Information processing method, medical image diagnostic apparatus, and information processing system
An information processing method of an embodiment is a processing method of information acquired by imaging performed by a medical image diagnostic apparatus, the information processing method includes the steps of: acquiring noise data by imaging a phantom using a medical imaging apparatus; based on first subject projection data acquired by the imaging performed by a medical image diagnostic modality of a same kind as the medical image diagnostic apparatus and the noise data, acquiring synthesized subject data in which noise based on the noise data is added to the first subject projection data; and acquiring a noise reduction processing model by machine learning using the synthesized subject data and second subject projection data acquired by the imaging performed by the medical image diagnostic modality.
1. An information processing method of processing information acquired by imaging performed by a medical image diagnostic apparatus, the information processing method comprising the steps of:
acquiring noise data by imaging a phantom using a medical imaging apparatus, wherein the noise data is extracted based on a first reconstructed image and a second reconstructed image acquired by respectively reconstructing first and second image data sets of the phantom taken at different times;
based on first subject projection data acquired by imaging performed by a medical image diagnostic modality of a same kind as the medical image diagnostic apparatus and the noise data, acquiring synthesized subject data in which noise based on the noise data is added to the first subject projection data; and
acquiring a noise reduction processing model by machine learning using the synthesized subject data and second subject projection data acquired by the imaging performed by the medical image diagnostic modality.
2. The information processing method according to claim 1 , wherein the first subject projection data is acquired by the imaging performed by the medical image diagnostic apparatus.
3. The information processing method according to claim 1 , wherein the first subject projection data is acquired by the imaging performed by another medical image diagnostic apparatus, which is a same kind as the medical image diagnostic apparatus and is not the medical image diagnostic apparatus.
4. The information processing method according to claim 1 , wherein the medical image diagnostic apparatus and the medical imaging apparatus are the same apparatus.
5. The information processing method according to claim 1 , wherein the second subject projection data is data acquired by imaging a same subject as was imaged to obtain the second subject projection data.
6. The information processing method according to claim 1 , further comprising acquiring synthesized subject data in which noise based on other noise data is added to the first subject projection data, wherein the other noise data is data obtained by scaling reconstructed data obtained by imaging the phantom using low dose radiation.
7. The information processing method according to claim 1 , wherein the noise reduction processing model is acquired by training the model by deep learning, an input of which is the synthesized subject data and a target of which is the second subject projection data acquired by the imaging performed by the medical image diagnostic apparatus.
8. The information processing method according to claim 1 , further comprising acquiring synthesized subject data in which noise based on other noise data is added to the first subject projection data wherein the other noise data is (1) data obtained by scaling reconstructed data obtained by imaging the phantom using low-dose radiation, and (2) data obtained by not scaling reconstructed data obtained by imaging the phantom using the low-dose radiation.
9. The information processing method according to claim 1 , wherein phantom comprises a water phantom.
10. The information processing method according to claim 1 , wherein the phantom comprises a cylindrical phantom.
11. The information processing method according to claim 1 , wherein the first subject projection data is a reconstructed image acquired by reconstructing subject data by a first reconstruction method, and
the second subject projection data is a reconstructed image acquired by reconstructing the subject data by a second reconstruction method with higher accuracy than the first reconstruction method.
12. The information processing method according to claim 1 , wherein the first subject projection data and the second subject projection data are data acquired by low-dose imaging.
13. A medical image diagnostic apparatus, comprising:
processing circuitry configured to:
acquire input subject data by imaging a subject,
acquire denoised data by reducing noise on the input subject data by a noise reduction processing model acquired by training a model by machine learning using synthesized subject data and first subject projection data, the synthesized subject data being acquired based on noise data generated by imaging a phantom, wherein the noise data is extracted based on a first reconstructed image and a second reconstructed image acquired by reconstructing first and second image data sets of the phantom taken at different times, and
output an image of the subject based on the denoised data.
14. An information processing system, comprising:
processing circuitry configured to:
acquire input subject data by imaging a subject,
acquire denoised data by reducing noise on the input subject data by a noise reduction processing model acquired by training a model by machine learning using synthesized subject data and first subject projection data, the synthesized subject data being acquired based on noise data generated by imaging a phantom, wherein the noise data is extracted based on a first reconstructed image and a second reconstructed image acquired by reconstructing first and second image data sets of the phantom taken at different times, and
output an image of the subject based on the denoised data.
15. An information processing method of information acquired by imaging performed by a medical image diagnostic apparatus, the information processing method comprising:
acquiring noise data by imaging a phantom using a medical imaging apparatus, wherein the noise data is extracted based on a first reconstructed image and a second reconstructed image acquired by reconstructing first and second image data sets of the phantom taken at different times;
based on subject data of a subject acquired by imaging performed by a medical image diagnostic modality of a same kind as the medical image diagnostic apparatus, generating first subject projection data corresponding to a first subset of the subject data and second subject projection data corresponding to second subset of the subject data, which is different from the first subset;
generating synthesized subject data by synthesizing the first subject projection data and the noise data; and
acquiring a noise reduction processing model by machine learning, an input of which is the synthesized subject data and a target of which is the second subject projection data.