pixel intensity, such as the lightness. Lightness of a colour may be measured in several colour models, such as additive (describes mixing of light) or subtractive (describes mixing of pigment or removal of light). Examples of the common colour spaces are: L*a*b*, RGB, CIE 1931 xyz, CMYK, HSV, HSLuv, CIELAB and CIELUV. The pixels may be classified according to their L*a*b* values. As such, measuring 304 intensities of the pixels can comprise measuring the L* values of the pixels in the image 103. The method 300 may comprise calibrating the image 103 based on one or more calibration barcodes, such as the first calibration barcode 119 and the second calibration barcode 129. This may comprise calibrating the colour and / or lightness of the image 103. This step may be configured to account for the variable way in which different cameras capture images. Calibrating the image 103 may be performed using any of the methods and apparatus described in European Patent Application EP4113429A1. The step of locating 306 the plurality of test lines in the image may comprise analysing the measured intensities of the pixels. This may comprise identifying characteristic features of the biomarker test in the image 103 to locate the test strip containing the corresponding test lines. For example, the calibration barcodes 119, 129 represent easily identifiable features of the biomarker test 100 in the image 103. Since the positions of the first window 117 and the second window 127 are predefined with respect to at least one of the calibration barcodes 119, 129, then the positions of the test lines in the corresponding test strip can be identified based on the position of at least one of the calibration barcodes. It will be appreciated that the identifier 104 (see Figure 1), or other characteristic features of the biomarker test 100, could instead or additionally be used to locate the plurality of test lines in the image 103. The step of locating 306 the plurality of test lines may comprise identifying peaks in the measured intensities of the pixels. With reference to Figure 3A, the peaks may be identified as regions of the image 103 that contrast with a background colour. This is because each test strip is configured to change colour in the presence of a target substance in the region of the test lines. In particular, where a test strip has a light background colour (such as white) before a sample is measured, then the test lines may turn red, or another colour, when the target substance is detected. Where intensities of pixels are measured in a unit of lightness, such as the L* value, it will be appreciated that a darker pixel may have a lower intensity. As such, an intensity peak may be described as a trough of intensity. Identifying the peaks may comprise analysing the measured intensities of the pixels. This may be performed along a portion of the biomarker test, such as a one-dimensional portion which spans across at least one test line. In the example shown in Figure 3A, a portion 118 of the first test strip 110 spans across all five test lines 111-115 and the control line 116. Likewise, a portion 128 of the second test strip 120 spans across all five test lines 121-125 and the control line 126. As such, the method 300 may include identifying regions of changing intensity to identify intensity peaks. This may include calculating a rate of change of pixel intensity along the portions 118, 128 to identify minima and maxima of pixel intensity. Locating 306 the plurality of test lines in the image may comprise using a neural network to process the image 103. The neural network may be configured to receive information from the image 103, such as pixel intensities, and output information corresponding to the location of the test lines in the image. Figure 3B shows a graph 200 of pixel intensity against distance. The values of pixel intensity may be related to the lightness of a pixel, such that a darker pixel (i.e., representing a part of an image of a test line that indicates the presence of the target substance) has a lower pixel intensity. The distance along the x-axis may represent the distance in pixels along the first test strip 110 or the second test strip 120 in the image 103. For example, the graph 200 may be generated by plotting the measured intensity of each pixel along a portion of one of the test lines, such as the portion 118 of the first test line 110 or the portion 128 of the second test line 120. In the example shown in Figure 3B, the pixel distance along the x-axis of the graph 200 is the pixel distance from the second end of a test strip at which the control line is positioned, such that moving along the x-axis (towards the right-hand side) represents moving along the portion 118 or the portion 128 opposite to the flow direction (i.e., downwards in the image 103 of Figure 3A). Figure 3B illustrates an example result after identifying peaks in the pixel intensity. In particular, the method 300 may identify the peaks that are labelled with an ‘x’, for example by analysing the rate of change of pixel intensity. It will be appreciated that not every peak corresponds to a test line of the image 103. In order to aid the identification of the test lines, the biomarker test 100 may be configured such that the presence of the target substance is detected particularly strongly by the first test line and / or the second test line. This may be achieved by including a higher concentration of antigen in the region of the first test line and / or the second test line. As such, a relatively strong test line is produced upon reaction with a sample containing the target analyte. This is illustrated in Figure 3B by the strong first test line peak 211 and the strong second test line peak 212, which may correspond to the positions of the first test line 111 and the second test line 112 of the first test trip 110. As such, the step of locating 306 the test lines may comprise identifying the relatively strong peaks. In the example described above, once the first test line 111 and the second test line 112 have been identified, then a pixel distance therebetween can be measured. This may be measured using the peak-to-peak distance in pixels between the first test line peak 211 and the second test line peak 212. In some examples, this can comprise identifying the location of the strongest pixel intensity associated with each of the first test line 111 and the second test line 112, and measuring the number of pixels therebetween. In the illustrated example, the separation between the first test line peak 211 and the second test line peak 212 is represented by the distance D1. Locating 306 the plurality of test lines may be based on a stored predefined separation between a pair of test lines. In particular, it will be appreciated that in a given biomarker test 100, the distance between each pair of test lines is predefined