# US 12,006,529 B2 · Method and system for identifying the gram type of a bacterium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,006,529 B2 - Document ID: US12006529B2 - Status: Granted - Kind: B2 - Application number: 16/772,402 - Filed: 2018-12-20 - Published: 2021-03-18 - Granted: 2024-06-11 - Earliest claimed: 2017-12-21 - Assignees: Biomerieux - Applicants: Biomerieux - Inventors: Jordane Lallemand; Denis Leroux; Manuel Petit - Web page: https://patentgrep.com/patent/US12006529B2 ## CPC - C12Q 1/04: Determining presence or kind of microorganism; Use of selective media for testing antibiotics or bacteriocides; Compositions containing a chemical indicator therefor {(C12Q1/6897 takes precedence)} - G01J 3/2803: {using photoelectric array detector} - G01J 3/2823: {Imaging spectrometer} - G01N 21/31: Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry {(G01N21/72 takes precedence)} - G06V 10/58: relating to hyperspectral data - G06V 20/698: {Matching; Classification} - G01N 15/01: specially adapted for biological cells, e.g. blood cells (investigating sedimentation of particle suspensions in blood G01N15/05) - G01N 2201/129: Using chemometrical methods - C12Q: MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS (immunoassay G01N33/53); COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES - G01J: MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY (light sources F21, H01J, H01K, H05B; investigating properties of materials by optical means G01N) - G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract The invention relates to a method for detecting the Gram type and the fermenting character of a bacterial strain, said method comprising: illuminating, in the wavelength range of 390 nm to 900 nm, at least one bacterium from said strain which has a natural electromagnetic response within said range; acquiring, within said range, a light intensity reflected by, or transmitted through, said illuminated bacterium; and determining the Gram type and the fermenting character of the bacterial strain according to the light intensity acquired within said range. ## Claims ### Claim 1 A process for detecting the Gram type and the fermentative character of a bacterial strain, comprising: illuminating in a wavelength range 390 nm-900 nm at least one bacterium of the strain having a natural electromagnetic response in the range, wherein the illuminating covers one or more wavelength ranges, wherein the one or more wavelength ranges cover at least two distinct spectral bands whose center wavelengths are more than 50 nm apart; acquiring, within the range, a luminous intensity reflected from or transmitted through the illuminated bacterium; and determining the Gram type and the fermentative character of the bacterial strain as a function of the luminous intensity acquired in the range, by applying an automated prediction model, wherein the prediction model includes at least ne class identifying the Gram type of the bacterial strain, wherein the prediction model includes at least ne class identifying the fermentative character of the bacterial strain, wherein the prediction model does not include a class identifying the microorganism at a species level. ### Claim 2 (depends on 1) The detection process as claimed in claim 1, wherein the determination of Gram type and fermentative character comprises applying a first classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a fermentative Gram-negative bacterial strain. ### Claim 3 (depends on 1) The detection process as claimed in claim 1, wherein the determination of the Gram type and the fermentative character comprises applying at least one selected from the group of: a classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a Gram-positive bacterial strain, and a classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a Gram-negative bacterial strain. ### Claim 4 (depends on 1) The detection process as claimed in claim 1, wherein the determination of the Gram type and the fermentative character further comprises applying a classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a yeast or a microbiological test distinguishing the bacterial strain from a yeast. ### Claim 5 (depends on 2) The detection process as claimed in claim 2, wherein: the illumination and acquisition are performed directly on a sample comprising a colony of the bacterial strain and a nutrient medium on which the colony has grown, the nutrient medium being a blood agar; if the acquired luminous intensity is not the luminous intensity of a fermentative Gram-negative bacterial strain, the determination of the Gram type and fermentative character further comprises applying a second classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a Gram-positive bacterial strain. ### Claim 6 (depends on 5) The detection process as claimed in claim 5, wherein, if the acquired luminous intensity is not the luminous intensity of a Gram-positive bacterial strain, the determination of the Gram type and of the fermentative character further comprises applying a third classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a yeast. ### Claim 7 (depends on 6) The process as claimed in claim 6, wherein the first classification predicting whether the acquired luminous intensity is the luminous intensity of a Gram-negative and fermentative bacterial strain is a classification distinguishing the luminous intensity of Gram-negative and fermentative