Systems and methods for processing electronic images to determine testing for unstained specimens
A computer-implemented method may include receiving a collection of unstained digital histopathology slide images at a storage device and running a trained machine learning model on one or more slide images of the collection to infer a presence or an absence of a salient feature. The trained machine learning model may have been trained by processing a second collection of unstained or stained digital histopathology slide images and at least one synoptic annotation for one or more unstained or stained digital histopathology slide images of the second collection. The computer-implemented method may further include determining at least one map from output of the trained machine learning model and providing an output from the trained machine learning model to the storage device.
1. A computer-implemented method, comprising:
receiving a collection of unstained digital histopathology slide images comprising one or more blocks at a storage device;
running a trained machine learning model on one or more slide images of the collection to infer a presence or an absence of a salient feature,
wherein the trained machine learning model has been trained by processing a second collection of unstained or stained digital histopathology slide images and at least one synoptic annotation for one or more unstained or stained digital histopathology slide images of the second collection;
determining whether the one or more slide images of a first block of the collection show a sufficient amount of tumor based on the presence of the salient feature and based on one or more tests to be performed on the amount of tumor;
selecting a second block of the collection for testing and re-performing the determining for the second block if none of the one or more slide images of the first block show the sufficient amount of tumor; and
determining, on the one or more slide images of the first block of the collection, where a salient region is located and indicating, on the one or more slide images, the salient region that is optimal for testing if the one or more slide images show the sufficient amount of tumor.
2. The computer-implemented method of claim 1 , wherein the one or more tests are associated with providing a continuous recurrence score for invasive breast cancer.
3. The computer-implemented method of claim 1 , wherein the at least one synoptic annotation comprises one or more labels of tests for which amounts of tumor shown in the second collection of unstained or stained digital histopathology slide images are sufficient.
4. The computer-implemented method of claim 1 , wherein the one or more tests are associated with providing a continuous score for recurrence of non-invasive breast cancer.
5. The computer-implemented method of claim 1 , further comprising:
outputting, based on determining that the one or more slide images of the collection do not show a sufficient amount of tumor, information indicating an amount by which the amount of tumor is insufficient.
6. The computer-implemented method of claim 1 , wherein the one or more tests are associated with providing a continuous score for a prostate cancer treatment recommendation.
7. The computer-implemented method of claim 1 , further comprising:
generating a recommendation for a specific test, of the one or more tests, based on determining that the one or more slide images of the collection show a sufficient amount of tumor.
8. The computer-implemented method of claim 1 , wherein the one or more tests are associated with providing a continuous score for a likelihood of malignancy of the tumor.
9. The computer-implemented method of claim 1 , further comprising:
running an artificial intelligence (AI) system to virtually stain the one or more slide images of the collection of unstained digital histopathology slide images;
partitioning the one or more slide images of the collection into a collection of tiles;
detecting or segmenting at least one tissue region of the one or more slide images of the collection from a non-tissue background of the one or more slide images of the collection to create a tissue mask;
removing one or more tiles of the collection of tiles detected to be the non-tissue background; and
wherein the running of the trained machine learning model further comprises:
running the trained machine learning model after removing the one or more tiles.
10. A system for using a trained machine learning model for tissue analysis includes memory storing instructions, and at least one processor executing the instructions to perform a process including:
receiving a collection of unstained digital histopathology slide images comprising one or more blocks at a storage device;
running a trained machine learning model on one or more slide images of the collection to infer a presence or an absence of a salient feature,
wherein the trained machine learning model has been trained by processing a second collection of unstained or stained digital histopathology slide images and at least one synoptic annotation for one or more unstained or stained digital histopathology slide images of the second collection;
determining whether the one or more slide images of a first block of the collection show a sufficient amount of tumor based on the presence of the salient feature and based on one or more tests to be performed on the amount of tumor;
selecting a second block of the collection for testing and re-performing the determining for the second block if none of the one or more slide images of the first block show the sufficient amount of tumor; and
determining, on the one or more slide images of the first block of the collection, where a salient region is located and indicating, on the one or more slide images, the salient region that is optimal for testing if the one or more slide images show the sufficient amount of tumor.
11. The system of claim 10 , wherein the one or more tests are associated with providing a continuous recurrence score for invasive breast cancer.
12. The system of claim 10 , wherein the at least one synoptic annotation comprises one or more labels of tests for which amounts of tumor shown in the second collection of unstained or stained digital histopathology slide images are sufficient.
13. The system of claim 10 , wherein the one or more tests are associated with providing a continuous score for recurrence of non-invasive breast cancer.
14. The system of claim 10 , further comprising:
outputting, based on determining that the one or more slide images of the collection do not show a sufficient amount of tumor, information indicating an amount by which the amount of tumor is insufficient.
15. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform a method for using a trained machine learning model for tissue analysis, the method including:
receiving a collection of unstained digital histopathology slide images comprising one or more blocks at a storage device;
running a trained machine learning model on one or more slide images of the collection to infer a presence or an absence of a salient feature,
wherein the trained machine learning model has been trained by processing a second collection of unstained or stained digital histopathology slide images and at least one synoptic annotation for one or more unstained or stained digital histopathology slide images of the second collection;
determining whether the one or more digital histopathology slide images of a first block of the collection show a sufficient amount of tumor based on the presence of the salient feature and based on one or more tests to be performed on the amount of tumor;
selecting a second block of the collection for testing and re-performing the determining for the second block if none of the one or more slide images of the first block show the sufficient amount of tumor; and
determining, on the one or more slide images of the first block of the collection of unstained digital histopathology slide images, where a salient region is located and indicating, on the one or more slide images, the salient region that is optimal for testing if the one or more slide images show the sufficient amount of tumor.
16. The non-transitory computer-readable medium of claim 15 , wherein the at least one synoptic annotation comprises one or more labels of tests for which amounts of tumor shown in the second collection of unstained or stained digital histopathology slide images are sufficient.
17. The non-transitory computer-readable medium of claim 15 , further comprising:
outputting, based on determining that the one or more slide images of the collection do not show a sufficient amount of tumor, information indicating an amount by which the amount of tumor is insufficient.
18. The non-transitory computer-readable medium of claim 15 , further comprising:
generating a recommendation for a specific test, of the one or more tests, based on determining that the one or more slide images of the collection show a sufficient amount of tumor.
19. The non-transitory computer-readable medium of claim 15 , further comprising:
running an artificial intelligence (AI) system to virtually stain the one or more slide images of the collection of unstained digital histopathology slide images;
partitioning the one or more slide images of the collection into a collection of tiles;
detecting or segmenting at least one tissue region of the one or more slide images of the collection from a non-tissue background of the one or more slide images of the collection to create a tissue mask;
removing one or more tiles of the collection of tiles detected to be the non-tissue background; and
wherein the running of the trained machine learning model further comprises:
running the trained machine learning model after removing the one or more tiles.
20. The non-transitory computer-readable medium of claim 15 , wherein the one or more tests are associated with providing at least one of:
a continuous recurrence score for invasive breast cancer,
a continuous score for recurrence of non-invasive breast cancer,
a continuous score for a prostate cancer treatment recommendation, or
a continuous score for a likelihood of malignancy of the tumor.