Medical information processing apparatus, magnetic resonance imaging apparatus, and medical information processing method

1. A medical information processing apparatus, comprising:

processing circuitry configured to

acquire first data generated as a result of a reception by a radio frequency coil and acquire information about reception characteristics of the radio frequency coil that contribute to a signal-to-noise ratio of the first data;

perform a denoise process on the first data to obtain second data from the first data, the second data being the denoised first data;

determine multiple mixing rates based on the reception characteristics of the radio frequency coil, each mixing rate being a rate of a mixture of a first medical image according to the first data and a second medical image according to the second data; and

generate a mixed image by mixing the first medical image and the second medical image according to the determined multiple mixing rates,

wherein the processing circuitry is further configured to

acquire a reception sensitivity map as the acquired information;

multiply respective pixel values of the first medical image and the second medical image by weighting factors according to the reception sensitivity map; and

generate the mixed image by summing the medical images having the pixel values multiplied by the weighting factors.

2. The medical information processing apparatus according to claim 1 , wherein the acquired information about the reception characteristics of the radio frequency coil includes a g-factor map.

3. The medical information processing apparatus according to claim 1 , wherein the processing circuitry is further configured to

generate a third map with which the weighting factors that differ according to positions are associated as the multiple mixing rates, based on the reception sensitivity map; and

multiply the respective pixel values of the first medical image and the second medical image by the weighting factors based on the generated third map.

4. The medical information processing apparatus according to claim 1 , wherein the processing circuitry is further configured to

acquire a g-factor map as the acquired information;

multiply the respective pixel values of the first medical image and the second medical image by weighting factors according to the g-factor map; and

generate the mixed image by summing the medical images having the pixel values multiplied by the weighting factors.

5. The medical information processing apparatus according to claim 4 , wherein the processing circuitry is further configured to

generate a third map with which the weighting factors that differ according to positions are associated as the multiple mixing rates based on the g-factor map; and

multiply the respective pixel values of the first medical image and the second medical image by the weighting factors based on the third map.

6. A medical information processing apparatus comprising:

processing circuitry configured to

acquire data and a g-factor map, the data being generated as a result of a reception by a radio frequency coil;

perform a denoise process on the acquired data by inputting the acquired data into a neural network having an activation function; and

determine a strength of the denoise process by determining a threshold of the activation function based on the acquired g-factor map.

7. The medical information processing apparatus according to claim 6 , wherein the processing circuitry is further configured to

determine a first threshold value and a second threshold value greater than the first threshold value with respect to a signal intensity of the data based on the acquired g-factor map; and

remove the data as noise when a signal intensity of the acquired data is equal to or greater than the first threshold value and equal to or less than the second threshold value.

8. The medical information processing apparatus according to claim 7 ,

wherein the processing circuitry is further configured to determine the threshold values based on a first parameter trained under a condition that noise of the data is spatially uniform, a second parameter varying according to the noise of the data, and g-factors included in the acquired g-factor map.

9. The medical information processing apparatus according to claim 8 ,

wherein the processing circuitry is further configured to determine the threshold values based on values obtained by multiplying a product of the first parameter and the second parameter by the g-factors.

10. A magnetic resonance imaging apparatus, comprising:

a medical image processing apparatus according to claim 6 ; and

a radio frequency coil configured to receive a magnetic resonance signal,

wherein the processing circuitry is further configured to generate the data as a result of a reception by the radio frequency coil.

11. A medical image processing method comprising, using a computer:

acquiring data and a g-factor map, the data being generated as a result of a reception by a radio frequency coil;

performing a denoise process on the acquired data by inputting the acquired data into a neural network having an activation function; and

determining a strength of the denoise process by determining a threshold of the activation function based on the acquired g-factor map.