Systems and methods for improving low dose volumetric contrast-enhanced mri

Abstract

Methods and systems are provided for improving model robustness and generalizability. The method may comprise: acquiring, using a medical imaging apparatus, a medical image of a subject; reformatting the medical image of the subject in multiple scanning orientations; applying a deep network model to the medical image to improve the quality of the medical image; and outputting an improved quality image of the subject for analysis by a physician.

1 . (canceled) 
     
     
         2 . A computer-implemented method for improving image quality of medical image, the method comprising:
 receiving a volumetric medical image of a subject;   generating one or more reformat volumetric medical images by reformatting the volumetric medical image of the subject in one or more orientations; and   feeding an input comprising the one or more reformat volumetric medical images to a deep network model to generate a predicted medical image with improved quality.   
     
     
         3 . The computer-implemented method of  claim 2 , wherein the volumetric medical image is acquired in a first direction of a scanning plane. 
     
     
         4 . The computer-implemented method of  claim 3 , wherein the one or more orientations comprise a second direction that is different from the first direction of the scanning plane. 
     
     
         5 . The computer-implemented method of  claim 2 , wherein the volumetric medical image is acquired using a transforming magnetic resonance (MR) device. 
     
     
         6 . The computer-implemented method of  claim 2 , wherein the volumetric medical image is a 2.5D volumetric image. 
     
     
         7 . The computer-implemented method of  claim 2 , wherein the volumetric medical image is created by stacking a number of image slices channel-wise. 
     
     
         8 . The computer-implemented method of  claim 2 , further comprising rotating the one or more reformat volumetric medical images into various angles to generate one or more rotated reformat medical images. 
     
     
         9 . The computer-implemented method of  claim 7 , further comprising applying the deep network model to the one or more rotated reformat medical images to output one or more predicted images. 
     
     
         10 . The computer-implemented method of  claim 2 , wherein the one or more predicted images are rotated to be aligned to a scanning plane. 
     
     
         11 . The computer-implemented method of  claim 2 , wherein the deep network model is a 2.5 D trained model. 
     
     
         12 . A non-transitory computer-readable storage medium including instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
 receiving a volumetric medical image of a subject;   generating one or more reformat volumetric medical images by reformatting the volumetric medical image of the subject in one or more orientations; and   feeding an input comprising the one or more reformat volumetric medical images to a deep network model to generate a predicted medical image with improved quality.   
     
     
         13 . The non-transitory computer-readable storage medium of  claim 12 , wherein the volumetric medical image is acquired in a first direction of a scanning plane. 
     
     
         14 . The non-transitory computer-readable storage medium of  claim 13 , wherein the one or more orientations comprise a second direction that is different from the first direction of the scanning plane. 
     
     
         15 . The non-transitory computer-readable storage medium of  claim 12 , wherein the volumetric medical image is acquired using a transforming magnetic resonance (MR) device. 
     
     
         16 . The non-transitory computer-readable storage medium of  claim 12 , wherein the volumetric medical image is a 2.5D volumetric image. 
     
     
         17 . The non-transitory computer-readable storage medium of  claim 12 , wherein the volumetric medical image is created by stacking a number of image slices channel-wise. 
     
     
         18 . The non-transitory computer-readable storage medium of  claim 12 , wherein the operations further comprise rotating the one or more reformat volumetric medical images into various angles to generate one or more rotated reformat medical images. 
     
     
         19 . The non-transitory computer-readable storage medium of  claim 12 , wherein the operations further comprise applying the deep network model to the one or more rotated reformat medical images to output one or more predicted images. 
     
     
         20 . The non-transitory computer-readable storage medium of  claim 12 , wherein the one or more predicted images are rotated to be aligned to a scanning plane. 
     
     
         21 . The non-transitory computer-readable storage medium of  claim 12 , wherein the deep network model is a 2.5 D trained model.