Systems and methods for imaging and image-based analysis of testing equipment

Abstract

Systems and methods for imaging and image-based analysis of testing equipment may include a background device for a transverse flow measurement test strip. In one aspect, the background device may include: a test strip portion sized and shaped to guide the placement of the transverse flow measurement test strip on the background device; a background portion at least partially surrounding the test strip portion; and one or more features of the background portion. The one or more features may include a line detection reference, a position reference, a modulation transfer function reference, a motion blur detection reference, and / or an RGB balance region for evaluating lighting conditions.

Description

[0001] Cross Reference to Related Applications

[0002] This application claims the benefit of U.S. Provisional Application Serial No. 63 / 105,146, filed October 23, 2020, entitled “SYSTEMS AND METHODS FOR IMAGING AND IMAGE-BASED ANALYSIS OF DIAGNOSTIC TEST DEVICES,” and U.S. Provisional Application Serial No. 63 / 126,437, filed December 16, 2020, entitled “SYSTEMS AND METHODS FOR IMAGING AND IMAGE-BASED ANALYSIS OF TEST DEVICES,” both of which are incorporated by reference herein in their entireties. TECHNICAL FIELD

[0003] The present technology relates to analyte testing, and more specifically to an application for reading cartridge analysis information and a test cartridge background device. BACKGROUND

[0004] Test analysis information can be obtained from images of test devices, such as lateral flow assays or other cartridge-based tests. Determination of analysis information based on image analysis can be susceptible to errors based on color management, alignment, image warping, and other error sources within the captured image. For example, such approaches can be susceptible to reading errors (such as false negative results) and / or algorithmic errors (e.g., poor homographies, poor normalization, etc.) due to image quality degradation (e.g., high noise levels, poor lighting, poor focus, significant motion blur, etc.). Analysis of images captured by a camera of a smartphone or other mobile device can be particularly susceptible to such errors. SUMMARY

[0005] To limit the occurrence of errors in image-based test analysis systems, imaging systems used for test analysis are typically operated in very controlled environments and are carefully calibrated and normalized. When calibration and controlled environments are not available, it can be necessary to extract image qualification and normalization metrics from the images themselves. Accordingly, the present technology provides systems and methods for including additional contextual information having known characteristics (e.g., size, intensity, color, spatial frequency, contrast, etc.) in the captured scene along with the test device itself. The systems and methods of the present technology include, among other things, a 3D box background device (e.g., a tray or card) on which a test box can be positioned to serve as a background for an image prior to information analysis of the image taken of the test box. The present technology further includes applications, such as mobile device applications, that are configured to perform image analysis on images of the box after standardization and normalization of the images using the 3D background device. The background can contain one or more fiducials to facilitate evaluation of image capture metrics based on the image. In some embodiments, some or all of the fiducials can be located in the same plane as a surface of a test strip, such as a test surface of a lateral flow assay test strip, and / or in the same plane as a surface of a box containing the test strip to allow for fine image capture for image capture condition verification, image normalization, and / or image standardization prior to analysis using evaluation software or manual interpretation. The 3D design of the 3D background device disclosed herein can further allow for de-warping of images taken at angles other than the top view of the box and 3D background device.

[0006] In one non-limiting example, a background device for assaying a test strip is provided. The background device includes a test strip portion sized and shaped to guide placement of a lateral flow assay test strip on the background device, a background portion at least partially surrounding the test strip portion, and a plurality of line detection fiducials disposed on the background portion, each of the plurality of line detection fiducials having a color different from a color of a surrounding area of the background portion and a width associated with an expected width of a line on the lateral flow assay test strip.

[0007] The plurality of line detection fiducials can include at least a first line detection fiducial having a width substantially equal to the expected width and a second line detection fiducial having a width greater or less than the expected width. The plurality of line detection fiducials can include at least a first line detection fiducial and a second line detection fiducial having a first shade of color. The plurality of line detection fiducials can include at least a third line detection fiducial having a second shade of color lighter or darker than the first shade of color. In one example, the color is gray and the expected color of the line on the lateral flow assay test strip is not gray.

[0008] The background device can also include one or more motion blur detection fiducials, each of the one or more motion blur detection fiducials including a dot of a first color surrounded by an area of a second color that contrasts with the first color. The background device can also include at least three position fiducials disposed on the background portion proximate to corners of the background portion to facilitate detecting at least one of a position, a pitch, or a roll of the image capture device relative to the background device. In one example, the background portion includes a red-green-blue (RGB) balance area having a color corresponding to equal red, green, blue values in an RGB color space.

[0009] The test strip portion can include an alignment mark configured to facilitate placement of the lateral flow assay test strip on the test strip portion of the background device. The test strip portion can include computer readable code positioned to be covered when the lateral flow assay test strip is placed on the test strip portion, the computer readable code identifying a software application configured to analyze an image of the lateral flow assay test strip for qualification for further analysis based at least in part on the line detection fiducials. The test strip portion can include computer readable code positioned to be covered when the lateral flow assay test strip is placed on the test strip portion, the computer readable code identifying a software application configured to analyze an image of the lateral flow assay test strip to determine a test result based at least in part on the line detection fiducials. The background device can be a three-dimensional background device including one or more recesses or one or more three-dimensional features protruding from the background portion. The lateral flow assay test strip can be housed in a cartridge, and wherein the test strip portion can be sized and shaped to guide placement of the cartridge on the test strip portion of the background device. Each of the plurality of line detection fiducials can have a width associated with an expected width of at least one of: a test line that changes intensity or color when an analyte of interest is present in a sample applied to the lateral flow assay test strip; and a control line that changes intensity or color when the sample applied to the lateral flow assay test strip is present.

[0010] In another non-limiting example, a computer-implemented method of determining a test result is provided. The method can include capturing, by an image capture device, an image of a lateral flow assay test strip disposed on a test strip portion of a background device, the background device including a background portion at least partially surrounding the test strip portion. The method can also include detecting, by one or more processors, a plurality of line detection fiducials disposed on the background portion of the background device based at least in part on the image, each of the plurality of line detection fiducials having a color different from a color of a surrounding area of the background portion and a width associated with an expected width of a line on the lateral flow assay test strip. The method can further include detecting, by the one or more processors, one or more control lines or test lines on the lateral flow assay test strip based at least in part on the plurality of line detection fiducials based at least in part on the image. The method can also include determining, by the one or more processors, a test result of the lateral flow assay test strip based at least in part on the one or more detected control lines or test lines.

[0011] The method can also include, prior to capturing the image, analyzing whether an image taken by the image capture device is eligible for detection of the test result. Detecting the one or more control lines or test lines can include detecting a presence of the one or more control lines or test lines based at least in part on a width of at least one of the plurality of line detection fiducials. The plurality of line detection fiducials can include at least a first line detection fiducial having a first darkness of color and a second line detection fiducial having a second darkness of color that is lighter or darker than the first darkness of color, and a third line detection fiducial having a third darkness of color that is lighter or darker than the first darkness of color or the second darkness of color, and detecting the one or more control lines or test lines can include detecting an intensity or color of the one or more control lines or test lines based at least in part on the first line detection fiducial, the second line detection fiducial, or the third line detection fiducial. In one example, the colors of the first fiducial, the second fiducial, and the third fiducial are gray and the color of the one or more control lines or test lines is not gray.

