Artificial intelligence-based assistant for concurrent review of needle core prostate biopsies
One example method includes receiving a digital image of a needle core prostate biopsy, displaying, using a display device, a magnified portion of the digital image, obtaining, from a deep learning model, Gleason scores corresponding to patches of the magnified portion of the digital image, and displaying, using the display device, a superimposed overlay on the magnified portion of the digital image based on the Gleason scores and corresponding confidence values of the Gleason scores, the superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the magnified portion of the digital image and comprising image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the magnified portion of the digital image and a confidence value of the corresponding Gleason score.
1 . A system comprising:
a non-transitory computer-readable medium; and
one or more processors communicatively coupled to the communications interface and the non-transitory computer-readable medium, the one or more processors configured to execute processor-executable instructions stored in the non-transitory computer-readable medium to:
receive a selection of a digital image of a needle core prostate biopsy having corresponding Gleason score information generated by a deep learning model, the Gleason score information comprising one or more Gleason scores and one or more confidence values corresponding to image patches in the digital image;
cause a display device to display a magnified portion of the digital image;
obtain a subset of the Gleason score information corresponding to image patches in the magnified portion of the digital image; and
cause the display device to display a superimposed overlay on the magnified portion of the digital image based on the Gleason score and corresponding confidence values of the Gleason score, the superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the magnified portion of the digital image and comprising image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the magnified portion of the digital image and an overlay providing confidence values for the image patches and the corresponding Gleason scores.
2 . The system of claim 1 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
receive a command from a user interface to change a view of the digital image, the changed view comprising a different magnified portion of the digital image; and
cause the display device to display an updated superimposed overlay on the different magnified portion of the digital image based on the Gleason scores and corresponding confidence values of the Gleason scores, the updated superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the different magnified portion of the digital image and comprising updated image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the different magnified portion of the digital image and a confidence value of the corresponding Gleason score.
3 . The system of claim 2 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to update the superimposed overlay in real-time as the view of the digital image changes.
4 . The system of claim 1 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
receive a plurality of digital images of the needle core prostate biopsy; and
cause the display device to display an interactive user interface element for navigating between the plurality of digital images.
5 . The system of claim 1 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
obtain, from the deep learning model, a prediction of a final Gleason Grade for the biopsy; and
cause the display device to display the prediction of the final Gleason Grade for the biopsy.
6 . The system of claim 1 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
obtain a relative proportion by area of the tumor of different Gleason scores and the amount of tumor as a percent of tissue for the biopsy; and
cause the display device to display the relative proportion by area of each Gleason score and the amount of tumor as a percent of tissue for the digital image.
7 . The system of claim 1 , further comprising the deep learning model, the deep learning model trained to make the predictions of Gleason scores of portions of the digital image, a relative proportion by area of the tumor of different Gleason scores, an amount of tumor as a percent of tissue for the biopsy, and a prediction of a final Gleason Grade for the biopsy, and wherein the deep learning model comprises a combination of a deep convolutional neural network trained to make the predictions of Gleason scores of portions of the digital image and a support vector machine to generate data for the prediction of a final Gleason Grade for the biopsy.
8 . The system of claim 1 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to:
cause the display device to display a viewing pane showing a portion of the magnified digital image,
cause the display device to display a thumbnail region showing the biopsy including a box indicating where in the biopsy the viewing pane is showing the magnified portion of the digital image; and
cause the display device to display a sidebar arranged peripheral to the viewing pane and comprising icons for activating and controlling which portion of the digital image is displayed in the viewing pane and an opacity of the colors overlaid on the image patches.
9 . The system of claim 8 , wherein the biopsy comprises multiple slices of prostate tissue, each having its associated digital image, and wherein the sidebar further includes thumbnail images of the digital images of the multiple slices.
10 . The system of claim 8 , wherein the one or more processors are configured to execute further processor-executable instructions stored in the non-transitory computer-readable medium to display user interface elements for panning around the digital image, making measurements within the digital image, and marking virtual pins in the digital image.
11 . A method comprising:
receiving a selection of a digital image of a needle core prostate biopsy having corresponding Gleason score information generated by a deep learning model, the Gleason score information comprising one or more Gleason scores and one or more confidence values corresponding to image patches in the digital image;
displaying, using a display device, a magnified portion of the digital image;
obtaining a subset of the Gleason score information corresponding to image patches in the magnified portion of the digital image; and
displaying, using the display device, a superimposed overlay on the magnified portion of the digital image based on the Gleason score and corresponding confidence values of the Gleason scores, the superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the magnified portion of the digital image and comprising image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the magnified portion of the digital image and an overlay providing confidence values for the image patches and the corresponding Gleason scores.
12 . The method of claim 11 , further comprising:
receiving a command from a user interface to change a view of the digital image, the changed view comprising a different magnified portion of the digital image; and
displaying an updated superimposed overlay on the different magnified portion of the digital image based on the Gleason scores and corresponding confidence values of the Gleason scores, the updated superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the different magnified portion of the digital image and comprising updated image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the different magnified portion of the digital image and a confidence value of the corresponding Gleason score.
13 . The method of claim 11 , further comprising updating the superimposed overlay in real-time as the view of the digital image changes.
14 . The method of claim 11 , further comprising:
receiving a plurality of digital images of the needle core prostate biopsy; and
causing the display device to display an interactive user interface element for navigating between the plurality of digital images.
15 . The method of claim 11 , further comprising:
obtaining, from the deep learning model, a prediction of a final Gleason Grade for the biopsy; and
displaying the prediction of the final Gleason Grade for the biopsy.