based on where the reactive molecules are placed on the test strip. Similarly, the distance between each test line and the control line is predefined. As such, the relative separations between the lines are also predefined. In a given test strip, for example the first test strip 110, the first test line 111 may be 5 mm away from the second test line 112, while the second test line 112 may be 6 mm away from the third test line 113. As such, the separation between the second test line 112 and the third 113 is 20% greater than the separation between the first test line 111 and the second test line 112. Therefore, by measuring the separation D1 between the first test line peak 211 and the second test line peak 212, the location of the third test line peak 213 can be inferred based the position of the second test line peak 212 and on a factor of D1 (in this case, a factor of 1.2 or a 20% increase). A successful biomarker test 100 will result in a colour change at the control line 116, for example. As such, the method 300 may comprise identifying a control line peak 216. The step of locating 306 the test lines may therefore comprise identifying the control line peak 216. Similarly to the method described above, a separation between the control line peak and one of the test line peaks could be used to locate the other test line peaks. In the illustrated example, the distance between the control line peak 216 and the first test line peak 211 is represented by the distance D2. As such, locating 306 the plurality of test lines may be based on a stored predefined separation between a control line and a test line. The step of summing 308 the intensities of the pixels over at least one test line is performed after the test lines in the image have been located. Summing the intensities of a given test line can be performed over a range, for example a line, of pixels which spans the test line. In other words, summing the intensities can be performed over a portion of a given test line, such as a line portion that is perpendicular to the test line. With reference to the image 103 in Figure 3A, the range of pixels may include pixels from a slice of the first test line 111 from a position just in front of the first test line 111 to a position just behind the first test line 111. The range of pixels over which the summation is performed may be along the portion 118, i.e. along a central region of the first test line 111. This may correspond to the first test line peak 211 as shown in Figure 3B. It will be appreciated that the intensities of all of the plurality of test lines may be summed in this way to produce a summed intensity for each test line. Figure 4 shows the graph 200 of Figure 3B including an illustration of a method for summing 308 the intensities of the pixels. Based on the graph 200 of pixel intensity against pixel distance, it will be appreciated that the area associated with a test line peak represents the summed intensity associated with that peak. For example, a first test line peak area 211a is associated with the first test line peak 211 and a second test line peak area 212a is associated with the second test line peak 212. As such, summing 308 the intensities of the pixels may comprise calculating the area associated with a peak from a test line. This may comprise integrating the pixel intensity over a test line. Such methods may require identification of a start and an end of a given test line, which may be determined by analysing a rate of change of pixel intensity across the test line, and / or by using a neural network, for example. As mentioned in relation to Figure 2, the method 300 further comprises calculating 310 a concentration of a biomarker based on summing the intensities. In other words, the summed intensity for a test line may be used to calculate the amount of biomarker in a sample. Given that a higher concentration of the target analyte in a sample results in a stronger chemical reaction and an associated stronger colour change, this step can therefore provide a quantification of the level of biomarker in a sample by quantifying the strength of a test line. Calculating the concentration comprises applying a multiplier to the summed intensity of a given line. The multiplier may be based on predefined characteristics of the biomarker test 100, such as the concentration of reactive molecules deposited at each test line. The multiplier may be based on at least one measured characteristic of the image 103, such as its resolution. The relationship between pixel intensity and concentration is not necessarily linear. In some examples, the method may comprise performing a regression analysis to determine the relationship. Linear trends can be observed for biomarkers at certain concentration range but it is common that for biomolecules such as antibodies and enzymes, the relationship between the pixel intensity and concentration follows polynomial or sigmoidal probability distribution curves. This is due to the natural complex system learning curves, specific to antibodies and enzymes, which can exhibit a progression from small beginnings that accelerate and approach a climax over time. The method 300 may include performing the steps for a first test trip 110 and a second trip 120. For example, after receiving 302 an image 103 of the biomarker test which contains an image of both test strips 110, 120 and measuring 304 the intensities of pixels of the image, the method 300 includes performing the steps of locating 306 the test lines associated with both test strips 110, 120, and summing 308 the intensities of the pixels over at least one test line of each test strip 110, 120. These steps may be performed using any of the methods described above. As such, a first set of test lines associated with the first test strip and a second set of test lines associated with the second strip can be identified, and a summed intensity can be calculated for each test line in a set. The step of calculating 310 the concentration may be performed based on a first summed intensity from the first set of test lines and a second summed intensity from the second set of test lines. In some examples, calculating the concentration of a biomarker is based on a ratio of the first summed intensity to the second summed intensity. This may be particularly relevant to measuring the level of glycated haemoglobin in a sample. As discussed in relation to Figure 1, the first test strip 110 can be configured to detect glycated haemoglobin (HbA1c) and the second test strip 120 can be configured to detect haemoglobin (Hb). By comparing the signal strengths of the test lines for each test strip, a concentration of the HbA1c biomarker can be compared to a concentration of the Hb biomarker in order to obtain a measurement of a concentration of HbA1c in a sample. Such a measurement may be used to diagnose whether a patient is