bacterial strains from the luminous intensity of the group consisting of Gram-negative and non-fermentative bacterial strains, Gram-positive bacterial strains, and yeasts; the second classification predicting whether the acquired luminous intensity is the luminous intensity of a Gram-positive bacterial strain is a classification distinguishing the luminous intensity of Gram-positive bacterial strains from the luminous intensity of the group consisting of negative and non-fermentative bacterial strains, and yeasts; the third classification predicting whether the acquired luminous intensity is the luminous intensity of a yeast is a classification distinguishing the luminous intensity of yeasts from the luminous intensity of the group consisting of Gram-negative and non-fermentative bacterial strains. ### Claim 8 (depends on 4) The process as claimed in claim 4, wherein, if the acquired luminous intensity is not the luminous intensity of a yeast, the determination of the Gram type and of the fermentative character further comprises applying a classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a Gram-positive bacterial strain. ### Claim 9 (depends on 8) The process as claimed in claim 8, wherein, if the acquired luminous intensity is not the luminous intensity of a Gram-positive bacterial strain, the determination of the Gram type and of the fermentative character further comprises applying a classification of the prediction model predicting whether the luminous intensity is the luminous intensity of a fermentative Gram-negative bacterial strain or of a non-fermentative Gram-negative bacterial strain. ### Claim 10 (depends on 9) The process as claimed in claim 9, wherein: the classification predicting whether the acquired luminous intensity is the luminous intensity of a yeast is a classification of the prediction model distinguishing the luminous intensity of yeasts from the luminous intensity of the group consisting of Gram-positive bacterial strains, Gram-negative and non-fermentative bacterial strains, and Gram-negative and fermentative bacterial strains; the classification predicting whether the luminous intensity is the luminous intensity of a Gram-positive bacterial strain is a classification of the prediction model distinguishing the luminous intensity of Gram-positive bacterial strains from the luminous intensity of the combination of Gram-negative and non-fermentative bacterial strains and Gram-negative and fermentative bacterial strains; the classification predicting whether the luminous intensity is the luminous intensity of a fermentative Gram-negative bacterial strain or a non-fermentative Gram-negative bacterial strain is a classification of the prediction model distinguishing the luminous intensity of Gram-negative non-fermentative bacterial strains from the luminous intensity of the group consisting of Gram-negative and fermentative bacterial strains. ### Claim 11 (depends on 3) The process as claimed in claim 3, wherein: the illumination and acquisition are performed directly on a sample comprising a colony of the bacterial strain and a nutrient medium on which the colony has grown, the nutrient medium being Tryptone-Soy agar; if the acquired luminous intensity is not that of a Gram-positive bacterial strain, the determination of the Gram type and the fermentative character further comprises applying a classification of the prediction model predicting whether the acquired luminous intensity is the luminous intensity of a yeast. ### Claim 12 (depends on 11) The process as claimed in claim 11, wherein, if the acquired luminous intensity is not the luminous intensity of a yeast, the determination of the Gram type and of the fermentative character further comprises applying a classification of the prediction model predicting whether the luminous intensity is the luminous intensity of a fermentative Gram-negative bacterial strain or of a non-fermentative Gram-negative bacterial strain. ### Claim 13 (depends on 12) The process as claimed in claim 12, wherein: the classification predicting whether the luminous intensity is the luminous intensity of a Gram-positive bacterial strain is a classification distinguishing the luminous intensity of Gram-positive bacterial strains from the luminous intensity of the group consisting of Gram-negative and non-fermentative bacterial strains, Gram-negative and fermentative bacterial strains, and yeasts; the classification predicting whether the acquired luminous intensity is the luminous intensity of the yeast is a classification distinguishing the luminous intensity of the yeast from the luminous intensity of the group consisting of negative and non-fermentative bacterial strains and negative and fermentative bacterial strains; the classification predicting whether the luminous intensity is the luminous intensity of a fermentative Gram-negative bacterial strain or a non-fermentative Gram-negative bacterial strain is a classification distinguishing the luminous intensity of Gram-negative non-fermentative bacterial strains from the luminous intensity of the group consisting of Gram-negative and fermentative bacterial strains. ### Claim 14 (depends on 2) The process as claimed in claim 2, wherein each of the classifications is learned from hyperspectral images in the range 390 nm-900 nm and according to an approach comprising stepwise increasing a set of spectral channels used in the classification until a predetermined