Method for early detection and identification of microbial-colonies, apparatus for performing the method and computer program

• US 10,643,058 B2
• Filed: 08/17/2016
• Issued: 05/05/2020
• Est. Priority Date: 09/16/2015
• Status: Active Grant

• Current Assignee: Centre National De La Recherche Scientifique, Universite De Strasbourg (Financière Meeschaert SA), Merck Patent GmbH (Merck & Co., Inc.), Millipore SAS (Merck & Co., Inc.)
• Original Assignee: Merck Patent GmbH (Merck & Co., Inc.)
• Inventors: Bouthillon, Marine, Felden, Luc, Ahmad, Ola, Collet, Christophe
• Primary Examiner(s): Azarian, Seyed H
• Application Number:
  US15/760,900
• Publication Number:
  US 20180349671A1
• Time in Patent Office: 1,357 Days
• Field of Search: 382100, 382103, 382106, 382128-134, 382162, 382168, 382172, 382173, 382181, 382190, 382199, 382219, 382224, 382232, 382254, 382260, 382274-276, 382286-291, 382305, 382312, 382321, 348 80, 348 79
• CPC Class Codes: C12Q 1/06 Quantitative determination
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/0016 involving temporal comparison
• CPC Class Codes: G06V 10/40 Extraction of image or video features
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation
• CPC Class Codes: G06V 20/698 Matching; Classification

Abstract

The present invention is directed to a method for detection of microbial colonies on a surface, comprising the steps of obtaining one or a plurality of digital images I0, I1, I2, I3, I4 of the surface, said digital images I0, I1, I2, I3, I4 being represented by at least two-dimensional matrices of pixel values, calculating a statistical noise distribution based on at least one of the digital images I0, applying the statistical noise distribution calculated in the calculation step to the one or the plurality of digital images I0, I1, I2, I3, I4, and detecting an object of interest as a candidate for a microbial colony based on deviation of pixel values from the noise distribution.

Claims

1. Method for detection of microbial colonies on a surface within an apparatus, comprising:

obtaining one or a plurality of digital images (I0, I1, I2, I3, I4) of the surface using a charge-coupled device (CCD) camera and a laser diode, said digital images (I0, I1, I2, I3, I4) being represented by at least two-dimensional matrices of pixel values;

applying a mathematical correction function to the pixel values for compensating any curvature of the surface;

calculating a statistical noise distribution based on at least one of the digital images (I0) received from the CCD camera;

applying the statistical noise distribution calculated in the calculation step to the one or the plurality of digital images (I0, I1, I2, I3, I4); and

detecting an object of interest as a candidate for a microbial colony on the surface based on deviation of pixel values from the noise distribution,

wherein the pixel values captured by the CCD camera are values representing the relative height of a point or area of the surface.

2. The method of claim 1, further comprising the step:

obtaining the plurality of digital images (I0, I1, I2, I3, I4) by imaging the same surface at different times (T0, T1, T2, T3, T4) using the CCD camera.

3. The method of claim 2 where each image (I0, I1, I2, I3, I4) obtained by imaging the same surface at different times (T0, T1, T2, T3, T4) is used to obtain a plurality of images ((I0a, I0b, I0c, . . . ), (I1a, I1b, I1c, . . . ), (I2a, I2b, I2c, . . . ), . . . ) by applying the steps of:

a) determining one of the images (I0, I1, I2, I3, I4) obtained by imaging the same surface at different times (T0, T1, T2, T3, T4) as a first digital image (Ii); and

b) obtaining any further digital image of the plurality of digital images (Iia, Iib, . . . ) by applying a respective smoothing filter to the first digital image (Ii);

c) repeating steps a) and b) for each image (I0, I1, I2, I3, I4) obtained by imaging the same surface at different times (T0, T1, T2, T3, T4).

4. The method of claim 1, wherein the object detecting process comprises:

identifying those objects of interest as microbial colonies, which show a time dependent growth based on two or more digital images (I0, I1, I2, I3, I4).

5. The method of claim 1, further comprising the step:

obtaining a first digital image (I0) by imaging the surface using the CCD camera;

and

obtaining any further digital image of the plurality of digital images (I0, I1, I2, I3, I4) by applying a respective smoothing filter to the first digital image (I0).

6. The method of claim 1, further comprising the step:

obtaining any further digital image of the plurality of digital images ((I0a, I0b, I0c, . . . ), (I1a, I1b, I1c, . . . ), (I2a, I2b, I2c, . . . ), . . . ) by applying a respective smoothing filter to the first digital images (I0, I1, I2, I3, I4).

7. The method of claim 1, wherein the statistical noise distribution is modelled as a Gaussian random field.

8. A non-transitory computer-readable medium having a computer program code encoded thereon, which when the computer program code is loaded into the internal memory of a computer will cause the computer to perform the method of claim 1.

9. An apparatus for detection of microbial colonies on a surface, comprising:

a device (1) for obtaining one or a plurality of digital images (I0, I1, I2, I3, I4) of the surface using a charge-coupled device (CCD) camera and a laser diode, said digital images (I0, I1, I2, I3, I4) being represented by at least two-dimensional matrices of pixel values;

a first processor configured to apply a mathematical correction function to the pixel values for compensating any curvature of the surface;

the first processor (3) configured to calculate a statistical noise distribution based on at least one of the digital images (I0) captured by the CCD camera;

a second processor (5) configured to apply the statistical noise distribution calculated in the calculation step to the plurality of digital images (I0, I1, I2, I3, I4); and

a detector (7) to detect an object of interest as a candidate for a microbial colony on the surface based on deviation of pixel values from the noise distribution,

wherein the pixel values captured by the CCD camera are values representing the relative height of a point or area of the surface.

10. The apparatus of claim 9, wherein the device for obtaining the plurality of digital images (I0, I1, I2, I3, I4) is configured to image the same surface at different times (T0, T1, T2, T3, T4) using the CCD camera.

11. The apparatus of claim 10, wherein the device for obtaining the plurality of digital images (I0, I1, I2, I3, I4) is configured to obtain a first digital image (I0) by imaging the surface using the CCD camera, and to obtain any further digital image of the plurality of digital images (I0, I1, I2, I3, I4) by applying a respective smoothing filter to the first digital image (I0).

12. The apparatus of claim 10, wherein the device for obtaining the plurality of digital images (I0, I1, I2, I3, I4) is configured to obtain the pixel values as the values representing the relative height of a point or area of the surface.

13. The apparatus of claim 10, wherein

the device obtains any further digital image of the plurality of digital images ((I0a, I0b, I0c, . . . ), (I1a, I1b, I1c, . . . ), (I2a, I2b, I2c, . . . ), . . . ) by applying a respective smoothing filter to the first digital images (I0, I1, I2, I3, I4).

14. The apparatus of claim 10, wherein the apparatus is configured to:

a) determine one of the images (I0, I1, I2, I3, I4) obtained by imaging the same surface at different times (T0, T1, T2, T3, T4) as a first digital image (I0); and

b) obtain any further digital image of the plurality of digital images (I0, I1, I2, I3, I4) by applying a respective smoothing filter to the first digital image (I0);

c) repeat steps a) and b) each image (I0, I1, I2, I3, I4) obtained by imaging the same surface at different times (T0, T1, T2, T3, T4).