Virtual ct images from magnetic resonance images

A computer program product comprising machine executable instructions for execution by a processor controlling a medical imaging system, wherein execution of the machine executable instructions causes the processor to:

• access multiple magnetic resonance images of respective different contrast types and reconstructed from multiple series of magnetic resonance data, from a subject acquired by several acquisition protocols, wherein each of the multiple magnetic resonance images comprises voxels;
• calculate an image segmentation for each of the multiple magnetic resonance images, wherein the image segmentation divides each of the multiple magnetic resonance images into regions;
• assign a tissue classification to each of the regions using an magnetic resonance imaging tissue classifier;
• calculate a Hounsfield unit map for each of the multiple magnetic resonance images by assigning a Hounsfield unit value to each of the voxels according to the tissue classification, wherein the Hounsfield mapping comprises a mapping between the tissue classification to Hounsfield units;
• and
• calculate a virtual CT image for each of the multiple magnetic resonance images using the Hounsfield unit mapping.

A medical imaging method, wherein the method comprises:

• accessing multiple magnetic resonance images of respective different contrast types and reconstructed from multiple series of magnetic resonance data, from a subject acquired by several acquisition protocols, wherein each of the multiple magnetic resonance images comprises voxels;
• calculating an image segmentation for each of the multiple magnetic resonance images, wherein the image segmentation divides each of the multiple magnetic resonance images into regions;
• assigning a tissue classification to each of the regions;
• calculating a Hounsfield unit map for each of the multiple magnetic resonance images by assigning a Hounsfield unit value to each of the voxels according to the tissue classification, wherein the Hounsfield mapping comprises a mapping between the tissue classification to Hounsfield units;
• and
• calculating a virtual CT image for each of the multiple magnetic resonance images using the Hounsfield unit mapping.

A magnetic resonance imaging system comprising:

• a memory for storing machine executable instructions and
• a processor for controlling the medical imaging system, wherein execution of the machine executable instructions causes the processor to:
• acquire multiple series of magnetic resonance data of respective different contrast types, from a subject acquired by several acquisition protocols
• reconstruct the multiple series of magnetic resonance data into the multiple magnetic resonance images, wherein each of the multiple magnetic resonance images comprises voxels;
• calculate an image segmentation for each of the multiple magnetic resonance images, wherein the image segmentation divides each of the multiple
• assign a tissue classification to each of the regions using an magnetic resonance imaging tissue classifier;
• calculate a Hounsfield unit map for the magnetic resonance image by assigning a Hounsfield unit value to each of the voxels according to the tissue classification, wherein the Hounsfield mapping comprises a mapping between the tissue classification to Hounsfield units;
• calculate a virtual CT image for each of the multiple magnetic resonance images using the Hounsfield unit mapping.