[0012] The background portion of the background device can include a red-green-blue (RGB) balance region having a color corresponding to equal red, green, blue values in an RGB color space. The method can also include, prior to determining the test result, evaluating lighting conditions of the background device based at least in part on the RGB balance region. Evaluating the lighting conditions can also include at least one of glare or shadow on the RGB balance region. The method can also include, prior to determining the test result, evaluating a modulation transfer function of the image capture device based at least in part on the plurality of line detection fiducials or based on one or more additional fiducials on the background portion of the background device. The method can also include estimating a detection level based on the modulation transfer function, wherein the test result of the lateral flow assay is determined based at least in part on the estimated detection level. The method can also include, prior to determining the test result, detecting a motion blur level of the image based at least in part on one or more motion blur detection fiducials disposed on the background portion of the background device. Each of the one or more motion blur detection fiducials can include a dot of a first color surrounded by a region of a second color that contrasts with the first color. The method can also include, prior to determining the test result, determining at least one of a position, a pitch, or a roll of the image capture device relative to the background device based at least in part on a plurality of position fiducials disposed on the background portion of the background device. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1A is a top view illustrating an example 3D box background device according to the present technology.

[0014] Figure 1B and Figure 1C is a cross-sectional view of the example 3D box background device of Figure 1A .

[0015] Figure 1D is a close-up view of a portion of the example background device of Figures 1A-1C , illustrating example fiducials thereon.

[0016] Figure 2A is a top view illustrating placement of a test cartridge against the example 3D box background device of Figures 1A-1C .

[0017] Figure 2B and Figure 2C is a cross-sectional view of the test cartridge and example 3D box background device of Figure 2A .

[0018] Figure 3 and Figure 4 illustrate display of Figure 1A and Figure 2Aa mobile device application's graphical user interface (GUI) of an image of an example 3D cartridge background device of Example 3.

[0019] Figure 5 is an example image of a test cartridge on a 3D cartridge background device taken at an angle by a mobile device camera.

[0020] Figure 6 is a de-warped version of the image of Figure 5

[0021] Figure 7 is an example image of a test cartridge on a 3D cartridge background device taken at an angle by a mobile device camera.

[0022] Figure 8 is a de-warped version of the image of Figure 7

[0023] Figure 9 is a flowchart illustrating an example method of performing and analyzing tests according to the present technology.

[0024] Figure 10 is a flowchart illustrating an example method of collecting and analyzing images to determine test results according to the present technology.

[0025] Figure 11A and Figure 11B is a flowchart illustrating an example method of performing image qualification and normalization according to the present technology.

[0026] Figures 12A-12D illustrates an example cartridge background device according to the present technology. DETAILED DESCRIPTION

[0027] ​​Embodiments of the present disclosure relate to systems and techniques for detecting analytes of interest that can be present in biological or non-biological samples, such as fluids. Analytes of interest can include any detectable substance, such as but not limited to antibodies, proteins, haptens, nucleic acids, amplicons, hormones, and dangerous or non-dangerous drugs or contaminants, such as anti-neoplastic drugs used to treat cancer. Throughout the present disclosure, example systems, devices, and methods will be described with reference to the collection, testing, and detection of analytes, such as those related to diagnostic tests associated with infectious diseases, but it will be understood that the present techniques can be used to collect, test, and detect any particle, molecule, or analyte of interest. Test strips and / or cartridges as described herein can be configured for the performance of diagnostic and / or non-diagnostic tests. In some embodiments, embodiments of the present disclosure can be implemented in conjunction with systems such as the BD Veritor system for rapid detection of SARS CoV-2, the BD Veritor system for rapid detection of Influenza A+B, the BD Veritor system for rapid detection of Respiratory Syncytial Virus (RSV), the BD Veritor system for rapid detection of Group A Streptococcus, the BD Veritor system, the BD Veritor Plus system, and / or components or operations thereof.

[0028] Systems and methods of the present technology include, among other things, a 3D cartridge background device (e.g., a tray or card) onto which a test cartridge can be placed to serve as a background for images taken of the cartridge prior to cartridge information analysis of the images. The present technology further includes applications, such as mobile device applications, that are configured to perform image analysis on images of cartridges and 3D background devices. The background can contain one or more fiducials to facilitate evaluation of image-based image capture metrics. In some embodiments, some or all fiducials can be located in the same plane as a test surface of a test strip, such as a test surface of a lateral flow assay test strip, to allow for fine image capture for image capture condition verification, image normalization, and / or image standardization prior to analysis using evaluation software or manual interpretation. The 3D design of the 3D background devices disclosed herein can further allow for de-warping of images taken at angles other than the top view of the cartridge and 3D background device.

[0029] Some pre-processing steps can be expected before analyzing the results of the assay and establishing the analysis image. In some aspects, an assessment of image metrics can be performed first, such as to determine the suitability of the image and / or to select the best image from a plurality of images taken by the testing device. Calibration and qualification steps can follow. In some embodiments, a series of images can be captured, and the testing result can be determined from the best or most suitable image of the series. One possible embodiment described here is to capture N images (e.g. N=3), evaluate one or a series of key metrics (e.g. motion blur, shadowing effects, distortion, etc.) for each image and keep only the best or optimal one (e.g. with less motion blur or one or more other optimal characteristics). This strategy can be applied to one metric or a series of metrics or all metrics considered together. Based on the preferred metric, the best image is then deemed eligible (e.g. suitable for subsequent analysis) or not eligible (e.g. requiring the user to take another step, such as attempting to capture the image again). If eligible, the image can be calibrated and / or normalized, such as to reduce or minimize variability due to external conditions (including, for example, the user of the phone, lighting) as well as phone characteristics (e.g. optics, electronics and software).

[0030] Figures 1A-2C An example 3D background device 100 according to the present technology is illustrated. Figure 1A The 3D background device 100 is illustrated separately, Figure 2A The 3D background device 100 is illustrated with a test cartridge 200 for imaging placed thereon. Figure 1B And Figure 2B are cross-sectional views taken along the line Figure 1A And Figure 2A are cross-sectional views taken along the line Figure 1C And Figure 2C are cross-sectional views taken along the line Figure 1A And Figure 2A are cross-sectional views taken along the line

[0031] The 3D background device 100 generally comprises a background portion 102 comprising one or more fiducials and a cartridge portion 104 at which a cartridge 200 can be received for analysis. In some embodiments, the cartridge portion 104 can be a recess or an insert that facilitates proper placement of the cartridge 200.

[0032] Various example fiducials are illustrated in Figures 1A-2C and described herein in connection with their function as enablers for image metric calculations and / or for image normalization and standardization. As will be referenced to Figures 10-11BMore details, the information associated with the fiducials can be digitally stored in a background device model, which can facilitate image analysis using the fiducials. For example, the background device model can store the position (e.g., as one or more x, y coordinates) of some or each fiducial relative to optically detectable features of the 3D background device 100 (e.g., relative to one or more corners or alignment fiducials of the background device 100, such as the additional corner fiducials 155). Thus, various embodiments of the 3D background device 100 can include more or fewer fiducials than those illustrated in FIG. 1, where the position of each fiducial can be reliably determined by an image analysis application based on the corresponding background device model. Figures 1A-2C

[0033] In some embodiments, fiducials having a height that matches the height of the top surface of the cartridge 200 placed within the cartridge portion 104 of the 3D background device 100, such as the 4 cylinders 105, allow image processing software to assess the tilt and roll of the cartridge 200 in the photo so that any such tilt and roll can be corrected. In addition, the intensity and / or direction of any shadows produced by the cylinders 105 can be observed to estimate the impact of the shadows on the test strip 205 within the cartridge 200.