diabetic, pre-diabetic or healthy. In operation, with reference to Figures 1 to 4, a user introduces a sample to the biomarker test 100 via the sample port 102. The sample is separated equally between the first test strip 110 and the second test strip 120. The sample travels along each test strip in the flow direction, during which the sample interacts with antigens at each test line until eventually reaching the control line. To analyse the completed biomarker test, the user takes an image of the biomarker test 100 with their smart phone. This may involve scanning the identifier 104 to obtain stored information about the particular biomarker test 100. This may also involve recognising the calibration barcodes 119, 129 and applying a calibration for the colour and / or lightness of the image, for example to produce a calibrated image. The intensities of the pixels of the image (or of the calibrated image) are measured and the test lines in each strip are located. Pixel intensities of each test line are summed across a given test line, for example along the portion 118 of the first test strip 110 and the portion 128 of the second test strip 120. The total summed intensity from the test lines of the first test strip 110 can then be divided by the total summed intensity from the test lines of the second test strip 120 in order to provide a relative concentration between the target analyte of the first test strip 110 (e.g., HbA1c) and the target analyte of the second test strip 120 (e.g., Hb). The method 300 has been described as including a step of summing the intensities of the pixels along a portion of the test strip, for example the central portion 118 of the first test strip 110. Summing the intensities of pixels in the central region may be preferable because it can avoid including the intensities from the edges of the strip at which the flow of the sample is slower. As such, the pixel intensities in the central region, such as along the central portion 118, may provide a more accurate measurement. However, the disclosure is not limited to using a single portion, such as a single portion along the central region. Rather, summing the intensities may be performed over a plurality of portions of the at least one test line. The plurality may comprise at least two portions. In this respect, Figure 5 illustrates an extract 103a of the image 103, showing the first test strip 110 and the second test strip 120 only. In the illustrated example, the portions over which the intensities of the pixels are summed are shown schematically. In particular, the plurality of portions for the first test strip 110 comprises a first portion 118a, second portion 118b, third portion 118c, fourth portion 118d and fifth portion 118e. Likewise, the plurality of portions for the second test strip comprises a first portion 128a, second portion 128b, third portion 128c, fourth portion 128d and fifth portion 128e. In each test strip, the plurality of portions may be parallel to one another. Each portion may be one-dimensional, in that each portion is one pixel thick and extends along the length of the test strip in the flow direction. The portions may be distributed across each test strip of the biomarker test 100, for example across its width as shown in Figure 5. The step of calculating 310 the concentration of a biomarker may be based on summing the intensities of at least two portions in each test strip, for example the first portion 118a and the second portion 118b of the first test strip 110 and the first portion 128a and the second portion 128b of the second test strip 120. The summed intensities for a given test line may be averaged across the portions. For example, an average of the summed pixel intensity for the first test line 111 of the first test strip 110 may be calculated based on the summed intensity of the first portion 118a and the second portion 118b in the region of the first test line 111. As mentioned above, a neural network may be used to process the image 103. In particular, a neural network may be used to identify a start and an end of a test line to aid the step of summing the intensities. Various machine learning models may be used, such as those provided by TensorFlow or PyTorch. A machine learning model may be trained based on datasets of known start and end points in a pixel intensity distribution. The accuracy of identification of the test lines, particularly the start and end of the peaks, may be evaluated using the Intersection over Union metric. This metric may be used to train the machine learning model. For line detection, the most common types of neural network model are the segmentation model and object detection neural model. Each of the models follows a traditional, supervised training protocol / process, in which the true labels may be provided by a human observer. Also disclosed is a computer program product. The computer program product comprise program instructions configured to program a programmable processor to perform the method 300 described above. Also disclosed is programmable logic configured to perform the method 300 described above. In some examples the functionality of the processor described herein may be provided by a general purpose processor, which may be configured to perform a method according to any one of those described herein. In some examples the processor may comprise digital logic, such as field programmable gate arrays, FPGA, application specific integrated circuits, ASIC, a digital signal processor, DSP, or by any other appropriate hardware. In some examples, one or more memory elements can store data and / or program instructions used to implement the operations described herein. Embodiments of the disclosure provide tangible, non-transitory storage media comprising program instructions operable to program a processor to perform any one or more of the methods described and / or claimed herein and / or to provide data processing apparatus as described and / or claimed herein. The processor may comprise an analogue control circuit which provides at least a part of this control functionality. An embodiment provides an analogue control circuit configured to perform any one or more of the methods described herein. Figure 6 is a block diagram of a computer system 1200 suitable for implementing one or more embodiments of the present disclosure, including for example any of the methods described above with reference to Figures 1 to 5. The computer system 1200 includes a bus 1212 or other communication mechanism for communicating information data, signals, and information between various components of the computer system 1200. The components include an input / output (I / O) component 1204 that processes a user (i.e., sender, recipient, service provider) action, such as selecting keys from a keypad / keyboard, selecting one or more buttons or links, etc., and sends a