16 . The method of claim 11 , further comprising:
obtaining a relative proportion by area of the tumor of different Gleason scores and the amount of tumor as a percent of tissue for the biopsy; and
displaying the relative proportion by area of each Gleason score and the amount of tumor as a percent of tissue for the digital image.
17 . The method of claim 11 , wherein the deep learning model is trained to make the predictions of Gleason scores of portions of the digital image, a relative proportion by area of the tumor of different Gleason scores, an amount of tumor as a percent of tissue for the biopsy, and a prediction of a final Gleason Grade for the biopsy, and wherein the deep learning model comprises a combination of a deep convolutional neural network trained to make the predictions of Gleason scores of portions of the digital image and a support vector machine to generate data for the prediction of a final Gleason Grade for the biopsy.
18 . The method of claim 11 , further comprising:
displaying a viewing pane showing a portion of the magnified digital image,
displaying a thumbnail region showing the biopsy including a box indicating where in the biopsy the viewing pane is showing the magnified portion of the digital image; and
displaying a sidebar arranged peripheral to the viewing pane and comprising icons for activating and controlling which portion of the digital image is displayed in the viewing pane and an opacity of the colors overlaid on the image patches.
19 . The method of claim 18 , wherein the biopsy comprises multiple slices of prostate tissue, each having its associated digital image, and wherein the sidebar further includes thumbnail images of the digital images of the multiple slices.
20 . The method of claim 18 , further comprising displaying user interface elements for panning around the digital image, making measurements within the digital image, and marking virtual pins in the digital image.
21 . A non-transitory computer-readable medium comprising processor-executable instructions configured to cause one or more processors to:
receive a selection of a digital image of a needle core prostate biopsy having corresponding Gleason score information generated by a deep learning model, the Gleason score information comprising one or more Gleason scores and one or more confidence values corresponding to image patches in the digital image;
cause a display device to display a magnified portion of the digital image;
obtain a subset of the Gleason score information corresponding to image patches in the magnified portion of the digital image; and
cause the display device to display a superimposed overlay on the magnified portion of the digital image based on the Gleason scores and corresponding confidence values of the Gleason score, the superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the magnified portion of the digital image and comprising image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the magnified portion of the digital image and an overlay providing confidence values for the image patches and the corresponding Gleason scores.
22 . The non-transitory computer-readable medium of claim 21 , further comprising processor-executable instructions configured to cause the one or more processors to:
receive a command from a user interface to change a view of the digital image, the changed view comprising a different magnified portion of the digital image; and
cause the display device to display an updated superimposed overlay on the different magnified portion of the digital image based on the Gleason scores and corresponding confidence values of the Gleason scores, the updated superimposed overlay comprising one or more outlines corresponding one or more Gleason scores associated with the different magnified portion of the digital image and comprising updated image patches having colors based on a Gleason score of the Gleason scores corresponding to an underlying portion of the different magnified portion of the digital image and a confidence value of the corresponding Gleason score.
23 . The non-transitory computer-readable medium of claim 22 , further comprising processor-executable instructions configured to cause the one or more processors to update the superimposed overlay in real-time as the view of the digital image changes.
24 . The non-transitory computer-readable medium of claim 21 , further comprising processor-executable instructions configured to cause the one or more processors to:
receive a plurality of digital images of the needle core prostate biopsy; and
cause the display device to display an interactive user interface element for navigating between the plurality of digital images.
25 . The non-transitory computer-readable medium of claim 21 , further comprising processor-executable instructions configured to cause the one or more processors to:
obtain, from the deep learning model, a prediction of a final Gleason Grade for the biopsy; and
cause the display device to display the prediction of the final Gleason Grade for the biopsy.
26 . The non-transitory computer-readable medium of claim 21 , further comprising processor-executable instructions configured to cause the one or more processors to:
obtain a relative proportion by area of the tumor of different Gleason scores and the amount of tumor as a percent of tissue for the biopsy; and
cause the display device to display the relative proportion by area of each Gleason score and the amount of tumor as a percent of tissue for the digital image.
27 . The non-transitory computer-readable medium of claim 21 , further comprising the deep learning model, the deep learning model trained to make the predictions of Gleason scores of portions of the digital image, a relative proportion by area of the tumor of different Gleason scores, an amount of tumor as a percent of tissue for the biopsy, and a prediction of a final Gleason Grade for the biopsy, and wherein the deep learning model comprises a combination of a deep convolutional neural network trained to make the predictions of Gleason scores of portions of the digital image and a support vector machine to generate data for the prediction of a final Gleason Grade for the biopsy.
28 . The non-transitory computer-readable medium of claim 21 , further comprising processor-executable instructions configured to cause the one or more processors to:
cause the display device to display a viewing pane showing a portion of the magnified digital image,
cause the display device to display a thumbnail region showing the biopsy including a box indicating where in the biopsy the viewing pane is showing the magnified portion of the digital image; and
cause the display device to display a sidebar arranged peripheral to the viewing pane and comprising icons for activating and controlling which portion of the digital image is displayed in the viewing pane and an opacity of the colors overlaid on the image patches.
29 . The non-transitory computer-readable medium of claim 28 , wherein the biopsy comprises multiple slices of prostate tissue, each having its associated digital image, and wherein the sidebar further includes thumbnail images of the digital images of the multiple slices.
30 . The non-transitory computer-readable medium of claim 28 , further comprising processor-executable instructions configured to cause the one or more processors to cause the display device to display user interface elements for panning around the digital image, making measurements within the digital image, and marking virtual pins in the digital image.