[0034] To well accommodate the camera's exposure (e.g., exposure time) and avoid saturation at the strip level, part of the scan card design can include some pure white areas 110. To have the 3D background device 100 facilitate standardization and / or normalization of the captured images, one or more of the following metrics can be used, individually or in combination with each other.

[0035] Focusing of the camera: To make proper focusing on the test strip 205, a grain texture 115 can be placed on the opposite side of the cartridge portion 104 on the 3D background device 100 within the same plane as the band on the test strip 205 to be detected. This texture 115, together with some or all of the other fiducials in the plane, can force the focusing (e.g., the focus of an autofocus camera such as a mobile device camera) to primarily be in the plane of the test strip 205, rather than in the plane of the cartridge 200, thus improving the ability to detect even faint test lines. In some embodiments, help such as a "focus here" prompt overlaid on the mobile device via augmented reality can prompt the user to manually focus the mobile device camera on the grain texture 115.

[0036] ​Dynamic range: The design can contain areas that span the entire dynamic range that can be used in each of the red, green, and blue color channels in many digital cameras. For example, in some embodiments, more than 2.5% of the design can correspond to 0 intensity (e.g., black level) and more than 2.5% can correspond to 255 (e.g., maximum) intensity. Thus, the histogram can be linearly expanded in each channel from 0, 255 (from 2.5 and 97.5 percent points).

[0037] Camera white balance: In addition to the box 200 itself, the vertical column 105, and the small R, G, B squares 120 used for color fidelity testing, the entire design can be an equal R, G, B balance. For example, the R, G, B values of each pixel in the remaining portion of the 3D background device surface can be equal, which simplifies the evaluation and correction of the scene's white balance regardless of the spectral characteristics of the incident light.

[0038] Color fidelity testing: The red, green, and blue squares 120 can be provided to determine the accuracy of the response in the red, green, and blue color channels. The known locations of the individual colors of the red, green, and blue squares 120 can be further utilized to confirm the Bayer pattern of a particular image sensor used to capture images of the 3D background device 100 and the box 200.

[0039] Linearity assessment / correction per channel: It can be desirable for the camera to have a linear response in all channels in order to normalize and standardize images captured from different devices (e.g., optics, electronics, and / or software of a smartphone, tables, etc.). To assess and correct the linearity of each of the R, G, B channels, a neutral patch 125 with a known expected intensity distribution can be placed on the design. The intensities of the individual subsections of the neutral patch 125 can be selected to span the entire dynamic range (e.g., 0, 16, 32, 48, 64, 80, 96, 128, 160, 192, 224, 240, 255, or other suitable set of intensities between 0 and 255), allowing the calculation and application of a linearity lookup table for each channel to linearization correction.

[0040] Motion blur assessment, deconvolution: The 3D background device can include two white disks 130 with a circular black spot in their center, specifically designed to help assess motion blur. The longer the exposure time (depending on the lighting intensity), the more likely the image can appear motion blurred. Recent mobile devices often contain image stabilization mechanisms (hardware and / or software) that seek to limit the occurrence of such motion blur, but the distance of the lens to the scene is quite small (in the range of one to a few inches), and for certain mobile devices that can lack such mechanisms, it can be important to assess and ultimately correct for such factors. If the shape and intensity of the small black spot is spread unevenly across the white disk 130, it can be determined that some motion occurred during acquisition that caused motion blur. The more spread, the more motion that occurred. If the spread is limited and the image qualifies for further analysis, deconvolution (scene stationary) can be applied to estimate the source image without motion blur.

[0041] Signal to noise ratio (SNR): The patches 125 used for linearity assessment have a known size and geometry that, once linearized, further allow the software to compute the standard deviation across all pixels, where each design is expected to have a given identical intensity. Evaluating this standard deviation, and thus the corresponding SNR across all different patches, can allow estimating the power function of the SNR as a function of the normalized and linearized intensity. This SNR can be used to correctly establish the limit of detection (LOD) of the test line as a function of the noise level.

[0042] Modulation transfer function (MTF): The focus of the image in the plane of the test strip 205 can be further assessed via horizontal and vertical MTF assessment of the two sides of the band to be detected. The modulation transfer function can be needed in conjunction with the signal to noise ratio to estimate the contrast degradation expected due to the focus and optical quality of the object of a given size. For example, if the MTF is only 25% at the spatial frequency corresponding to the expected width of the test band, there will only be 25% of the original signal present in the captured image. Thus, the MTF can be used in conjunction with the SNR to estimate the band LOD taking into account the captured image. Two regions 135 that allow X and Y MTF estimation using vertically and horizontally tilted edges can be placed symmetrically on both sides of the control and test band in order to balance the interpolation in the strip horizontal. These tilted edges are on both sides and in the same plane as the test strip 205, and the interpolated MTF from the left and right assessment is in fact the best possible estimate in the strip plane.

[0043] Belt detection capability verification: Despite best efforts to qualify, normalize, and standardize images for robust precise and accurate belt detection, additional checks can be desirable in some or all embodiments. Small test line fiducials 140 corresponding to different intensity of expected control lines and test line widths (e.g., ½ width, expected width, 2x expected width) can be added to the design to confirm correction efficacy. Post-correction, the SNR and / or contrast measured for these lines can be used as a final qualifier to confirm the LOD expected from the image.

[0044] In some embodiments, a QR code 145 or other computer readable code can be included on the 3D cartridge device 100 to direct the user to download software, unlock features of existing software, or provide other functionality or features related to imaging and / or interpretation of the test strip 205, and / or reporting of results of the test strip 205. In one example, the QR code 145 is used to load the correct version of the application and / or cartridge model of the 3D cartridge device 100. Barcodes or other computer readable codes can be present on the cartridge 200 and can be used to select and / or confirm the algorithm used for proper assessment of the particular test to be analyzed.

[0045] Arrows 150 can further be included in the cartridge area to help the user place and / or orient the cartridge correctly.

[0046] Additional corner fiducials 155 arranged in a rectangular configuration can be detected to verify the analysis window within the image and help with pitch and roll assessment.

[0047] Figure 3 and Figure 4 respectively. Graphical user interfaces (GUIs) of mobile device applications illustrating images of example 3D cartridge devices are shown. The GUIs can be implemented in an application executing on any suitable mobile device. The application can further include augmented reality features to guide image capture and / or interaction with the user. For example, as shown in Figure 1A and Figure 2A respectively. Graphical user interfaces (GUIs) of mobile device applications illustrating images of example 3D cartridge devices are shown. The GUIs can be implemented in an application executing on any suitable mobile device. The application can further include augmented reality features to guide image capture and / or interaction with the user. For example, as shown in Figure 3 and Figure 4 respectively. Graphical user interfaces (GUIs) of mobile device applications illustrating images of example 3D cartridge devices are shown. The GUIs can be implemented in an application executing on any suitable mobile device. The application can further include augmented reality features to guide image capture and / or interaction with the user. For example, as shown in

[0048] Figure 5 is an example image of a test cartridge on a 3D cartridge device taken at an angle by a mobile device camera.

