Systems, devices, and methods for providing feedback on and improving the accuracy of super-resolution imaging
Systems, methods, and computer-readable media for feedback on and improving the accuracy of super-resolution imaging. In some embodiments, a low resolution image of a specimen can be obtained using a low resolution objective of a microscopy inspection system. A super-resolution image of at least a portion of the specimen can be generated from the low resolution image of the specimen using a super-resolution image simulation. Subsequently, an accuracy assessment of the super-resolution image can be identified based on one or more degrees of equivalence between the super-resolution image and one or more actually scanned high resolution images of at least a portion of one or more related specimens identified using a simulated image classifier. Based on the accuracy assessment of the super-resolution image, it can be determined whether to further process the super-resolution image. The super-resolution image can be further processed if it is determined to further process the super-resolution image.
1. A method comprising:
obtaining, by a computing system, a training data set comprising a plurality of high confidence super-resolution images and a plurality of low confidence super-resolution images;
generating, by the computing system, a simulated image classifier for determining a confidence interval that measures a likelihood of a super-resolution image falls into a particular class by:
inputting the training data set into a machine learning algorithm; and
learning, via the machine learning algorithm, to classify super-resolution images of artifacts;
obtaining, by the computing system, a target super-resolution image of a specimen;
analyzing, by the computing system via the simulated image classifier, the target super-resolution image to determine a class associated with the target super-resolution image; and
determining, by the computing system via the simulated image classifier, the class of the target super-resolution image and a confidence interval of the class.
2. The method of claim 1 , wherein obtaining, by the computing system, the training data set comprising the plurality of high confidence super-resolution images and the plurality of low confidence super-resolution images comprises:
obtaining a set of high confidence super-resolution images, wherein each high confidence super-resolution image in the set of high confidence super-resolution images comprises an image grade or image score associated therewith.
3. The method of claim 1 , wherein determining, by the computing system via the simulated image classifier, the class of the target super-resolution image and the confidence interval of the class comprises:
comparing, by the computing system via the simulated image classifier, the target super-resolution image with an acceptable image tolerance.
4. The method of claim 3 , further comprising:
responsive to determining that the target super-resolution image falls below the acceptable image tolerance, indicating that artifacts in the super-resolution image is to be scanned using a higher resolution objective.
5. The method of claim 3 , further comprising:
responsive to determining that the target super-resolution image meets or exceeds the acceptable image tolerance, providing the target super-resolution image to an image assembly module.
6. The method of claim 3 , further comprising:
utilizing a feedback mechanism to dynamically adjust the acceptable image tolerance based on an accuracy of the simulated image classifier.
7. The method of claim 3 , further comprising:
dynamically adjusting, by the computing system, the acceptable image tolerance based on an importance of the specimen.
8. A non-transitory computer readable medium having one or more sequences of instructions, which, when executed by one or more processors, causes a computing system to perform operations, comprising:
obtaining, by the computing system, a training data set comprising a plurality of high confidence super-resolution images and a plurality of low confidence super-resolution images;
generating, by the computing system, a simulated image classifier for determining a confidence interval that measures a likelihood of a super-resolution image falls into a particular class by:
inputting the training data set into a machine learning algorithm; and
learning, via the machine learning algorithm, to classify super-resolution images of artifacts;
obtaining, by the computing system, a target super-resolution image of a specimen;
analyzing, by the computing system via the simulated image classifier, the target super-resolution image to determine a class associated with the target super-resolution image; and
determining, by the computing system via the simulated image classifier, the class of the target super-resolution image and a confidence interval of the class.
9. The non-transitory computer readable medium of claim 8 , wherein obtaining, by the computing system, the training data set comprising the plurality of high confidence super-resolution images and the plurality of low confidence super-resolution images comprises:
obtaining a set of high confidence super-resolution images, wherein each high confidence super-resolution image in the set of high confidence super-resolution images comprises an image grade or image score associated therewith.
10. The non-transitory computer readable medium of claim 8 , wherein determining, by the computing system via the simulated image classifier, that the class of the target super-resolution image and the confidence interval of the class comprises:
comparing, by the computing system via the simulated image classifier, the target super-resolution image with an acceptable image tolerance.
11. The non-transitory computer readable medium of claim 10 , further comprising:
responsive to determining that the target super-resolution image falls below the acceptable image tolerance, indicating that artifacts in the super-resolution image is to be scanned using a higher resolution objective.
12. The non-transitory computer readable medium of claim 10 , further comprising:
responsive to determining that the target super-resolution image meets or exceeds the acceptable image tolerance, providing the target super-resolution image to an image assembly module.
13. The non-transitory computer readable medium of claim 10 , further comprising:
utilizing a feedback mechanism to dynamically adjust the acceptable image tolerance based on an accuracy of the simulated image classifier.
14. The non-transitory computer readable medium of claim 10 , further comprising:
dynamically adjusting, by the computing system, the acceptable image tolerance based on an importance of the specimen.
15. A system comprising:
one or more processors; and
a memory having programming instructions stored thereon, which, when executed by the one or more processors, causes the system to perform one or more operations, comprising:
obtaining a training data set comprising a plurality of high confidence super-resolution images and a plurality of low confidence super-resolution images;
generating a simulated image classifier for determining a confidence interval that measures a likelihood of a super-resolution image falls into a particular class by:
inputting the training data set into a machine learning algorithm; and
learning, via the machine learning algorithm, to classify super-resolution images of artifacts;
obtaining a target super-resolution image of a specimen;
analyzing, via the simulated image classifier, the target super-resolution image to determine a class associated with the target super-resolution image; and
determining, via the simulated image classifier, the class of the target super-resolution image and a confidence interval of the class.
16. The system of claim 15 , wherein obtaining the training data set comprising the plurality of high confidence super-resolution images and the plurality of low confidence super-resolution images comprises:
obtaining a set of high confidence super-resolution images, wherein each high confidence super-resolution image in the set of high confidence super-resolution images comprises an image grade or image score associated therewith.
17. The system of claim 15 , wherein determining, via the simulated image classifier, the class of the target super-resolution image and the confidence interval of the class comprises:
comparing, via the simulated image classifier, the target super-resolution image with an acceptable image tolerance.
18. The system of claim 17 , further comprising:
responsive to determining that the target super-resolution image falls below the acceptable image tolerance, indicating that artifacts in the super-resolution image is to be scanned using a higher resolution objective.
19. The system of claim 17 , further comprising:
responsive to determining that the target super-resolution image meets or exceeds the acceptable image tolerance, providing the target super-resolution image to an image assembly module.
20. The system of claim 17 , further comprising:
utilizing a feedback mechanism to dynamically adjust the acceptable image tolerance based on an accuracy of the simulated image classifier.