Magnetic resonance imaging apparatus, image processing apparatus, and image processing method

1. A magnetic resonance imaging apparatus, comprising:

an MRI (magnetic resonance imaging) system including a receiving coil to receive magnetic resonance signals; and

processing circuitry configured to:

generate a plurality of images imaged at different times, based on the magnetic resonance signals, each of the plurality of images including a plurality of pixels;

calculate feature vectors of respective pixels included in one image of the plurality of images, the calculating the feature vectors includes calculating, as an element of each feature vector, a fluctuation quantity of a signal value for the respective pixel based on signal values of pixels included in the plurality of images;

correct the calculated feature vectors, including correcting the calculated fluctuation quantity of each feature vector based on a sensitivity of the receiving coil; and

reduce noise in the one image for output, based on distribution of the corrected feature vectors.

2. The apparatus according to claim 1 , wherein

the processing circuitry is configured to correct the feature vectors of the respective pixels, further based on a geometry factor used in high-speed imaging.

3. The apparatus according to claim 1 , wherein

the processing circuitry is configured to:

calculate a correction map based on the sensitivity of the receiving coil, and

correct the feature vectors of the respective pixels using the correction map.

4. The apparatus according to claim 3 , wherein

a value of the correction map is calculated by dividing a value of the sensitivity of the receiving coil by a geometry factor used in high-speed imaging.

5. The apparatus according to claim 3 , wherein

the correction map has a larger value in a position in which the receiving coil has higher sensitivity.

6. The apparatus according to claim 3 , wherein

the correction map has a larger value in a position having lower calculation accuracy of a signal value in image reconstruction, when the one image is obtained by high-speed imaging in which a plurality of receiving coils are used and imaging is performed using difference in sensitivity among the receiving coils.

7. The apparatus according to claim 3 , wherein

the processing circuitry is configured to correct at least one element of the feature vectors of the respective pixels such that an element corresponding to a position having a smaller value in the correction map has a smaller value and an element corresponding to a position having a larger value in the correction map has a larger value.

8. The apparatus according to claim 3 , wherein

the processing circuitry is configured to decrease an intensity of noise removal as a value of the correction map increases.

9. The apparatus according to claim 1 , wherein

the processing circuitry is configured to:

select a noise model from a plurality of noise models, based on the corrected feature vectors, and

reduce the noise in the one image using the selected noise model.

10. The apparatus according to claim 9 , wherein

the noise model outputs a noise quantity for an input signal value, outputs a smaller noise quantity as the input signal value is smaller, and outputs a larger noise quantity as the input signal value is larger, and the output noise quantity converges to a fixed value as the input signal value increases.

11. The apparatus according to claim 9 , wherein

the processing circuitry is configured to select the noise model that most approximates to a data point group indicated by the corrected feature vectors, from the plurality of noise models.

12. The apparatus according to claim 9 , wherein

the processing circuitry is configured to increase an intensity of noise removal as a noise quantity that is output from the selected noise model increases.

13. The apparatus according to claim 9 , wherein

the processing circuitry is configured to:

generate a moving image as the plurality of images,

detect a motion region and a static region including movement smaller than movement in the motion region from the moving image, and

reduce noise by different methods for the motion region and the static region.

14. The apparatus according to claim 13 , wherein

the processing circuitry is configured to detect one of the plurality of pixels having a larger temporal fluctuation quantity of a signal value of the one of the plurality of pixels in an identical position in a plurality of frames included in the moving image than a threshold based on a noise quantity obtained from the noise model, as a pixel of the motion region.

15. The apparatus according to claim 13 , wherein

the processing circuitry is configured to set an intensity of noise removal for pixels included in the motion region to be lower than an intensity of noise removal for pixels included in the static region, when the pixels included in the motion region and the pixels included in the static region have an equal noise quantity obtained from the noise model.

16. The apparatus according to claim 1 , wherein

the processing circuitry is configured to calculate the fluctuation quantity of the signal value for the respective pixel, based on the signal value of the respective pixel and signal values of other pixels located in positions spatially or temporally close to the respective pixel.

17. The apparatus according to claim 1 , wherein

each feature value vector includes a plurality of elements.

18. The apparatus according to claim 17 , wherein

the fluctuation quantity comprises a standard deviation of a respective signal value.

19. An image processing apparatus, comprising:

a processor; and

a memory that stores processor-executable instructions that, when executed by the processor, cause the processor to:

acquire a plurality of images imaged at different times and generated based on magnetic resonance signals, each of the plurality of images including a plurality of pixels;

calculate feature vectors of respective pixels included in one image of the plurality of images, the calculating the feature vectors includes calculating, as an element of each feature vector, a fluctuation quantity of a signal value for the respective pixel based on signal values of pixels included in the plurality of images;

correct the calculated feature vectors, including correcting the calculated fluctuation quantity of each feature vector based on a sensitivity of the receiving coil to receive the magnetic resonance signal; and

reduce noise in the one image for output based on distribution of the corrected feature vectors.

20. An image processing method, comprising:

acquiring a plurality of images imaged at different times and generated based on magnetic resonance signals, each of the plurality of images including a plurality of pixels;

calculating feature vectors of respective pixels included in one image of the plurality of images, the calculating the feature vectors includes calculating, as an element of each feature vector, a fluctuation quantity of a signal value for the respective pixel based on signal values of pixels included in the plurality of images;

correcting the calculated feature vectors, including correcting the calculated fluctuation quantity of each feature vector based on a sensitivity of the receiving coil to receive the magnetic resonance signal; and

reducing noise in the one image for output based on distribution of the corrected feature vectors.