Medical image processing apparatus and medical image processing method
A medical image processing apparatus and a medical image processing method that can reduce the noise bias of a corrected image in which high absorber artifacts included in a reconstructed image are corrected. The medical image processing apparatus has an arithmetic unit for correcting high absorber artifacts. The arithmetic unit comprises: a projection data generating section that generates projection data corresponding to a high absorber area in a reconstructed image with high absorber artifacts; a noise image generating section that generates a noise image using the projection data; and a weighted combining section that makes a weighted combining of the noise image with a corrected image in which high absorber artifacts are corrected.
1 . A medical image processing apparatus comprising a processor and a program storage device tangibly embodying a program of instructions executable by the processor, the medical image processing apparatus including the processor performing a method comprising:
generating projection data corresponding to a high absorber area in a reconstructed image with high absorber artifacts;
generating a noise image using the projection data; and
making a weighted combining of the noise image with a corrected image in which high absorber artifacts are corrected, in the weighted combining the corrected image being added to a noise-adjusted image generated by weighting the noise image.
2 . The medical image processing apparatus according to claim 1 , wherein in the method, the processor generates the projection data corresponding to the high absorber area from a difference between the projection data corresponding to the reconstructed image and projection data of areas other than the high absorber area in the reconstructed image.
3 . A medical image processing apparatus comprising a processor and a program storage device tangibly embodying a program of instructions executable by the processor, the medical image processing apparatus including the processor performing a method comprising:
generating projection data corresponding to a high absorber area in a reconstructed image with high absorber artifacts, the projection data including odd-numbered projection data and even-numbered projection data generated by even-odd division of the projection data corresponding to the high absorber area;
generating an odd-numbered reconstructed image and an even-numbered reconstructed image by reconstructing the odd-numbered projection data and the even-numbered projection data;
generating a noise image using the projection data, the noise image being generated based on a difference between the odd-numbered reconstructed image and the even-numbered reconstructed image; and
making a weighted combining of the noise image with a corrected image in which high absorber artifacts are corrected.
4 . The medical image processing apparatus according to claim 1 , wherein the method further comprises
adding an image obtained by multiplying the noise image by a noise coefficient image generated based on a noise distribution of the reconstructed image, to the corrected image.
5 . The medical image processing apparatus according to claim 1 , further comprising:
a machine learning processor that is generated by machine-learning taking a reconstructed image without high absorber artifacts as a teacher image, and a reconstructed image with high absorber artifacts added to the teacher image and a noise-added image generated as input images.
6 . The medical image processing apparatus according to claim 1 , further comprising:
a machine learning processor that is generated by machine-learning taking a reconstructed image without high absorber artifacts as a teacher image, and a reconstructed image with high absorber artifacts added to the teacher image and a noise image generated as input images.
7 . A medical image processing method for correcting high absorber artifacts, comprising:
a projection data generating step of generating projection data corresponding to a high absorber area in a reconstructed image with high absorber artifacts;
a noise image generating step of generating a noise image using the projection data; and
a weighted combining step of making a weighted combining of the noise image with a corrected image in which high absorber artifacts are corrected, in the weighted combining the corrected image being added to a noise-adjusted image generated by weighting the noise image.
8 . The medical image processing method according to claim 7 , wherein the projection data corresponding to the high absorber area is generated from a difference between the projection data corresponding to the reconstructed image and projection data of areas other than the high absorber area in the reconstructed image.
9 . The medical image processing method according to claim 7 , wherein the noise-adjusted image is obtained by multiplying the noise image by a noise coefficient image generated based on a noise distribution of the reconstructed image, to the corrected image.
10 . The medical image processing apparatus according to claim 3 , wherein the projection data corresponding to the high absorber area is generated from a difference between the projection data corresponding to the reconstructed image and projection data of areas other than the high absorber area in the reconstructed image.
11 . The medical image processing apparatus according to claim 3 , wherein the noise-adjusted image is obtained by multiplying the noise image by a noise coefficient image generated based on a noise distribution of the reconstructed image, to the corrected image.