[0049] Figure 6 is an example image of a test cartridge on a 3D cartridge device taken at an angle by a mobile device camera. Figure 5a de-warped version of the image of

[0050] Figure 7 is an example image of a test cartridge on a 3D cartridge background device taken at an angle by a mobile device camera.

[0051] Figure 8 is a de-warped version of the image of Figure 7

[0052] Figure 9 is a flowchart illustrating an example method 900 of performing and analyzing a test according to the present technology. The method 900 can be performed at least in part by a clinician, healthcare provider, or other trained or untrained user performing a test in conjunction with one or more devices including an imaging device and a computing device including a processor and a memory storing instructions causing the processor to perform the computer-implemented operations described herein. In some embodiments, the one or more devices can include a smartphone, tablet, digital camera, or other computing device including an imaging device, processor, and memory. The method 900 is merely one non-limiting example test method, and the systems, devices, and methods of the present technology can equally be used according to any other test method. Advantageously, aspects of the method 900 can be performed by a user without special training or expertise in performing tests. For example, the user can include an untrained operator who collects a sample and performs aspects of the method 900 in a non-clinical setting, such as the user’s home. Accordingly, embodiments of the method 900 can include collection by the individual being tested at home and display of the test results to the individual in a home collection setting. It will be understood that the method 900 is not limited to collection by the individual being tested at home, and tests of the present technology can be performed in any point-of-care (POC) setting (e.g., a doctor’s office, hospital, urgent care center, and emergency room).

[0053] The method 900 begins at block 902, where the user opens a test kit and downloads an application configured for image normalization, identification, and analysis described herein. In some embodiments, the test kit can include a background device (such as the 3D background device 100 disclosed herein), a cartridge (such as the cartridge 200), and / or one or more test strips (such as the test strip 205). The image capture and analysis application can be pre-installed on the user’s computing device or can be downloaded upon opening the kit. For example, as referenced above with respect to FIG. 1, the user can open the kit and launch the application on the user’s computing device. Figure 1A ​As described, in some embodiments, a user can capture an image of a QR code 145 or other computer-readable code located on a background device to enable a computing device to download an appropriate application. The QR code 145 or other computer-readable code on the background device can further enable the computing device to download additional supplementary data, such as background device model files and / or box model files corresponding to the background device and / or box included in the kit. It will be understood that the location of the QR code 145 is not limited to the background device 100, and the QR code can be located in any suitable location, such as, but not limited to, the kit box packaging and box 200.

[0054] At box 904, the user can complete any necessary documentation associated with the execution of the test. At box 906, the user can perform one or more operations to evaluate the imaging device, such as determining whether the imaging device and / or environmental conditions are suitable for capturing images for test analysis. At box 908, an image of the background device is captured in the absence of a box (i.e., no box on box section 104). At box 910, an application executing on the computing device obtains an image of the background device. At box 912, the application performs one or more image normalization and evaluation operations, as described in more detail below. Based on the image normalization and evaluation operations, the application can provide the user with feedback or instructions regarding the quality of the image and can guide the user to capture new images based on the feedback or instructions from the application, for example, at different angles or distances from the background device, under different lighting conditions, etc. If the image normalization and evaluation operations determine that the imaging device is suitable, method 900 proceeds to box 914 and method 900 can continue to execute and analyze the test.

[0055] At box 916, remove the sample, such as a biological sample, environmental sample, pollutant sample, etc., and apply it to the box (e.g., directly or indirectly to a container such as...). Figure 2AThe test strip 205 within the box 200 shown is, for example, a sample input well (such as when a sample is placed into the box containing the test strip). At box 918, a timestamp associated with the start of the test processing time is recorded. In one example implementation, the user is prompted to take a photograph of the background equipment before performing the test to determine if the user, imaging equipment, and environmental conditions are sufficient to perform the test, and a timer button to be selected is unlocked when the test is performed. The timer button in various example embodiments may be a software button within a graphical user interface, which, when pressed, indicates that a test is being performed and can start a counter. The counter may be configured with one or more predetermined time thresholds corresponding to the test being performed. After a predetermined time, such as at or slightly before a predetermined initial test result threshold time, an alarm or other warning may be provided to the user at box 920 to notify the user that it is time to take a photograph of the test for result determination. For example, if the test strip is configured to be read between 15 and 20 minutes after sample application to provide accurate results, the warning at box 920 may be provided 14 minutes from the timestamp to indicate that the user should be prepared to capture a result photograph within 1 minute.

[0056] At box 922, based on the warning provided at box 920, the user places the box containing the test strips on the background device and captures an image of the field of view containing the background device, the box on the background device, and the test strips within the box. Continuing to box 924, the application executing on the user's computing device determines whether the image was captured at a time greater than or equal to an initial test result threshold time (e.g., at least 15 minutes after the timestamp in this non-limiting example). If the image was not captured at least after the timestamp (at least the initial threshold time), method 900 continues to box 926, where method 900 returns to box 922. The application may instruct the computing device to wait an additional time period (e.g., the time difference between the image capture time and the initial test result threshold time) before capturing another image of the background device and the box. If the image was captured at least after the timestamp (at least the initial threshold time), the method continues to box 928.

[0057] At box 928, the application determines whether the image was captured before the test result expiration time (e.g., 20 minutes from the timestamp in this non-limiting example). If the image was not captured before the test result expiration time, method 900 terminates at box 930 and may notify the user that the test is invalid because the image acquisition occurred too late to accurately analyze the results. If the image was captured before the test result expiration time, method 900 continues to box 932. In various embodiments, the application or other software may be configured to run multiple iterations of method 900 at least partially concurrently. The application or other software may include control, tracking, and / or separate test identification features to enable tracking and verification to ensure that overlapping iterations of method 900 are paired with the correct test strip 205, such as by registration of each test strip 205 and the start clock associated with applying the sample to each test strip 205.

[0058] In boxes 932, 934, and 936, the application receives images from the imaging device and performs image normalization and identification processes, as shown in the reference... Figures 10-11B The description is more detailed, determining whether the image is appropriate, and if so, determining the test result. At box 938, the application outputs the result. For example, if the image was captured within the required time window after the test inoculation, the image identification factors are acceptable, and the test is valid (e.g., control lines are visible), then the result can be considered valid. At box 940, the test results can be reported, such as by transmitting the results to patients, doctors, health institutions, etc.

[0059] Figure 10 This diagram illustrates a flowchart of an example method 1000 for acquiring and analyzing images to determine test results according to this technology. Some or all of method 1000 can correspond to, for example, methods used for implementation. Figure 9 The image acquisition, normalization, and identification operations of blocks 908-912 and / or blocks 932-936, and the determination of results that can be reported at block 940, are performed. In some embodiments, method 1000 may be performed at least in part by an application running on a smartphone, tablet, digital camera, or other computing device including an imaging device, processor, and memory. Method 1000 is merely a non-limiting example of an image acquisition and analysis method, and the systems, devices, and methods of this technology can be used similarly according to any other test method. Various example methods of image identification and normalization may include fewer than all the operations described in method 1000, may include additional operations not described herein, and / or may include operations of method 1000 in the same or different order without departing from the scope of this disclosure. Furthermore, while method 1000 is a reference... Figures 1A-2CThe 3D background device 100, box 200 and test strip 205 illustrated in the figure are described, but the operation of method 1000 can also be implemented with any other background device, box and / or test strip.

