Software and algorithms for use in remote assessment of disease diagnostics
A method to assess a subject diagnosis is provided. The method includes receiving an image from an image-capturing device, the image comprising an area of interest in a test cartridge, and finding a border of the area of interest of the test cartridge and applying a geometrical transformation on an area delimited by the border of the test cartridge to bring the image of the area of interest in the test cartridge to a selected size and a selected shape. The method also includes identifying a target region within the area of interest of the test cartridge, evaluating a quality of the image based on a characteristic feature of the target region, and providing commands to adjust an optical coupling in the image-capturing device when the quality of the image is lower than a selected threshold.
1. A computer-implemented method, comprising:
receiving one or more images from an image-capturing device, the one or more images comprising an area of interest in a test cartridge;
finding a border of the area of interest of the test cartridge and applying a geometrical transformation on an area delimited by the border of the test cartridge to bring an image of the area of interest in the test cartridge to a selected size and a selected shape;
identifying a target region within the area of interest of the test cartridge;
evaluating a quality of the one or more images based on a characteristic feature of the target region;
providing commands to adjust an optical coupling and a frequency rate for collecting the one or more images in the image-capturing device when the quality of the one or more images is lower than a selected threshold; and
when the quality of the one or more images meets the selected threshold, providing the one or more images to a processor that contains software designed to assess a subject diagnostics based on a digital analysis of the one or more images.
2. The computer-implemented method of claim 1 , wherein evaluating the quality of the one or more images further comprises evaluating a second image, wherein the second image is one of a background sample or a blank sample.
3. The computer-implemented method of claim 1 , further comprising normalizing an intensity value for a pixel in an image of the area of interest relative to a selected intensity value from multiple pixels in the image of the area of interest.
4. The computer-implemented method of claim 1 , wherein the target region comprises a process control area including at least one of a positive control area or a negative control area, and evaluating a quality of the one or more images comprises evaluating a signal intensity in the process control area.
5. The computer-implemented method of claim 1 , wherein identifying a target region within the area of interest comprises identifying at least a test line and a control line in the area of interest of the test cartridge within a field of view of the one or more images.
6. The computer-implemented method of claim 1 , further comprising displaying the one or more images in a computer display, and including a viewing guide in the computer display, the viewing guide overlapping at least a portion of the digital analysis of the one or more images.
7. The computer-implemented method of claim 1 , further comprising displaying a test result in a computer display, and not displaying the one or more images.
8. The computer-implemented method of claim 1 , wherein evaluating a quality of the one or more images comprises comparing a selected feature of the one or more images with a value associated with selected features of multiple images having known quality values.
9. A computer-implemented method, comprising:
receiving a first image from an image-capturing device, the first image comprising an area of interest in a test cartridge, at a first time;
receiving a second image from the image-capturing device, the second image comprising a same area of interest in the test cartridge, at a second time;
forming a volume of interest, having a time dimension between the first image and the second image;
associating a dynamic value for an assay in the test cartridge based on the time dimension;
providing a first identifier code identifying the image-capturing device to a processor that contains software designed to assess a subject diagnostics based on a digital analysis of the first image;
identifying a target region within the area of interest of the test cartridge;
evaluating a quality of the first image based on a characteristic feature of the target region, the dynamic value for the assay, and on the first identifier code; and
when a quality of the first image meets a selected threshold, providing the first image to the processor.
10. The computer-implemented method of claim 9 , further comprising retrieving a calibration table from a remote server using the first identifier code, the calibration table associated with the image-capturing device, and indicative of a signal value that is a threshold for evaluating the quality of the first image.
11. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises selecting a threshold for a signal intensity in a process control area within the target region based on the first identifier code.
12. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises verifying that a signal intensity at an end of a test channel in the test cartridge is higher than a selected threshold indicative that a sample flowed to the end.
13. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises verifying that a reference line appears at a selected location.
14. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises verifying that a signal intensity of a negative control is less than a selected threshold indicative of an assay interference.
15. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises verifying an exposure and a focus.
16. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises assessing whether the first image is appropriate to send to an artificial intelligence model for inferencing.
17. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises verifying that a valid crop in the first image contains features from a test cassette.
18. The computer-implemented method of claim 9 , wherein evaluating a quality of the first image comprises monitoring a physical attribute of the image-capturing device.
19. The computer: implemented method of claim 18 , wherein the physical attribute is internal temperature of the image-capturing device.
20. A computer-implemented method, comprising:
retrieving a first image associated with an assay in a test cartridge carrying a biological sample from a user for a disease diagnostic;
receiving a second image comprising a same area of interest in the test cartridge, at a second time;
forming a volume of interest, having a time dimension between the first image and the second image;
selecting a digital portion including the volume of interest;
modifying, with a model, the digital portion of the first image to obtain a weighted value of the digital portion associating a dynamic value for an assay in the test cartridge based on the time dimension;
determining, based on the weighted value of the digital portion and the model, a diagnostic value; and
determining a certainty level for the diagnostic value based on a second weighted value from a second digital portion of the second image and the dynamic value for the assay in the test cartridge.
21. The computer-implemented method of claim 20 , wherein retrieving the first image comprises receiving an image from a client device via a remote network communication channel.
22. The computer-implemented method of claim 20 , wherein retrieving the first image comprises accessing a database including multiple images of multiple assays including different biological samples from multiple users.
23. The computer-implemented method of claim 20 , further comprising updating the model when the certainty level for the diagnostic value is less than a predetermined value.
24. The computer-implemented method of claim 20 , wherein modifying the digital portion of the first image comprises convoluting a value of the digital portion of the first image with multiple values of adjacent digital portions of the first image according to a weighting coefficient in the model.
25. The computer-implemented method of claim 20 , wherein the assay is an immunoassay.
26. The computer-implemented method of claim 20 , wherein the assay is a lateral flow immunoassay.
27. The computer-implemented method of claim 20 , wherein the assay is a molecular diagnostic assay.
28. The computer-implemented method of claim 27 , wherein the molecular diagnostic assay comprises amplification of a molecular target and binding of the molecular target to a surface for visualization.