Medical image processing apparatus, and nuclear medicine diagnosis apparatus
A medical image processing apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured to obtain Time-of-Flight (TOF) depiction image data generated on the basis of an annihilation point of a gamma ray. The processing circuitry is configured to output reconstructed Positron Emission computed Tomography (PET) image data on the basis of the TOF depiction image data and a trained model that outputs the reconstructed PET image data on the basis of an input of the TOF depiction image data.
1. A medical image processing apparatus, comprising:
processing circuitry configured to:
obtain Time-of-Flight (TOF) depiction image data generated on a basis of an annihilation point of a gamma ray, the TOF depiction image data representing a TOF depiction image in which annihilation point indicating a position of occurrence of annihilation event is depicted as a point or a line; and
output reconstructed Positron Emission computed Tomography (PET) image data on a basis of the TOF depiction image data and a trained model that outputs the reconstructed PET image data on a basis of an input of the TOF depiction image data,
wherein the processing circuitry outputs the reconstructed PET image data, on a basis of the trained model corresponding to a type of drug administered for a subject in a scan at a time of acquiring coincidence counting information from which the obtained TOF depiction image data is derived, the trained model being one of a plurality of trained models corresponding to a plurality of types of drugs.
2. The medical image processing apparatus according to claim 1 , wherein the processing circuitry outputs the reconstructed PET image data, on the basis of the trained model generated by learning a plurality of pieces of TOF depiction image data having a plurality of mutually-different noise levels.
3. The medical image processing apparatus according to claim 1 , wherein the processing circuitry obtains the TOF depiction image data based on coincidence counting information acquired with respect to each of sections of a subject.
4. The medical image processing apparatus according to claim 1 , wherein the processing circuitry obtains the TOF depiction image data based on coincidence counting information acquired while a subject is being moved relative to a PET gantry device.
5. The medical image processing apparatus according to claim 1 , wherein the processing circuitry outputs the reconstructed PET image data, on a basis of the trained model corresponding to a site subject to a scan performed at a time of acquiring coincidence counting information from which the obtained TOF depiction image data is derived, the trained model being one of a plurality of trained models corresponding to a plurality of sites of a subject.
6. The medical image processing apparatus according to claim 5 , wherein
the processing circuitry derives site information indicating a site, on a basis of the obtained TOF depiction image data and another trained model that outputs the site information on a basis of an input of the TOF depiction image data, and
the processing circuitry outputs the reconstructed PET image data, on a basis of one of the plurality of trained models that corresponds to the site indicated by the derived site information.
7. The medical image processing apparatus according to claim 5 , wherein
the processing circuitry detects a site from image data in a position that matches a position of the obtained TOF depiction image data, and
the processing circuitry outputs the reconstructed PET image data on a basis of one of the plurality of trained models that corresponds to the detected site.
8. The medical image processing apparatus according to claim 1 , wherein the processing circuitry outputs the reconstructed PET image data, on a basis of the TOF depiction image data, an attenuation coefficient map, and the trained model that outputs the reconstructed PET image data on a basis of inputs of the TOF depiction image data and the attenuation coefficient map.
9. The medical image processing apparatus according to claim 1 , wherein the processing circuitry outputs the reconstructed PET image data, on a basis of the TOF depiction image data, morphological image data, and the trained model that outputs the reconstructed PET image data on a basis of inputs of the TOF depiction image data and the morphological image data.
10. The medical image processing apparatus according to claim 1 , comprising an interface capable of selecting whether the processing circuitry is caused to reconstruct the reconstructed PET image data by implementing a successive approximation reconstruction method or the processing circuitry is caused to derive the reconstructed PET image data by using the TOF depiction image data.
11. A nuclear medicine diagnosis apparatus comprising:
processing circuitry configured to:
acquire Time-of-Flight (TOF) depiction image data generated on a basis of an annihilation point of a gamma ray, the TOF depiction image data representing a TOF depiction image in which annihilation point indicating a position of occurrence of annihilation event is depicted as a point or a line; and
output reconstructed Positron Emission computed Tomography (PET) image data on a basis of the TOF depiction image data and a trained model that outputs the reconstructed PET image data on a basis of an input of the TOF depiction image data,
wherein the processing circuitry outputs the reconstructed PET image data, on a basis of the trained model corresponding to a type of drug administered for a subject in a scan at a time of acquiring coincidence counting information from which the obtained TOF depiction image data is derived, the trained model being one of a plurality of trained models corresponding to a plurality of types of drugs.