[0060] Method 1000 begins with image acquisition at box 1002. As described above with reference to method 900, the user can acquire one or more images of a field of view including a background device (such as 3D background device 100), a box on the background device (such as box 200), and / or a test strip within the box (such as test strip 205). At box 1004, the application receives one or more images from an imaging device. In some embodiments, the acquired images include images of the background device without boxes and images of the background device including boxes. At box 1006, an image identification metric extractor can identify features of the image to be used for image identification and normalization. In some embodiments, the application can determine at box 1006 whether the image is suitable for analysis (e.g., having sufficient illumination intensity, uniformity, in focus, and / or without excessive pitch and roll). If the image is unsuitable, method 1000 can proceed to box 1008 to provide feedback to the user and possibly capture another more suitable image based on that feedback. If the image is suitable for analysis, method 1000 proceeds to box 1010.

[0061] At box 1010, the application performs an early image normalization operation, which can be performed solely on the image without using a specific background device model. For example, the application can determine the boundaries of the background device (e.g., based on a reference to one or more corners, edges, and / or viewing areas of the detected background device). After identifying the image portion corresponding to the background device, the application can perform image normalization operations such as evaluating white balance and / or detecting glare from the surfaces of the background device that may suppress accurate image analysis. At box 1012, the application can output the initially normalized image.

[0062] At box 1014, the application performs background device detection. In some embodiments, the application may detect QR codes, barcodes, or other computer-readable markings, including identifying information specifying the type of background device and / or specifying identification information corresponding to a specific background device model corresponding to the imaged background device. The application may then obtain the corresponding background device model 1016, such as from the memory of a computing device and / or from a remote computing device via a network connection. Method 1000 continues to box 1020, where a more extensive image normalization operation may be performed, at least in part, based on the image of the background device (alone or together with the received box) and the obtained background device model corresponding to the background device. The image normalization operation performed at box 1020 may include operations based on various benchmarks located on the background device (e.g., evaluations of illumination uniformity, linearity, scaling, and other aspects), such as referencing Figure 11A and Figure 11B In more detail, the image normalization operation performed at box 1020 is applied at least to the portion of the image containing the background device to produce a normalized background device image 1022.

[0063] If the image includes a box set on a background device, the method continues to box 1024, where the application performs box detection. In some embodiments, the application may detect a QR code, barcode, or other computer-readable mark, including identifying information specifying the type of box and / or specifying a particular box model corresponding to the imaged box. A box model 1026 corresponding to the imaged box can be obtained, and the application may generate a normalized box image 1028 based at least in part on the image and the box model. In some embodiments, the normalized box image 1028 may be generated based on a normalized background device image by applying further image normalization operations only to the regions in the image that include the box. Method 1000 may return to box 1006 to identify the normalized box image 1028. The application may similarly detect the position and / or type of a bar (e.g., test bar 205) set within the box at box 1030, and may obtain a bar model 1032 corresponding to the detected bar to generate a normalized bar image 1034 by applying further image normalization operations only to the regions in the image that include the bar.

[0064] Method 1000 can return to box 1006 to identify the normalized bar image 1034, and can analyze the identified normalized bar image 1036 at box 1038 using one or more band detection algorithms to detect the presence, intensity, color, and / or any other characteristics of visible bands on the test bars to determine the test result. In some embodiments, band detection may include detecting one or more control lines at box 1040, detecting one or more test lines at box 1042, determining the signal-to-noise ratio (SNR) at the test lines at box 1044, and / or determining test validity and / or positivity at box 1046. The positivity estimated at box 1046 can be performed at least in part based on a known level of detection (LOD), which can be determined based on analysis of an image of the background device under the environmental conditions present at the time of testing (e.g., lighting). For example, LOD can be determined based on: a contrast rating of pixel resolution from the background device region of the image, a test line reference 140 ( Figure 1A The detection of the modulation transfer function determined from the background device and / or the noise assessment based on linearity and / or noise determination (e.g., based on neutral patch reference 125).

[0065] Figure 11A and Figure 11B This is a flowchart illustrating an example method 1100 for performing various image identification and normalization operations according to the present technology. Some or all of methods 1100 may correspond to, for example, methods used to implement Figure 9 The image acquisition, normalization, and identification operations of blocks 908-912 and / or blocks 932-936, and the determination of results that can be reported at block 940, are performed. In some embodiments, method 1100 may be performed at least in part by an application running on a smartphone, tablet, digital camera, or other computing device including an imaging device, processor, and memory. Method 1100 is merely a non-limiting example of an image identification and normalization method, and the systems, devices, and methods of this technology can be used similarly according to any other test method. Various example methods of image identification and normalization may include fewer than all the operations described in method 1100, may include additional operations not described herein, and / or may include operations of method 1100 in the same or different order without departing from the scope of this disclosure. Furthermore, while method 1100 is a reference... Figures 1A-2C The 3D background device 100, box 200 and test strip 205 illustrated in the figure are described, but the operation of method 1100 can also be implemented with any other background device, box and / or test strip.

[0066] When image 1102 is received for application evaluation, method 1100 begins at box 1104. The field of view of image 1102 includes background devices, such as background device 100 ( Figures 1A-2C In some embodiments, image 1102 may include a background device 100 with or without box 200. At box 1104, the application evaluates the dynamic range of a portion of the image corresponding to the background device 100 (including or excluding the portion of the image corresponding to box 200). The application may evaluate the dynamic range in one, two, or all three channels of the red, green, and blue channels. In some embodiments, it may be advantageous to evaluate the dynamic range in the green channel as a reference, as the green channel may contain the most information, since most Bayer patterns contain twice as many green pixels as blue or red pixels. Furthermore, in embodiments where test strip 205 has a test strip with a micro-red or red tone control to be detected, calibrating the green and blue channels may be advantageous, as these channels tend to have maximum contrast at the location of the strip. The application may output and / or store the dynamic range 1106 corresponding to some or all of the evaluated color channels (e.g., as one or more values, such as one or more digital ranges and / or ratios), and / or may determine whether the determined dynamic range 1106 is within acceptable limits.

[0067] Method 1100 continues to box 1110, where the application performs background device detection and evaluates the white balance of the image. The application may identify a portion of image 1102 containing background device 100, such as by detecting one or more edges, corners, and / or references of background device 100. The application may determine a specific configuration of background device 100 based on a background device model 1108 corresponding to background device 100. In some embodiments, an appropriate background device model 1108 may be determined based on identifying a computer-readable identifier on background device 100 and / or by selecting a corresponding background device model 1108 from a plurality of available background device models corresponding to various background device configurations.

[0068] The application can then perform a white balance evaluation on the portion of the image containing the background device 100. In some embodiments, the white balance evaluation can be performed at least in part based on the background device model 1108. For example, the background device model 1108 can indicate which portions of the background device are equally R, G, B balanced and therefore available for white balance evaluation. Thus, the application can evaluate the white balance of the background device portion of the image by identifying any differences between the red, green, and blue channels on the R, G, B balanced regions.

[0069] At box 1112, based on background device model 1108, the application can determine whether the entire region of interest (ROI) (e.g., the entire background device 100) is visible within image 1102 (e.g., based on the detection of the four corner references 155 within image 1102). The application can also determine the detection confidence based at least in part on an evaluation of the white balance in image 1102.

