Classifying images using deep neural network with integrated acquisition information

1. A system, comprising a processor to:

receive an image with corresponding acquisition information; and

classify the image using the corresponding acquisition information as input into a deep neural network that comprises a learned mask, wherein the learned mask is learned by operating a fully connected layer in the deep neural network and the same mask is then applied to a plurality of layers of the deep neural network to integrate acquisition information into the deep neural network.

2. The system of claim 1 , wherein the integrated acquisition information comprises the learned mask that is learned by operating a plurality of fully connected layers and applied to different convolutional layers in different depths of the deep neural network.

3. The system of claim 1 , wherein the integrated acquisition information comprises the learned mask that is learned individually for each layer of the deep neural network.

4. The system of claim 1 , wherein the learned mask that integrates the acquisition information is concatenated to a feature before a decision layer of the deep neural network.

5. The system of claim 1 , wherein the classification of the image comprises a cancer detection, a breast density, a risk, or any combination thereof.

6. The system of claim 1 , wherein the acquisition information comprises information related to an acquisition process of the image.

7. The system of claim 1 , wherein the acquisition information and image is received as Digital Imaging and Communications in Medicine (DICOM®) data.

8. A computer-implemented method, comprising:

receiving, at a trained deep neural network comprising integrated acquisition information, an image with corresponding acquisition information; and

classifying, via the trained deep neural network, the image using the corresponding acquisition information as input into a deep neural network that comprises a learned mask, wherein the learned mask is learned by operating a fully connected layer in the deep neural network and the same mask is then applied to a plurality of layers of the deep neural network to integrate acquisition information into the deep neural network.

9. The computer-implemented method of claim 8 , comprising learning the learned mask by operating a plurality of fully connected layers and applying the learned mask to different convolutional layers at different depths of the deep neural network to integrate the acquisition information into the trained deep neural network.

10. The computer-implemented method of claim 8 , comprising training the deep neural network by concatenating the learned mask to a feature before a decision layer of the deep neural network.

11. The computer-implemented method of claim 8 , comprising learning the learned mask individually for each layer of the deep neural network to integrate the acquisition information into the trained deep neural network.

12. The computer-implemented method of claim 8 , wherein classifying the image comprises generating comprises classifying the images as a cancer.

13. The computer-implemented method of claim 8 , wherein classifying the image comprises generating comprises generating a breast density.

14. The computer-implemented method of claim 8 , wherein classifying the image comprises generating comprises generating a risk.

15. A computer program product for classifying images, the computer program product comprising a computer-readable storage medium having program code embodied therewith, the program code executable by a processor to cause the processor to:

receive an image with corresponding acquisition information at a trained deep neural network comprising integrated acquisition information; and

classify the image using the corresponding acquisition information as input into a deep neural network that comprises a learned mask, wherein the learned mask is learned by operating a fully connected layer in the deep neural network and the same mask is then applied to a plurality of layers of the deep neural network to integrate acquisition information into the deep neural network.

16. The computer program product of claim 15 , further comprising program code executable by the processor to operate a plurality of fully connected layers to learn the learned mask that integrates the acquisition information and apply the learned mask to different convolutional layers at different depths of the deep neural network.

17. The computer program product of claim 15 , further comprising program code executable by the processor to concatenate the integrated acquisition information to a feature before a decision layer of the deep neural network.

18. The computer program product of claim 15 , further comprising program code executable by the processor to learn the learned mask individually for each layer of the deep neural network to integrate the acquisition information.

19. The computer program product of claim 15 , further comprising program code executable by the processor to classify the image as cancerous.

20. The computer program product of claim 15 , further comprising program code executable by the processor to generate a breast density.