corresponding signal to the bus 1212. It may also include a camera for obtaining image data. The I / O component 1204 may also include an output component, such as a display 1202 and a cursor control 1208 (such as a keyboard, keypad, mouse, etc.). The display 1202 may be configured to present a login page for logging into a user account, and is configured to display a user interface. An optional audio input / output component 1206 may also be included to allow a user to use voice for inputting information by converting audio signals. The audio I / O component 1206 may allow the user to hear audio. A transceiver or network interface 1220 transmits and receives signals between the computer system 1200 and other devices, such as another user device, a merchant server, or a service provider server via network 1222. In one embodiment, the transmission is wireless, although other transmission mediums and methods may also be suitable. A processor 1214, which can be a micro-controller, digital signal processor (DSP), or other processing component, processes these various signals, such as for display on the computer system 1200 or transmission to other devices via a communication link 1224. The processor 1214 may also control transmission of information, such as cookies or IP addresses, to other devices. The components of the computer system 1200 also include a system memory component 1210 (e.g., RAM), a static storage component 1216 (e.g., ROM), and / or a disk drive 1218 (e.g., a solid-state drive, a hard drive). The computer system 1200 performs specific operations by the processor 1214 and other components by executing one or more sequences of instructions contained in the system memory component 1210. For example, the processor 1214 can run the method 300 described above. Logic may be encoded in a computer readable medium, which may refer to any medium that participates in providing instructions to a processor for execution. Such a medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. In various implementations, non-volatile media includes optical or magnetic disks, volatile media includes dynamic memory, such as a system memory component, and transmission media includes coaxial cables, copper wire, and fibre optics. In one embodiment, the logic is encoded in non-transitory computer readable medium. In one example, transmission media may take the form of acoustic or light waves, such as those generated during radio wave, optical, and infrared data communications. Some common forms of computer readable media include, for example, floppy disk, flexible disk, hard disk, magnetic tape, any other magnetic medium, CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with patterns of holes, RAM, PROM, EPROM, FLASH-EPROM, any other memory chip or cartridge, or any other medium from which a computer is adapted to read. In various embodiments of the present disclosure, execution of instruction sequences to practice the present disclosure may be performed by a computer system. In various other embodiments of the present disclosure, a plurality of computer systems coupled by a communication link to a network (e.g., such as a LAN, WLAN, PTSN, and / or various other wired or wireless networks, including telecommunications, mobile, and cellular phone networks) may perform instruction sequences to practice the present disclosure in coordination with one another. It will also be understood that aspects of the present disclosure may be implemented using hardware, software, or combinations of hardware and software. Also, where applicable, the various hardware components and / or software components set forth herein may be combined into composite components comprising software, hardware, and / or both without departing from the spirit of the present disclosure. Where applicable, the various hardware components and / or software components set forth herein may be separated into subcomponents comprising software, hardware, or both without departing from the scope of the present disclosure. In addition, where applicable, it is contemplated that software components may be implemented as hardware components and vice-versa. Software in accordance with the present disclosure, such as program code and / or data, may be stored on one or more computer readable mediums. It is also contemplated that software identified herein may be implemented using one or more general purpose or specific purpose computers and / or computer systems, networked and / or otherwise. Where applicable, the ordering of various steps described herein may be changed, combined into composite steps, and / or separated into sub-steps to provide features described herein. The various features and steps described herein may be implemented as systems comprising one or more memories storing various information described herein and one or more processors coupled to the one or more memories and a network, wherein the one or more processors are operable to perform steps as described herein, as non-transitory machine-readable medium comprising a plurality of machine-readable instructions which, when executed by one or more processors, are adapted to cause the one or more processors to perform a method comprising steps described herein, and methods performed by one or more devices, such as a hardware processor, user device, server, and other devices described herein. As utilized herein, terms “component,” “system,” “interface,” “unit” and the like are intended to refer to a computer-related entity, hardware, software (e.g., in execution), and / or firmware. For example, a component can be a processor, a process running on a processor, an object, an executable, a program, a storage device, and / or a computer. By way of illustration, an application running on a server and the server can be a component. One or more components can reside within a process, and a component can be localized on one computer and / or distributed between two or more computers. Further, these components can execute from various computer readable media having various data structures stored thereon. The components can communicate via local and / or remote processes such as in accordance with a signal having one or more data packets (e.g., data from one component interacting with another component in a local system, distributed system, and / or across a network, e.g., the Internet, a local area network, a wide area network, etc. with other systems via the signal). It will be appreciated from the above description that many features of the different examples are interchangeable and combinable. The disclosure extends to further examples comprising features from different examples combined together in ways not specifically mentioned. Indeed, there are many features presented in the above examples and it will be apparent to the skilled person that these may be advantageously combined with one another.