[0070] At box 1114, the application determines the original resolution of image 1102, indicating the linear dimensions of the region of the background device included within each pixel of the background device portion of the image. Background device model 1108 may include a known region (e.g., length and width dimensions or a calculated region) of the imaged side of background device 100. The application may determine the number of pixels covering the background device portion of the image (e.g., based on the background device detection operation described above), and thus determine the original resolution value 1116 by dividing the square root of the known background device area by the number of pixels representing the background device in the image. Therefore, the determined original resolution value 1116 corresponds to the side length of the square region of the background device corresponding to each pixel in image 1102. The application may further determine whether the determined original resolution value 1116 is within an acceptable range.

[0071] At box 1118, the application performs one or more dedistortion operations to produce a dedistorted image 1120 suitable for further analysis. For example, based at least in part on the background device model 1108, the application may determine one or more orientation characteristics, such as the pitch, roll, and / or distance of the background device relative to the imaging device that captured image 1102. Some or all of these orientation characteristics may be determined as linear or angular deviations from a standard or optimal orientation (e.g., an orientation in which one or more edges of the background device 100 are imaged at a predetermined distance along an axis perpendicular to the front surface of the background device, provided that one or more edges of the background device 100 are parallel to the edges of the image). The application may further detect distortions in the image due to, for example, lens effects of the imaging device, artifacts of other optics within the imaging device, motion during image capture, etc. Based on the detected pitch, roll, distance, and / or other distortion characteristics, the application edits image 1102 to generate a dedistorted image 1120 for further processing. At box 1122, the application may also compare the amounts of pitch and roll in the original image and / or the dedistorted image to determine whether the pitch and roll are within acceptable limits. For example, pitch and / or roll exceeding a predetermined threshold may result in different focus characteristics at different areas of the background device and / or box, and may potentially cause partial occlusion of a portion of the image of the 3D background device and / or box.

[0072] At box 1126, the application evaluates the saturation of the background device portion of the image and flattens the illumination of the background device in the image. For example, the application may apply a correction factor including a quadratic polynomial fit to correct the illumination flatness in the image to produce a dedistorted, flattened image 1130. In some embodiments, any saturated pixel (e.g., a value of 255 corresponding to the maximum pixel intensity value in an 8-bit image) may remain unchanged independently of flattening for further glare assessment.

[0073] After correcting the illumination flatness of the image, at box 1132, the application evaluates the presence of shadows or glare on the background device. In one example, the application uses the background device model 1108 in conjunction with the dedistorted, flattened image 1130 to identify the neutral patch 125 and / or modulation transfer function region 135 on the 3D background device 100 for linearity evaluation. The appearance of these regions can be compared to the expected appearance based on the background device model 1108 to determine whether excessive shadows and / or glare exist within the image. The application can generate a result 1134 indicating the impact of any shadows or glare on the dedistorted, flattened image 1130.

[0074] At box 1136, the application evaluates the linearity of the intensity response in each color channel based on neutral patches 125. In some embodiments, each neutral patch 125 comprises a hue with equivalent white balance. Neutral patches 125 can be arranged to form a series of patches with linearly increasing intensity values, thereby enabling the evaluation of linearity for each channel. Based on the evaluation of linearity, the application can apply corrections to any detected non-linearities and can further estimate the signal-to-noise ratio (SNR) as a function of normalized and linearized intensity. The application can then output and / or save a linearity lookup table (LUT) corresponding to the SNR at various intensity values ​​at box 1138.

[0075] Continue to Figure 11BFollowing the linearity assessment at box 1136, method 1100 proceeds to box 1140 to evaluate the image's focus and any motion blur effects. For example, the presence of a defocused image and / or motion-induced blur can be detected based on a white disk 130 containing small black dots or other suitable benchmarks with relatively small detectable features. In the example of the 3D background device 100, the application can determine the lateral extent of the black dots within the white disk 130 in the x and y directions. When determining the extent of black dots in the image, the application can determine a region 1142 where pixel values ​​are within 5% of the expected intensity value based on the background device model 1108. The x and y effects of motion blur or focus can be compared to predetermined thresholds to determine whether the amount of motion blur is acceptable. In some embodiments, one of the x or y directions may have a lower threshold, for example, in the direction in which the test strip 205 is oriented, because motion blur in that direction will tend to make it difficult to determine the width of the detected band on the test strip 205.

[0076] At box 1144, the application evaluates the modulation transfer function (MTF) of the imaging device that generated image 1102. The MTF can be evaluated using region 135, which includes slanted edges configured for horizontal and vertical MTF evaluation. The MTF evaluation can be repeated on both sides of the background device 100, opposite the location of test strip 205. Region 135 can be located substantially coplanar with the surface of test strip 205, allowing the application to estimate the expected average MTF at the strip level of the control and / or test strip to be detected. The MTF as a function of spatial frequency (MTF(f)) can be derived from the slanted edges of region 135 using procedures such as those in ISO 12233 or other suitable mathematical procedures. The MTF metric 1146 can be saved for estimating the level of detection (LOD) for image-based test analysis. The application can further validate the MTF evaluation based on test strip reference 140.

[0077] At box 1148, the application estimates the LOD based on a determined MTF metric. Using box 200 with a known antigen load imaged under various imaging conditions (different phone and lighting intensities and environmental conditions), taking into account the median bandwidth, it is possible to plot the expected raw contrast of the test band once normalized according to the evaluated contrast MTF measured in the captured images. Taking into account the detection limit (e.g., 0.5%) and the known width and natural contrast of the known antigen load, the expected LOD can be estimated by considering the observed image quality (from a measurement metric using a background device benchmark). Therefore, the LOD can be determined as the theoretical antigen load such that the degraded test band contrast evaluated according to the MTF equals the detection limit. The application can output or save the estimated LOD 1150.

[0078] Method 1100 continues to box 1154, where the application performs box detection to detect box 200 located on the background device. The application may identify a portion of the dedistorted, flattened image 1130 containing box 200, such as by selecting a portion of the image corresponding to a box region indicated in the background device model 1108. The application may determine a specific configuration, test type, or other characteristic of box 200 based on box model 1152 corresponding to box 200. In some embodiments, an appropriate box model 1152 may be determined based on identifying a computer-readable identifier on box 200 and / or by selecting a corresponding box model 1152 from a plurality of available box models corresponding to various boxes compatible with background device 100. Thus, the application may determine box location 1156 and / or detection confidence of box location determination. The application may also perform one or more normalization operations on the box portions of the image based on previous image normalization and analysis performed on the background device to produce a normalized box image 1158.

[0079] At box 1160, barcode 1162 or other computer-readable identifiers are detected and read from the box to determine test information, such as the type of test, antigen, contaminant, or condition detectable by the test, test and / or control band location, and / or other information associated with determining the test result based on the image of test strip 205. In some embodiments, the barcode may contain a test identifier, and the application may enable the retrieval of test information from the memory of a computing device or from remote memory based on the test identifier.

[0080] At box 1164, the application can detect a region of the image corresponding to the test strip 205 to be analyzed. In some embodiments, the application can similarly perform one or more normalization operations on the strip portion of the image to produce a normalized strip image 1166, which can be analyzed to detect the presence of control and / or test strips. At box 1168, the application can further perform additional shadow / glare detection operations to determine whether there is any excessive shadow or glare on the test strip 205 that is not detected at the background device level. If the shadow / glare assessment result 1170 indicates that there is no significant amount of glare or shadow that would affect the ability to determine test results based on the image, the method continues to decision state 1172.

