Systems and methods for processing electronic images
An image processing method including identifying, using a machine learning system, an area of interest of a target image by analyzing features extracted from image regions in the target image, the machine learning system being generated by processing a plurality of training images each comprising an image of human tissue and a diagnostic label characterizing at least one of a slide morphology, a diagnostic value, and a pathologist review outcome; determining, using the machine learning system, a probability of a target feature being present in the area of interest of the target image based on an average probability; determining, using the machine learning system, a prioritization value, of a plurality of prioritization values, of the target image based on the probability of the target feature being present in the target image.
1. An image processing method, comprising:
identifying, using a machine learning system, an area of interest of a target image by analyzing features extracted from image regions in the target image, the machine learning system being generated by processing a plurality of training images each comprising an image of human tissue and a diagnostic label characterizing at least one of a slide morphology, a diagnostic value, and a pathologist review outcome;
determining, using the machine learning system, a probability of a target feature being present in the area of interest of the target image based on an average probability;
determining, using the machine learning system, a prioritization value, of a plurality of prioritization values, of the target image based on the probability of the target feature being present in the target image, the prioritization value comprising a first prioritization value determined based on preferences of a first user and a second prioritization value determined based on preferences of a second user; and
outputting, using the machine learning system, a sequence of digitized pathology images, wherein a placement of the target image in the sequence is based on the prioritization values.
2. The method of claim 1 , wherein the diagnostic label comprises a preparation value corresponding to a likelihood that further preparation is to be performed by preparing a new slide for the target image.
3. The method of claim 1 , wherein the diagnostic label comprises a preparation value corresponding to a likelihood that further preparation is to be performed by preparing a new slide for the target image.
4. The method of claim 3 , wherein further preparation is performed by preparing a new slide for the target image based on at least one of a specimen recut, an immunohistochemical stain, additional diagnostic testing, additional consultation, and/or a special stain, being performed prior to a user review.
5. The method of claim 1 , further comprising: upon determining that the target feature comprises a feature in the target image indicating that further preparation is to be performed, preparing a new slide for the target image prior to a user review.
6. The method of claim 5 , the diagnostic label comprising at least one of cancer presence, cancer grade, treatment effects, precancerous lesions, biomarkers for treatment selection, and/or presence of infectious organisms.
7. The method of claim 1 , wherein the diagnostic label comprises an artifact label corresponding to at least one of scanning lines, missing tissue, and/or blur.
8. An image processing system, comprising:
a memory storing instructions; and
a processor configured to execute the instructions to perform operations comprising:
identifying, using a machine learning system, an area of interest of a target image by analyzing features extracted from image regions in the target image, the machine learning system being generated by processing a plurality of training images each comprising an image of human tissue and a diagnostic label characterizing at least one of a slide morphology, a diagnostic value, and an analytic difficulty;
determining, using the machine learning system, a probability of a target feature being present in the area of interest of the target image based on an average probability;
determining, using the machine learning system, a prioritization value, of a plurality of prioritization values, of the target image based on the probability of the target feature being present in the target image, the prioritization value comprising a first prioritization value determined based on preferences of a first user and a second prioritization value determined based on preferences of a second user; and
outputting, using the machine learning system, a sequence of digitized pathology images, wherein a placement of the target image in the sequence is based on the prioritization values.
9. The system of claim 8 , wherein the diagnostic label comprises a preparation value corresponding to a likelihood that further preparation is to be performed by preparing a new slide for the target image.
10. The system of claim 8 , wherein the diagnostic label comprises a preparation value corresponding to a likelihood that further preparation is to be performed by preparing a new slide for the target image.
11. The system of claim 8 , wherein further preparation is performed for the target image based on at least one of a specimen recut, an immunohistochemical stain, additional diagnostic testing, additional consultation, and/or a special stain, being performed prior to a user review.
12. The system of claim 8 , wherein the diagnostic label comprises a diagnostic feature of the target image.
13. The system of claim 12 , the diagnostic feature comprising at least one of cancer presence, cancer grade, treatment effects, precancerous lesions, biomarkers for treatment selection, and presence of infectious organisms.
14. The system of claim 8 , wherein the diagnostic label comprises an artifact label corresponding to at least one of scanning lines, missing tissue, and blur.
15. A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform image processing operations, the operations comprising:
identifying, using a machine learning system, an area of interest of a target image by analyzing features extracted from image regions in the target image, the machine learning system being generated by processing a plurality of training images each comprising an image of human tissue and a diagnostic label characterizing at least one of a diagnostic value, a pathologist review outcome, and an analytic difficulty;
determining, using the machine learning system, a probability of a target feature being present in the area of interest of the target image based on an average probability;
determining, using the machine learning system, a prioritization value, of a plurality of prioritization values, of the target image based on the probability of the target feature being present in the target image, the prioritization value comprising a first prioritization value determined based on a first user and a second prioritization value determined based on a second user; and
outputting, using the machine learning system, a sequence of digitized pathology images, wherein a placement of the target image in the sequence is based on the prioritization values.
16. The non-transitory computer-readable medium of claim 15 , wherein the diagnostic label comprises a preparation value corresponding to a likelihood that further preparation is to be performed by preparing a new slide for the target image.
17. The non-transitory computer-readable medium of claim 15 , wherein the diagnostic label comprises a preparation value corresponding to a likelihood that further preparation is to be performed by preparing a new slide for the target image.
18. The non-transitory computer-readable medium of claim 17 , wherein further preparation is performed by preparing a new slide for the target image based on at least one of a specimen recut, an immunohistochemical stain, additional diagnostic testing, additional consultation, and/or a special stain, being performed prior to a user review.
19. The non-transitory computer-readable medium of claim 15 , wherein the diagnostic label comprises a diagnostic feature of the target image, the diagnostic feature comprising at least one of cancer presence, cancer grade, treatment effects, precancerous lesions, biomarkers for treatment selection, and/or presence of infectious organisms.