Claims

1. A computer-implemented method of analysing a biomarker test having a plurality of test lines, the method comprising:receiving an image of the biomarker test;measuring intensities of pixels of the image;locating the plurality of test lines in the image;summing the intensities of the pixels over at least a portion of at least one test line of the plurality of test lines; andcalculating a concentration of a biomarker based on summing the intensities.

2. The method of claim 1, wherein the biomarker test is for measuring the amount of glycated haemoglobin (HbA1c) in a sample.

3. The method of claim 1 or claim 2, wherein locating the plurality of test lines in the image comprises analysing the measured intensities of the pixels.

4. The method of any preceding claim, wherein locating the plurality of test lines in the image comprises identifying peaks in the measured intensities of the pixels.

5. The method of any preceding claim, wherein the plurality of test lines comprises a first test line, a second test line and a third test line, and wherein locating the third test line is based on locating the first test line and / or the second test line.

6. The method of claim 5, wherein locating the third test line in the image is based on a predetermined separation between the first test line and the second test line.

7. The method of any preceding claim, wherein summing the intensities of the pixels over the at least one test line comprises identifying a start and an end of the at least one test line by analysing a rate of change of pixel intensity across the at least one test line.

8. The method of any preceding claim, wherein summing the intensities of the pixelsover the at least one test line comprises identifying a start and an end of the at least one test line using a neural network.

9. The method of any preceding claim, wherein summing the intensities is performed over a central portion of the at least one test line.

10. The method of any preceding claim, wherein summing the intensities is performed over a first portion of the at least one test line and a second portion of the at least one test line.

11. The method of claim 10, wherein calculating the concentration is based on summing the intensities of the first portion and the second portion.

12. The method of claim 11, wherein calculating the concentration comprises applying a multiplier to the summed intensities of the first portion and the second portion.

13. The method of any preceding claim, wherein the biomarker test comprises a first test strip and a second test strip, and wherein the method further comprises analysing the intensities of the pixels to locate a first set of test lines of the first test strip and a second set of test lines of the second test strip.

14. The method of claim 13, wherein summing the intensities is performed over at least a portion of at least one test line of the first set of test lines and at least a portion of at least one test line of the second set of test lines.

15. The method of claim 14, wherein calculating the concentration is performed based on a comparison of a first summed intensity of the first set of test lines and a second summed intensity of the second set of test lines.

16. The method of claim 15, wherein calculating the concentration of the biomarker is based on a ratio of the first summed intensity to the second summed intensity.

17. The method of claim 15 or claim 16, wherein calculating the concentrationcomprises applying multipliers to the first summed intensity and the second summed intensity.

18. The method of any of claims 13 to 17, wherein the first set of test lines comprises 5 five test lines and the second set of test lines comprises five test lines.

19. The method of any preceding claim, wherein summing the intensities is performed over all of the plurality of test lines.10 20. A computer program product comprising program instructions configured toprogram a programmable processor to perform the method of any of claims 1 to 19.

21. Programmable logic configured to perform the method of any of claims 1 to 19.

Images