[0081] In decision state 1172, the application makes a final determination to continue test analysis. If the strip image 1166 does not have excessive glare and shadows, and the various normalization and identification operations performed at the background device, box, and test strip levels indicate that test results can be obtained from the image of test strip 205 with appropriate confidence at the appropriate LOD, then method 1100 continues to box 1174, where one or more strip detection algorithms are performed on the normalized strip image 1166 to identify the presence and location of control strips and / or any number of test strips on test strip 205, as referenced above. Figure 10 The method described in boxes 1038-1046 of method 1000.

[0082] Figures 12A-12D Another example box background device according to the present technology is shown, including exemplary non-limiting dimensions of the example device. Figure 12A It has been modified to remove the color from R, G, B blocks 120 and... Figure 1A The line drawing in the diagram represents an example background device. Figure 12B yes Figure 12A The illustration shows images of the example background device and the actual test box, both of which are shown in [image format]. Figure 2A The line graph is represented as such. Figure 12B The illustration of the example background device has been modified to remove color from R, G, B, and square 120.

[0083] The analytical apparatus described herein can accurately measure a variety of analytes of interest in many different types of samples. Samples may include specimens or cultures obtained from any source, as well as biological and environmental samples. Biological samples may be obtained from animals (including humans) and include fluids, solids, tissues, and gases. Biological samples include urine, saliva, and blood products such as plasma, serum, etc. However, such examples should not be construed as limiting the types of samples applicable to this disclosure.

[0084] In some embodiments, the sample is an environmental sample used to detect one or more analytes in the environment. In some embodiments, the sample is a biological sample from a subject. In some embodiments, the biological sample may include peripheral blood, serum, plasma, ascites, urine, cerebrospinal fluid (CSF), sputum, saliva, bone marrow, synovial fluid, aqueous humor, amniotic fluid, cerumen, breast milk, bronchoalveolar lavage fluid, semen (including prostatic fluid), Copper's fluid or preejaculate fluid, female semen, sweat, feces, hair, tears, cystic fluid, pleural and peritoneal fluid, pericardial fluid, lymph, chyme, chyle, bile, interstitial fluid, menstrual blood, pus, sebum, vomitus, vaginal secretions, mucosal secretions, fecal water, pancreatic juice, sinus lavage fluid, bronchopulmonary aspirate, or other lavage fluid.

[0085] As used herein, "analyte" generally refers to the substance to be detected. For example, analytes may include antigens, haptens, antibodies, and combinations thereof. Analytes include, but are not limited to, toxins, organic compounds, proteins, peptides, microorganisms, amino acids, nucleic acids, hormones, steroids, vitamins, drugs (including drugs for therapeutic purposes and drugs for illicit purposes), pharmaceutical intermediates or byproducts, bacteria, viral particles, and metabolites or antibodies of any of the above substances. Specific examples of some analytes include ferritin; creatinine kinase MB (CK-MB); human chorionic gonadotropin (hCG); digoxin; phenytoin; phenobarbital; carbamazepine; vancomycin; gentamicin; theophylline; valproic acid; quinidine; luteinizing hormone (LH); follicle-stimulating hormone (FSH); estradiol, progesterone; C-reactive protein (CRP); lipid transport protein; IgE antibody; cytokines; TNF-associated apoptosis-inducing ligand (TRAIL); vitamin B2 microglobulin; interferon-γ-induced protein 10 (IP-10); interferon-induced GTP-binding proteins (also known as myxovirus (influenza virus) resistance 1, MX1, MxA, IFI-78K, IFI78, MX, MX dynein, such as GTPase 1); procalcitonin (PCT); glycated hemoglobin (Glycated hemoglobin). Hb); cortisol; digitalisin; N-acetylprocainamide (NAPA); procainamide; rubella antibodies, such as rubella-IgG and rubella IgM; Toxoplasma gondii antibodies, such as Toxoplasma gondii IgG (Toxo-IgG) and Toxoplasma gondii IgM (Toxo-IgM); testosterone; salicylates; acetaminophen; hepatitis B surface antigen (HBsAg); hepatitis B core antigen antibodies, such as anti-hepatitis B core antigen IgG and IgM (Anti-HBC); Human immunodeficiency virus 1 and 2 (HIV1 and 2); human T-cell leukemia virus 1 and 2 (HTLV); hepatitis B e antigen (HBeAg); hepatitis B e antigen antibody (Anti-HBe); influenza virus; thyroid-stimulating hormone (TSH); thyroxine (T4); total triiodothyronine (total T3); free triiodothyronine (free T3); carcinoembryonic antigen (CEA); lipoproteins, cholesterol and triglycerides; and alpha-fetoprotein (AFP). Abused and controlled substances include, but are not limited to, amphetamines; methamphetamine; barbiturates such as amobarbital, secobarbital, pentobarbital, phenobarbital, and barbiturates; benzodiazepines such as nitrazepam and tranquilizers; cannabinoids such as hemp and cannabis; cocaine; fentanyl; LSD; methimazole; opioids such as heroin, morphine, codeine, hydromorphone, hydrocodone, methadone, oxycodone, hydroxymorphone, and opium; phencyclidine; and propoxy. Additional analytes may be included for the purpose of identifying biological or environmental substances of interest.

[0086] This disclosure relates to crossflow assay apparatus, testing systems, and methods for determining the presence and concentration of multiple analytes in a sample, including when one or more analytes of interest are present at high concentrations and when one or more analytes of interest are present at low concentrations. As described above, and as used herein, "analyte" generally refers to a substance to be detected, such as a protein. Examples of proteins that can be detected by the crossflow assay apparatus, testing systems, and methods described herein include, but are not limited to:

[0087] TRAIL: A TNF-related apoptosis-inducing ligand (also known as Apo2L, Apo-2 ligand, and CD253); representative RefSeq DNA sequences are NC_000003.12; NC_018914.2; NT_005612.17, and representative RefSeq protein sequence accession numbers are NP_001177871.1; NP_001177872.1; and NP_003801.1. TRAIL proteins belong to the tumor necrosis factor (TNF) ligand family.

[0088] CRP: C-reactive protein; representative RefSeq DNA sequences are NC_000001.11; NT_004487.20; and NC_018912.2, and representative RefSeq protein sequence accession number is NP_000558.2.

[0089] IP-10: Chemokinetic (CXC motif) ligand 10; representative RefSeq DNA sequences are NC_000004.12; NC_018915.2; and NT_016354.20, and the RefSeq protein sequence is NP_001556.2.

[0090] PCT: Procalcitonin is the peptide precursor of calcitonin hormone. The representative RefSeq amino acid sequence of this protein is NP_000558.2. Representative RefSeq DNA sequences include NC_000001.11, NT_004487.20, and NC_018912.2.

[0091] MX1: Interferon-induced GTP-binding protein Mx1 (also known as interferon-induced protein p78, interferon-regulated resistance GTP-binding protein, MxA). Representative RefSeq amino acid sequences of this protein are NP_001138397.1; NM_001144925.2; NP_001171517.1; and NM_001178046.2.

[0092] The transverse flow assay apparatus, testing system, and method according to this disclosure can measure TRAIL proteins in soluble and / or membrane forms. In one embodiment, only the soluble form of TRAIL is measured.