accuracy threshold or a predetermined maximum number of channels is reached. ### Claim 15 (depends on 2) The detection process as claimed in claim 2, wherein the first classification distinguishes between luminous intensities in the wavelength range 470 nm-500 nm and the wavelength range 575-645 nm only. ### Claim 16 (depends on 3) The detection process as claimed in claim 3, wherein the applied classification distinguishes between luminous intensities as a function of the wavelength range 535 nm-675 nm and the wavelength range 850 nm-875 nm only. ### Claim 17 (depends on 5) The detection process as claimed in claim 5, wherein the second classification distinguishes luminous intensities according to the wavelength range 415 nm-500 nm and the wavelength range 535 nm-675 nm only. ### Claim 18 (depends on 1) The detection process as claimed in claim 1, wherein the luminous intensity is acquired on a number of spectral channels less than or equal to 24. ### Claim 19 (depends on 5) The detection process as claimed in claim 5, wherein the luminous intensity is acquired on a number of spectral channels less than or equal to 5 for each of the first and second classifications. ### Claim 20 (depends on 1) The detection process as claimed in claim 1, wherein the acquisition of the luminous intensity comprises the acquisition of a hyperspectral or multispectral image of a colony of bacteria of the strain, and wherein the luminous intensity is determined as a function of at least one pixel of the image corresponding to the colony. ### Claim 21 (depends on 20) The detection process as claimed in claim 20, wherein a respective first prediction of the Gram type and the fermentative character is carried out as a function of the light intensity of each pixel of a set of pixels of the colony, and wherein the Gram type and the fermentative character are determined by a majority vote of results of the first predictions. ### Claim 22 (depends on 21) The detection process as claimed in claim 21, wherein the majority vote is a 70% or more pixel vote. ### Claim 23 (depends on 1) A process for producing an antibiogram of a bacterial strain to an antibiotic comprising: determining the Gram type and the fermentative character of the bacterial strain according to the process as claimed in claim 1; providing at least one sample comprising the bacterial strain, a culture medium and a concentration of an antibiotic as a function of the Gram type and the selected fermentative character; and determining the sensitivity of the bacterial strain to the antibiotic as a function of the growth of the strain in the sample. ### Claim 24 (depends on 1) A process for identifying a bacterial strain to an antibiotic comprising: determining the Gram type and the fermentative character of the bacterial strain according to the process as claimed in claim 1; selecting at least one colorimetric medium according to the Gram type and the selected fermentative character; and culturing the bacterial strain in the medium. ### Claim 25 A system for detecting the Gram type and the fermentative character of a bacterial strain, comprising: an illumination device configured to illuminate, in a wavelength range 390 nm-900 nm, at least one bacterium of the strain; a sensor configured to acquire, in the 390 nm-900 nm range, a luminous intensity reflected from or transmitted through the illuminated bacterium, wherein the illuminating by the sensor covers one or more wavelength ranges, wherein the one or more wavelength ranges cover at least two distinct spectral bands whose center wavelengths are more than 50 nm apart; and a computer unit configured to determine the Gram type and the fermentative character of the bacterial strain as a function of the luminous intensity acquired in the 390 nm-900 nm range, by applying an automated prediction model, wherein the prediction model includes at least one class identifying the Gram type of the bacteria strain, wherein the prediction model includes at least one class identifying the fermentative character of the bacterial strain, wherein the prediction model does not include a class identifying the microorganism at a species level. ### Claim 26 (depends on 25) The system as claimed in claim 25, configured to implement a process for detecting the Gram type and the fermentative character of a bacterial strain, comprising: illuminating in the wavelength range 390 nm-900 nm at least one bacterium of the strain having a natural electromagnetic response in the range, wherein the illuminating covers one or more wavelength ranges. wherein the one or more wavelength ranges cover at least two distinct spectral bands whose center wavelengths are more than 50 nm apart; acquiring, within the range, a luminous intensity reflected from or transmitted through the illuminated bacterium; and determining the Gram type and the fermentative character of the bacterial strain as a function of the acquired luminous intensity in the range, wherein the determination of Gram type and fermentative character comprises applying a classification of the prediction model predicting whether the acquired luminous intensity is the light intensity of a fermentative Gram-negative bacterial strain. ### Claim 27 (depends on 25) The system as claimed in claim 25, configured to illuminate, and acquire the image of, a sample comprising a colony of bacteria of the strain and a nutrient medium on which the colony has grown.