Mobile chemical analysis

Patent metadata

Field	Value
Publication number	US-11060968-B2
Application number	US-201815941692-A
Country	US
Kind code	B2
Filing date	Mar 30, 2018
Priority date	Mar 30, 2018
Publication date	Jul 13, 2021
Grant date	Jul 13, 2021

Abstract

A target sample to be chemically analyzed is received at a microfluidic device which is part of an analysis card. The microfluidic device includes at least one input layer configured to receive the target sample, at least one intermediary layer, and at least one readout layer configured to present one or more color attributes at one or more readout regions in response to one or more colorimetric reactions to the target sample. An image of the readout layer of the microfluidic device is obtained. A calibration file that corresponds to the analysis card is obtained. The calibration file includes a calibrated model of at least one color attribute based on data from at least one known chemical attribute of at least one standard sample processed by a calibration card. A data processing function is executed on the obtained image by comparing the one or more color attributes at the one or more readout regions to the calibrated model of the at least one color attribute to determine at least one chemical attribute of the target sample.

First claim

What is claimed is:

1. A method comprising:
receiving a target sample to be chemically analyzed at a microfluidic device which is part of an analysis card, wherein the microfluidic device comprises at least one input layer configured to receive the target sample, at least one intermediary layer, and at least one readout layer configured to present one or more color attributes at one or more readout regions in response to one or more colorimetric reactions to the target sample;
obtaining one or more images of the readout layer of the microfluidic device;
obtaining a calibration file that corresponds to the analysis card, wherein the calibration file comprises at least one calibrated model of one or more color attributes based on data from at least one known chemical attribute of at least one standard sample processed by a calibration card; and
determining at least one chemical attribute of the target sample utilizing at least one data processing function on the obtained image, wherein the data processing function comprises comparing the one or more color attributes at the one or more readout regions in the obtained image of the readout layer to the calibrated model of one or more color attributes of the obtained calibration file based on the data from at least one known chemical attribute of at least one standard sample processed by the calibration card;
wherein obtaining the calibration file that corresponds to the analysis card further comprises:
obtaining a code from the analysis card; and
retrieving the calibration file from a set of calibration files that corresponds to the code obtained from the analysis card; and
wherein one or more steps of the method are performed by at least one processing device.

2. The method of claim 1 , wherein the code comprises a unique identifier for tracking the analysis card.

3. The method of claim 1 , wherein the code comprises a quick-response code, bar code, radio frequency identification chip content or plain text.

4. The method of claim 1 , wherein the code comprises information corresponding to a manufacturing lot of the analysis card.

5. The method of claim 1 , wherein the code comprises information corresponding to at least one type of test performable on a sample applied to the analysis card.

6. The method of claim 1 , further comprising adding information comprising temporal, geographical or user data.

7. The method of claim 1 , wherein the calibration file is generated prior to chemical analysis of the target sample using a calibration procedure.

8. The method of claim 7 , wherein the calibration procedure comprises:
applying at least one standard sample to a calibration card, wherein the calibration card comprises one or more readout regions that present one or more color attributes in response to one or more colorimetric reactions to the standard sample, and wherein the one or more color attributes correlate to at least one chemical attribute associated with the at least one standard sample;
creating the at least one calibrated model of the one or more color attributes of the readout regions in the calibration card for the standard sample; and
storing the at least one calibrated model of the one or more color attributes as data in a calibration file for retrieval during chemical analysis of the target sample.

9. The method of claim 8 , wherein the calibration card comprises a circle matrix wherein each circle in the circle matrix is pre-treated with detection chemistry for use in detecting a given chemical element.

10. The method of claim 9 , wherein two or more circles in the circle matrix are configured to detect two or more attributes of the same chemical element.

11. The method of claim 8 , wherein the at least one calibrated model comprises a machine learning algorithm.

12. The method of claim 1 , wherein the at least one data processing function is executed on a mobile device.

13. The method of claim 12 , wherein the mobile device that executes the at least one data processing function also performs the step of obtaining the image of the readout layer of the microfluidic device.

14. The method of claim 1 , wherein the at least one data processing function is executed on a computing platform remote from the microfluidic device.

15. The method of claim 14 , wherein the computing platform comprises a cloud computing platform.

16. The method of claim 1 , wherein determining the at least one chemical attribute of the target sample utilizing the at least one data processing function on the image of the readout layer of the microfluidics device further comprises:
generating at least one location mask based on a given readout region and a size of the given readout region;
generating at least one masked image based on the at least one location mask and the size of the given readout region; and
determining at least one color space value based on the at least one masked image.

17. The method of claim 1 , wherein the intermediary layer of the microfluidic device further comprises one or more analysis layers operatively disposed between the input layer and the readout layer.

18. The method of claim 17 , wherein the analysis layer is configured to provide one or more colorimetric analyses on the target sample received at the input layer such that the one or more color attributes are presented at the readout layer in response to the colorimetric reaction.

19. The method of claim 16 , wherein the data processing function determines color gradient or spatial color distribution properties.

20. The method of claim 6 , wherein the calibration file selection depends on a sensor reading on the analysis card.

21. An article of manufacture comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by the processing device causes the processing device to perform the steps of claim 1 .

22. A system comprising:
one or more processing devices, wherein each processing device comprises a memory and a processor operatively coupled to the memory, and wherein the one or more processing devices are configured to:
obtain an image of a readout layer of a microfluidic device, wherein the microfluidic device is part of an analysis card and comprises at least one input layer configured to receive a target sample to be chemically analyzed, at least one intermediary layer, and at least one readout layer configured to present one or more color attributes at one or more readout regions in response to one or more colorimetric reactions to the target sample;
obtain a calibration file that corresponds to the analysis card, wherein the calibration file comprises at least one calibrated model of one or more color attributes based on data from at least one known chemical attribute of at least one standard sample processed by a calibration card; and
determine at least one chemical attribute of the target sample utilizing at least one data processing function on the obtained image, wherein the data processing function comprises comparing the one or more color attributes at the one or more readout regions in the obtained image of the readout layer to the calibrated model of one or more color attributes of the obtained calibration file based on the data from at least one known chemical attribute of at least one standard sample processed by the calibration card;
wherein the one or more processing devices are further configured to obtain the calibration fi

Assignees

• IBM

Inventors

• Ohta Ricardo Luis
• Silva Ademir Ferreira Da
• De Oliveira Botelho Andre
• Ferreira Matheus Esteves
• Tirapu Azpiroz Jaione
• Engel Michael
• Steiner Mathias

Classifications

• G01N2201/126

  Microprocessor processing · CPC title

• G01N2201/13

  Standards, constitution · CPC title

• G01N2201/0221

  Portable; cableless; compact; hand-held · CPC title

• G01N21/251

  Colorimeters; Construction thereof · CPC title

• G01N2201/102

  Video camera · CPC title