Method of determining the concentration of an analyte in a sample of a body fluid using a camera and a color reference card

Patent metadata

Field	Value
Publication number	US-12542859-B2
Application number	US-202218063492-A
Country	US
Kind code	B2
Filing date	Dec 8, 2022
Priority date	Jun 9, 2020
Publication date	Feb 3, 2026
Grant date	Feb 3, 2026

Abstract

A method of determining the concentration of an analyte in a sample of a body fluid with a mobile device having a camera is disclosed. The camera captures an image of a color reference card and of a reagent test field of an optical test strip having a sample applied to it. A predetermined pixel-based mean tone map correction is applied to the image obtained, which results in a first intensity-corrected image. Local brightness information is derived from the first intensity-corrected image. A mobile device-specific tone map correction is applied to the first intensity-corrected image, taking into account the local brightness information. A second intensity-corrected image is thereby obtained. Analyte concentration is determined based on a color formation reaction of the test field by using the second intensity-corrected image. Optionally, a color correction may be derived and applied to the second intensity-corrected image to obtain an intensity-corrected and color-corrected image.

First claim

What is claimed is:

1 . A method of determining the concentration of an analyte in a sample of a body fluid using a mobile device having a camera, the method comprising:
i) using the camera to capture an image of at least a part of a color reference card and of at least a part of a reagent test field of an optical test strip having the sample applied thereto, wherein the test field is in a defined position in the image relative to the color reference card and the color reference card has a first plurality of different gray reference fields locally assigned to the test field, wherein the first plurality of different gray reference fields and the test field are locally assigned to each other by being placed in neighboring positions, or wherein the first plurality of different gray reference fields locally assigned to the test field are arranged on the color reference card such that the first plurality of different gray reference fields surrounds the test field, wherein the color reference card has a plurality of different color reference fields having known reference color values and a second plurality of different gray reference fields locally assigned to the plurality of different color reference fields, and wherein the second plurality of different gray reference fields and the plurality of different color reference fields are locally assigned to each other by being placed in neighboring positions or wherein the second plurality of different gray reference fields locally assigned to the plurality of different color reference fields are arranged on the color reference card such that the second plurality of different gray reference fields surrounds the plurality of different color reference fields, ii) applying a predetermined pixel-based mean tone map correction to the image obtained in step i), thereby obtaining a first intensity-corrected image, wherein the predetermined pixel-based mean tone map correction comprises an assignment of a second brightness value to a first brightness value, wherein the first brightness value is recorded by the camera, wherein each pixel of the recorded image is corrected individually by the predetermined pixel-based mean tone map correction; iii) deriving, from the first intensity-corrected image, local brightness information for at least one of the plurality of different color reference fields and for the test field by using gray reference fields of the first plurality of different gray reference fields and the second plurality of different gray reference fields which are locally assigned to the at least one color reference field and the test field, respectively, wherein the local brightness information comprises a numerical indication describing the local intensity of at least one RGB color of the at least one color reference field and the test field, respectively; iv) applying a mobile device-specific tone map correction to the first intensity-corrected image taking into account the local brightness information and thereby obtaining at least one second intensity-corrected image; and v) determining the analyte concentration based on a color formation reaction of the test field by using the second intensity-corrected image.

2 . The method according to claim 1 , further comprising:
vi) deriving a color correction by using at least one of the color reference fields in the second intensity-corrected image; and vii) applying the color correction to the second intensity-corrected image and thereby obtaining an intensity-corrected and color-corrected image.

3 . The method according to claim 2 , further comprising, prior to step vi), local white balancing the at least one of the color reference field of step vi) and the test field.

4 . The method according to claim 2 , wherein, in step vi), the color correction is derived using a selection of the color reference fields based on at least one of:
selecting the color reference fields in a predetermined color subspace of colors similar to the color formation reaction of the test field; dynamically selecting the color reference fields by selecting color reference fields having a color value similar to the color formation reaction of the test field; weighting the color reference fields differently; and selecting the color reference fields in accordance with an accuracy of the color correction.

5 . The method according to claim 1 , wherein the predetermined pixel-based mean tone map correction comprises: a mean tone map correction derived by combining a plurality of tone map corrections for different types of mobile devices, or an inverse of a standard Gamma correction.

6 . The method according to claim 1 , further comprising performing a statistical modification of the first intensity-corrected image to produce a statistically modified first intensity-corrected image, wherein step iv) is performed on the basis of the statistically modified first intensity-corrected image.

7 . The method according to claim 1 , wherein the first plurality of different gray reference fields locally assigned to the test field surround the test field and form a test field group.

8 . The method according to claim 7 , wherein the test field group comprises a plurality of redundant gray reference fields having at least three different gray levels, wherein at least two of the redundant gray reference fields are provided for each gray level, wherein the redundant gray reference fields are arranged symmetrically around the test field, wherein, in step iii), the local brightness information for the test field is determined by averaging over redundant gray reference fields having a common gray level.

9 . The method according to claim 1 , wherein the second plurality of different gray reference fields locally assigned to the plurality of different color reference fields surround each of the plurality of different color reference fields and wherein each color reference field and the gray reference fields locally assigned thereto form a local color reference field group.

10 . The method according to claim 9 , wherein each local color reference field group comprises a plurality of redundant gray reference fields having at least three different gray levels, wherein at least two of the plurality of redundant gray reference fields are provided for each gray level, wherein the plurality of redundant gray reference fields are arranged symmetrically around the color reference field of each local color reference field group.

11 . The method according to claim 10 , wherein, in step iii), the local brightness information for a respective local color reference field group is determined by averaging over the plurality of redundant gray reference fields of the respective local color reference field group having a common gray level.

12 . A mobile device having a camera and a processor, the mobile device configured for performing a method of determining the concentration of an analyte in a sample of a body fluid wherein the method comprises:
i) using the camera to capture an image of at least a part of a color reference card and of at least a part of a reagent test field of an optical test strip having the sample applied thereto, wherein the test field is in a defined position in the image relative to the color reference card and the color reference card has a first plurality of different gray reference fields locally assigned to the test field, wherein the first plurality of different gray reference fields and the test field are locally assigned to each other by being placed in neighboring positions, or wherein the first plurality of different gray reference fields locally assigned to the test field are arranged on the colo

Assignees

• Roche Diabetes Care Inc

Inventors

• Berg Max
• Hailer Fredrik
• Limburg Bernhard
• Tuerck Volker
• Winkelnkemper Momme

Classifications

• G06T2207/10024

  Color image · CPC title

• H04N9/73

  Colour balance circuits, e.g. white balance circuits or colour temperature control (camera processing pipelines for colour balance H04N23/88) · CPC title

• G06V10/60

  relating to illumination properties, e.g. using a reflectance or lighting model · CPC title

• G06T5/92

  based on global image properties · CPC title

• A61B5/145

  Measuring characteristics of blood in vivo, e.g. gas concentration or pH-value {; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid or cerebral tissue} (non-radiation detecting or locating of foreign bodies in blood A61B5/06) · CPC title