Systems and methods for processing electronic images for auto-labeling for computational pathology
Systems and methods are described herein for processing electronic medical images to determine a first machine learning system, the first machine learning system having been trained to identify regions of electronic medical images; receive a plurality of electronic medical images, each of the electronic medical images being associated with one or more subcategories; determine a subset of the plurality of electronic medical images that are associated with only one subcategory of the one or more subcategories; provide the subset of the plurality of electronic medical images to the first machine learning system, the first machine learning system identifying regions within the subset of the plurality of electronic medical images associated with the subcategory; and train a second machine learning system, using the identified regions and the subset of the plurality of electronic medical images.
1 . A computer-implemented method for processing electronic medical images, comprising:
receiving a plurality of electronic medical images, each of the electronic medical images being associated with one or more subcategories of health grades;
determining a first machine learning system of a plurality of machine learning training systems, the first machine learning system having been trained to identify one subcategory of the one or more subcategories;
providing the plurality of electronic medical images to the first machine learning system to determine a subset of the plurality of electronic medical images associated with the one subcategory that the first machine learning system has been trained to identify;
training a second machine learning system, using the subset of the plurality of electronic medical images; and
repeating the determining, providing and training steps with remaining machine learning systems of the plurality of machine learning training systems, each of the remaining machine learning systems being trained identify a different subcategory of the one or more subcategories.
2 . The method of claim 1 , wherein the subcategories are Gleason pattern scores.
3 . The method of claim 2 , wherein the second machine learning system may classify tissue within new electronic images into one of the Gleason pattern scores.
4 . The method of claim 3 , wherein the second machine learning system outputs a slide level Gleason score based on proportions of each Gleason score assigned to the tissue within the electronic images.
5 . The method of claim 1 , wherein the first machine learning system creates a segmentation mask for each subset of the plurality of electronic medical images associated with each subcategory.
6 . The method of claim 1 , wherein the second machine learning system is capable of identifying one or more subcategories on additional electronic medical images.
7 . The method of claim 1 , wherein the plurality of electronic medical images are images that have had a domain shift applied.
8 . A system for processing electronic medical images, the system comprising:
at least one memory storing instructions; and
at least one processor configured to execute the instructions to perform operations comprising:
receiving a plurality of electronic medical images, each of the electronic medical images being associated with one or more subcategories of health grades;
determining a first machine learning system of a plurality of machine learning training systems, the first machine learning system having been trained to identify one subcategory of the one or more subcategories;
providing the plurality of electronic medical images to the first machine learning system to determine, a subset of the plurality of electronic medical images associated with the one subcategory that the first machine learning system has been trained to identify;
training a second machine learning system, using the subset of the plurality of electronic medical images; and
repeating the determining, providing and training steps with remaining machine learning systems of the plurality of machine learning training systems, each of the remaining machine learning systems being trained identify a different subcategory of the one or more subcategories.
9 . The system of claim 8 , wherein the subcategories are Gleason pattern scores.
10 . The system of claim 9 , wherein the second machine learning system may classify tissue within new electronic images into one of the Gleason pattern scores.
11 . The system of claim 10 , wherein the second machine learning system outputs a slide level Gleason score based on proportions of each Gleason score assigned to the tissue within the electronic images.
12 . The system of claim 8 , wherein the first machine learning system creates a segmentation mask utilizing the identified regions.
13 . The system of claim 8 , wherein the second machine learning system is capable of identifying one or more subcategories on additional electronic medical images.
14 . The system of claim 8 , wherein the plurality of electronic medical images are images that have had a domain shift applied.
15 . A non-transitory computer-readable medium storing instructions that, when executed by a processor, perform operations processing electronic digital medical images, the operations comprising:
receiving a plurality of electronic medical images, each of the electronic medical images being associated with one or more subcategories of health grades;
determining a first machine learning system of a plurality of machine learning training systems, the first machine learning system having been trained to identify one subcategory of the one or more subcategories;
providing the plurality of electronic medical images to the first machine learning system to determine, a subset of the plurality of electronic medical images associated with the one subcategory that the first machine learning system has been trained to identify;
training a second machine learning system, using the subset of the plurality of electronic medical images; and
repeating the determining, providing and training steps with remaining machine learning systems of the plurality of machine learning training systems, each of the remaining machine learning systems being trained identify a different subcategory of the one or more subcategories.
16 . The computer-readable medium of claim 15 , wherein the subcategories are Gleason pattern scores.
17 . The computer-readable medium of claim 16 , wherein the second machine learning system may classify tissue within new electronic images into one of the Gleason pattern scores.
18 . The computer-readable medium of claim 17 , wherein the second machine learning system outputs a slide level Gleason score based on proportions of each Gleason score assigned to the tissue within the electronic images.
19 . The computer-readable medium of claim 15 , wherein the first machine learning system creates a segmentation mask utilizing the identified regions.
20 . The computer-readable medium of claim 15 , wherein the second machine learning system is capable of identifying one or more subcategories on additional electronic medical images.