[0093] The prior description of the disclosed embodiments is provided to enable those skilled in the art to make or use the technology. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments without departing from the spirit or scope of the technology. Therefore, the technology is not intended to be limited to the embodiments shown herein, but is accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A background apparatus for use in assaying a test strip, the background apparatus comprising: a test strip portion sized and shaped to guide placement of a lateral flow assay test strip on the background apparatus; a background portion at least partially surrounding the test strip portion; a plurality of line detection fiducials disposed on the background portion, each of the plurality of line detection fiducials having a color different from a color of a surrounding area of the background portion and a width associated with an expected width of a line on the lateral flow assay test strip; and one or more additional fiducials disposed on the background portion for evaluating a modulation transfer function of an image capture device.

2. The background apparatus of claim 1, wherein the plurality of line detection fiducials includes at least a first line detection fiducial having a width substantially equal to the expected width and a second line detection fiducial having a width greater or less than the expected width.

3. The background apparatus of claim 1, wherein the plurality of line detection fiducials includes at least a first line detection fiducial and a second line detection fiducial having a first darkness of a fiducial color.

4. The background apparatus of claim 3, wherein the plurality of line detection fiducials includes at least a third line detection fiducial having a second darkness of the fiducial color that is lighter or darker than the first darkness of the fiducial color.

5. The background apparatus of claim 3, wherein the fiducial color is gray and an expected color of the line on the lateral flow assay test strip is not gray.

6. The background apparatus of claim 1, further comprising one or more motion blur detection fiducials, each of the one or more motion blur detection fiducials including a dot of a first color surrounded by an area of a second color that contrasts with the first color.

7. The background apparatus of claim 1, further comprising at least three position fiducials disposed on the background portion proximate to corners of the background portion to facilitate detecting at least one of a position, a pitch, or a roll of an image capture device relative to the background apparatus.

8. The background apparatus of claim 1, wherein the background portion includes a red-green-blue balance region having a color corresponding to equal red, green, blue values in a red-green-blue color space.

9. The background apparatus of claim 1, wherein the test strip portion includes an alignment mark configured to facilitate placement of the lateral flow assay test strip on the test strip portion of the background apparatus.

10. The background apparatus of claim 1, wherein the test strip portion includes computer readable code positioned to be covered when the lateral flow assay test strip is placed on the test strip portion, the computer readable code identifying a software application configured to analyze whether an image of the lateral flow assay test strip qualifies for further analysis based at least in part on the line detection fiducials. ​ 11. The background apparatus of claim 1, wherein the test strip portion includes computer readable code positioned to be covered when the lateral flow assay test strip is placed on the test strip portion, the computer readable code identifying a software application configured to analyze an image of the lateral flow assay test strip based at least in part on the line detection fiducials to determine a test result.

12. The background apparatus of claim 1, wherein the background apparatus is a three- dimensional background apparatus that includes one or more recesses or one or more three- dimensional features that protrude from the background portion.

13. The background apparatus of claim 1, wherein the lateral flow assay test strip is housed in a cartridge, and wherein the test strip portion is sized and shaped to guide placement of the cartridge on the test strip portion of the background apparatus.

14. The background apparatus of any of the preceding claims, wherein each of the plurality of line detection fiducials has a width associated with an expected width of at least one of: a test line that changes intensity or color when an analyte of interest is present in a sample applied to the lateral flow assay test strip; and a control line that changes intensity or color when the sample applied to the lateral flow assay test strip is present.

15. A computer-implemented method of determining a test result, the method comprising: capturing, by an image capture device, an image of a lateral flow assay test strip disposed on a test strip portion of a background apparatus, the background apparatus including a background portion at least partially surrounding the test strip portion; detecting, by one or more processors, a plurality of line detection fiducials disposed on the background portion of the background apparatus based at least in part on the image, each of the plurality of line detection fiducials having a color different from a color of a surrounding area of the background portion and a width associated with an expected width of a line on the lateral flow assay test strip; evaluating, by the one or more processors, a modulation transfer function of the image capture device based at least in part on one or more additional fiducials on the background portion of the apparatus; detecting, by the one or more processors, one or more control lines or test lines on the lateral flow assay test strip based at least in part on the plurality of line detection fiducials based at least in part on the image; and determining, by the one or more processors, a test result for the lateral flow assay test strip based at least in part on the one or more detected control lines or test lines.

16. The computer-implemented method of claim 15, further comprising, prior to capturing the image, analyzing whether an image taken by the image capture device is eligible for detection of the test result.

17. The computer-implemented method of claim 15, wherein detecting one or more control lines or test lines includes detecting a presence of the one or more control lines or test lines based at least in part on a width of at least one of the plurality of line detection fiducials.

18. The computer-implemented method of claim 15, wherein the plurality of line detection fiducials includes at least a first line detection fiducial having a first darkness of color and a second line detection fiducial having a second darkness of the color that is lighter or darker than the first darkness of the color, and wherein detecting one or more control lines or test lines includes detecting an intensity or color of the one or more control lines or test lines based at least in part on the first line detection fiducial, the second line detection fiducial, or the third line detection fiducial.

19. The computer-implemented method of claim 18, wherein the colors of the first line detection fiducial, the second line detection fiducial, and the third line detection fiducial are gray and the color of the one or more control lines or test lines is not gray.

20. The computer-implemented method of claim 15, wherein the background portion of the background device includes a red-green-blue balance region having a color corresponding to equal red, green, and blue values in a red-green-blue color space.

21. The computer-implemented method of claim 20, further comprising evaluating lighting conditions of the background device based at least in part on the red-green-blue balance region prior to determining the test result.

22. The computer-implemented method of claim 21, wherein evaluating the lighting conditions includes detecting at least one of glare or shadow on the red-green-blue balance region.

23. The computer-implemented method of claim 15, further comprising estimating a detection level based on the modulation transfer function, wherein the test result of the lateral flow assay is determined based at least in part on the estimated detection level.

24. The computer-implemented method of claim 15, further comprising detecting a motion blur level of the image based at least in part on one or more motion blur detection fiducials disposed on the background portion of the background device prior to determining the test result.

25. The computer-implemented method of claim 24, wherein each of the one or more motion blur detection fiducials includes a dot of a first color surrounded by a region of a second color that contrasts with the first color.

26. The computer-implemented method of claim 15, further comprising determining at least one of a position, pitch, or roll of the image capture device relative to the background device based at least in part on a plurality of position fiducials disposed on the background portion of the background device prior to determining the test result.

27. The background device of claim 1, wherein the one or more additional fiducials include at least one region having a color that is different from a color of a surrounding region of the background portion and at least one contrasting edge that is skewed relative to a dimension of the background device.

28. The background device of claim 27, wherein the at least one contrasting edge includes a first contrasting edge disposed at a first oblique angle for evaluation of an X-axis modulation transfer function and a second contrasting edge disposed at a second oblique angle for evaluation of a Y-axis modulation transfer function.

29. The computer-implemented method of claim 15, wherein the one or more additional fiducials include at least one region having a color different from a color of a surrounding region of the background portion and at least one contrasting edge oblique with respect to a dimension of the background device.

30. The computer-implemented method of claim 29, wherein the at least one contrasting edge includes a first contrasting edge disposed at a first oblique angle for evaluation of an X-axis modulation transfer function and a second contrasting edge disposed at a second oblique angle for evaluation of a Y-axis modulation transfer function.

Images