System and methods for qualifying medical images
Disclosed is a system for qualifying medical images submitted by user for diagnostic analysis comprising: an image input module, configured to receive one or more image input by a user; an image protocol conformation module, configured to receive the one or more images from the image input module, and further configured to analyze each of the one or more images for conformity with a predefined protocol and wherein images that do not conform to the predefined protocol are flagged as non-conforming images; an image output module, configured to identify to the user each of the one or more images flagged as non-conforming and prompting the user to resubmit a new image for each of the non-conforming images; and an image resubmission module, configured to receive the user resubmitted image and provide the resubmitted image to the image protocol conformation module.
1. A system for qualifying medical images submitted by a user for diagnostic analysis comprising:
a non-transitory computer-readable medium comprising memory with instructions encoded thereon; and
one or more processors configured to, when executing the instructions, perform operations comprising:
receiving a plurality of retinal images input by a user, each of the plurality of retinal images capturing a different portion of a retina, each different portion of a retina corresponding to a respective predefined protocol requiring an anatomical object to be at a particular location within the respective retinal image, the respective predefined protocol being of a plurality of predefined protocols;
determining that a retinal image of the plurality of retinal images is a non-conforming image based on the retinal image not conforming to its respective predefined protocol, while determining that other ones of the plurality of retinal images conform to their respective predefined protocols;
generating for display a graphical interface to the user, the graphical interface comprising:
for each respective predefined protocol of the plurality of predefined protocols, a respective section comprising a description of the respective predefined protocol and an area for an image corresponding to the respective predefined protocol;
for each of the other ones of the plurality of retinal images that conform to their respective predefined protocols, a copy of the retinal respective retinal image in its respective area of its respective section;
for the retinal image that is determined to be a non-conforming image, omitting the retinal image from its respective section and instead including a generic placeholder cell in its respective area that excludes any retinal image;
receiving a resubmitted image; and
replacing the placeholder cell with the resubmitted image responsive to determining that the resubmitted image conforms to the respective predefined protocol corresponding to the captured portion of the retina.
2. The system of claim 1 , wherein each respective predefined protocol requires its respective anatomical object to be at a center within the respective image.
3. The system of claim 1 , wherein the anatomical object is an optic disk or a fovea.
4. The system of claim 1 , wherein each retinal image is at least one of a fundus camera image or an optical coherence tomography (OCT) image.
5. The system of claim 1 , wherein each respective protocol comprises one or more retinal image parameters including image quality, field size, and/or retinal coverage.
6. The system of claim 1 , wherein the operations further comprise guiding the user in selecting or generating the user resubmitted image that corrects the non-conformity of the non-conforming image.
7. The system of claim 6 , wherein the system is configured to perform the guiding iteratively through multiple candidate replacement images.
8. The system of claim 1 , wherein determining that the retinal image is a non-conforming image comprises implementing machine learning from example protocols and determining non-conformity with the respective predefined protocol based on output from the implemented machine learning.
9. The system of claim 1 , wherein determining that a retinal image is a non-conforming image comprises utilizing rule based heuristics.
10. The system of claim 1 , wherein the one or more processors are at least partially housed in a server.
11. A non-transitory computer-readable medium comprising memory with instructions encoded thereon for qualifying medical images submitted by a user for diagnostic analysis, the instructions, when executed, causing one or more processors to perform operations, the instructions comprising instructions to:
receive a plurality of retinal images input by a user, each of the plurality of retinal images capturing a different portion of a retina, each different portion of a retina corresponding to a respective predefined protocol requiring an anatomical object to be at a particular location within the respective retinal image, the respective predefined protocol being of a plurality of predefined protocols;
determine that a retinal image of the plurality of retinal images is a non-conforming image based on the retinal image not conforming to its respective predefined protocol, while determining that other ones of the plurality of retinal images conform to their respective predefined protocols;
generate for display a graphical interface to the user, the graphical interface comprising:
for each respective predefined protocol of the plurality of predefined protocols, a respective section comprising a description of the respective predefined protocol and an area for an image corresponding to the respective predefined protocol;
for each of the other ones of the plurality of retinal images that conform to their respective predefined protocols, a copy of the retinal respective retinal image in its respective area of its respective section;
for the retinal image that is determined to be a non-conforming image, omitting the retinal image from its respective section and instead including a generic placeholder cell that excludes any retinal image;
receive a resubmitted image; and
replace the placeholder cell with the resubmitted image responsive to determining that the resubmitted image conforms to the respective predefined protocol corresponding to the captured portion of the retina.
12. The non-transitory computer-readable medium of claim 11 , wherein each respective predefined protocol requires its respective anatomical object to be at a center within the respective image.
13. The non-transitory computer-readable medium of claim 11 , wherein the anatomical object is an optic disk or a fovea.
14. The non-transitory computer-readable medium of claim 11 , wherein each retinal image is at least one of a fundus camera image or an optical coherence tomography (OCT) image.
15. The non-transitory computer-readable medium of claim 11 , wherein each respective protocol comprises one or more retinal image parameters including image quality, field size, and/or retinal coverage.
16. A method for qualifying medical images submitted by a user for diagnostic analysis, the method comprising:
receiving a plurality of retinal images input by a user, each of the plurality of retinal images capturing a different portion of a retina, each different portion of a retina corresponding to a respective predefined protocol requiring an anatomical object to be at a particular location within the respective retinal image, the respective predefined protocol being of a plurality of predefined protocols;
determining that a retinal image of the plurality of retinal images is a non-conforming image based on the retinal image not conforming to its respective predefined protocol, while determining that other ones of the plurality of retinal images conform to their respective predefined protocols;
generating for display a graphical interface to the user, the graphical interface comprising:
for each respective predefined protocol of the plurality of predefined protocols, a respective section comprising a description of the respective predefined protocol and an area for an image corresponding to the respective predefined protocol;
for each of the other ones of the plurality of retinal images that conform to their respective predefined protocols, a copy of the retinal respective retinal image in its respective area of its respective section;
for the retinal image that is determined to be a non-conforming image, omitting the retinal image from its respective section and instead including a generic placeholder cell that excludes any retinal image;
receiving a resubmitted image; and
replacing the placeholder cell with the resubmitted image responsive to determining that the resubmitted image conforms to the respective predefined protocol corresponding to the captured portion of the retina.
17. The method of claim 16 , wherein the method further comprises guiding the user in selecting or generating the user resubmitted image that corrects the non-conformity of the non-conforming image.
18. The method of claim 17 , wherein further comprising performing the guiding iteratively through multiple candidate replacement images.
19. The method of claim 16 , wherein determining that the retinal image is a non-conforming image comprises implementing machine learning from example protocols and determining non-conformity with the respective predefined protocol based on output from the implemented machine learning.
20. The method of claim 16 , wherein determining that a retinal image is a non-conforming image comprises utilizing rule based heuristics.