Systems and methods for processing whole slide images using machine-learning
According to systems and techniques disclosed herein, a method for generating a navigable three-dimensional image of a tissue sample may include receiving a plurality of whole slide images (WSI) associated with the tissue sample. The method may further include providing the plurality of whole slide images to a machine-learning model. The machine-learning model may have been trained to identify one or more positional features within the plurality of whole slide images and output a plurality of relative positional relationships corresponding to each of the plurality of whole slide images. The method may further include generating the navigable three-dimensional image of the tissue sample based on the plurality of relative positional relationships. The method may further include generating an interactive display incorporating the navigable three-dimensional image. The method may further include providing, to a user interface, the interactive display.
1 . A computer-implemented method for generating a navigable three-dimensional image of a tissue sample, the method comprising:
receiving, by one or more processors, a plurality of whole slide images (WSI) associated with the tissue sample;
providing, by the one or more processors, the plurality of whole slide images to a machine-learning model, wherein the machine-learning model has been trained, using one or more prior patient and/or synthetically generated sets of whole slide images, to identify one or more positional features within the plurality of whole slide images and output a plurality of relative positional relationships corresponding to each of the plurality of whole slide images;
generating, by the one or more processors, the navigable three-dimensional image of the tissue sample based on the plurality of relative positional relationships;
generating, by the one or more processors, an interactive display incorporating the navigable three-dimensional image; and
providing, to a user interface and by the one or more processors, the interactive display.
2 . The computer-implemented method of claim 1 , further comprising co-registering two or more whole slide images of the plurality of whole slide images based on the plurality of relative positional relationships.
3 . The computer-implemented method of claim 2 , further comprising:
generating, by the one or more processors, a side-by-side display incorporating graphical representations of the two or more whole slide images based on the co-registering; and
providing, to the user interface and by the one or more processors, the side-by-side display.
4 . The computer-implemented method of claim 3 , wherein displayed pixels of the side-by-side display are based on an average of color information.
5 . The computer-implemented method of claim 1 , further comprising:
providing, by the one or more processors, the plurality of whole slide images to the machine-learning model, wherein the machine-learning model has been trained, using one or more prior patient and/or simulated sets of whole slide images, to identify one or more positional features within the plurality of whole slide images and output a sample level corresponding to each of the plurality of whole slide images.
6 . The computer-implemented method of claim 5 , further comprising:
ordering, by the one or more processors, each whole slide image relative to the plurality of whole slide images based on the sample level corresponding to each whole slide image.
7 . The computer-implemented method of claim 6 , further comprising:
generating, by the one or more processors, the navigable three-dimensional image using an image stitching of the plurality of whole slide images, the image stitching based on the ordering.
8 . The computer-implemented method of claim 7 , further comprising:
generating, by the one or more processors, an interactive display incorporating the navigable three-dimensional image, the interactive display operable to navigate sample levels.
9 . A system for generating a navigable three-dimensional image of a tissue sample, the system comprising:
a memory storing instructions and a processor operatively connected to the memory and configured to execute the instructions to perform operations comprising:
receiving, by one or more processors, a plurality of whole slide images (WSI) associated with the tissue sample;
providing, by the one or more processors, the plurality of whole slide images to a machine-learning model, wherein the machine-learning model has been trained, using one or more prior patient and/or simulated sets of whole slide images, to identify one or more positional features within the plurality of whole slide images and output a plurality of relative positional relationships corresponding to each of the plurality of whole slide images;
generating, by the one or more processors, the navigable three-dimensional image of the tissue sample based on the plurality of relative positional relationships;
generating, by the one or more processors, an interactive display incorporating the navigable three-dimensional image; and
providing, to a user interface and by the one or more processors, the interactive display.
10 . The system of claim 9 , the operations further comprising co-registering two or more whole slide images of the plurality of whole slide images based on the plurality of relative positional relationships.
11 . The system of claim 10 , the operations further comprising:
generating, by the one or more processors, a side-by-side display incorporating graphical representations of the two or more whole slide images based on the co-registering; and
providing, to the user interface and by the one or more processors, the side-by-side display.
12 . The system of claim 9 , the operations further comprising:
providing, by the one or more processors, the plurality of whole slide images to the machine-learning model, wherein the machine-learning model has been trained, using one or more prior patient and/or simulated sets of whole slide images, to identify one or more positional features within the plurality of whole slide images and output a sample level corresponding to each of the plurality of whole slide images.
13 . The system of claim 12 , the operations further comprising:
ordering, by the one or more processors, each whole slide image relative to the plurality of whole slide images based on the sample level corresponding to each whole slide image.
14 . The system of claim 13 , the operations further comprising:
generating, by the one or more processors, the navigable three-dimensional image using an image stitching of the plurality of whole slide images, the image stitching based on the ordering.
15 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, perform operations including:
receiving, by one or more processors, a plurality of whole slide images (WSI) associated with a tissue sample;
providing, by the one or more processors, the plurality of whole slide images to a machine-learning model, wherein the machine-learning model has been trained, using one or more prior patient and/or simulated sets of whole slide images, to identify one or more positional features within the plurality of whole slide images and output a plurality of relative positional relationships corresponding to each of the plurality of whole slide images;
generating, by the one or more processors, a navigable three-dimensional image of the tissue sample based on the plurality of relative positional relationships;
generating, by the one or more processors, an interactive display incorporating the navigable three-dimensional image; and
providing, to a user interface and by the one or more processors, the interactive display.
16 . The non-transitory computer-readable medium of claim 15 , the operations further comprising co-registering two or more whole slide images of the plurality of whole slide images based on the plurality of relative positional relationships.
17 . The non-transitory computer-readable medium of claim 16 , the operations further comprising:
generating, by the one or more processors, a side-by-side display incorporating graphical representations of the two or more whole slide images based on the co-registering; and
providing, to the user interface and by the one or more processors, the side-by-side display.
18 . The non-transitory computer-readable medium of claim 15 , the operations further comprising:
providing, by the one or more processors, the plurality of whole slide images to the machine-learning model, wherein the machine-learning model has been trained, using one or more prior patient and/or simulated sets of whole slide images, to identify one or more positional features within the plurality of whole slide images and output a sample level corresponding to each of the plurality of whole slide images.
19 . The non-transitory computer-readable medium of claim 18 , the operations further comprising:
ordering, by the one or more processors, each whole slide image relative to the plurality of whole slide images based on the sample level corresponding to each whole slide image.
20 . The non-transitory computer-readable medium of claim 19 , the operations further comprising:
generating, by the one or more processors, the navigable three-dimensional image using an image stitching of the plurality of whole slide images, the image stitching based on the ordering.