Deep-Learning-based T1-Enhanced Selection of Linear Coefficients (DL-TESLA) for PET/MR Attenuation Correction
AssigneeWashington University
InventorsHongyu An, Chunwei Ying, Yasheng Chen
Life of this patent
Estimating the term…
Abstract
Systems and methods for deep-learning-based T1-enhanced selection of linear attenuation coefficients (DL-TESLA) for PET/MR attenuation are described.
Claims8
A system for obtaining a linear attenuation coefficient map based on at least one MR image, the system comprising a computing device with at least one processor and a non-volatile computer-readable memory, the non-volatile computer-readable memory containing a plurality of instructions executable on the at least one processor, the plurality of instructions configured to:
- receive the at least one MR image, each MR image comprising an FA1TE1 map, an FA2TE2 map, and an R1 map;
- transform each MR image into a pseudo-CT map using a DL-TESLA model based on the FA1TE1 map, the FA2TE2 map, and the R1 map;
- and
- transform the pseudo-CT map into the linear attenuation coefficient map using piecewise linear scaling.
A computer-implemented method for obtaining a linear attenuation coefficient map based on at least one MR image, the method comprising:
- receiving, using the computing device, the at least one MR image, each MR image comprising an FA1TE1 map, an FA2TE2 map, and an R1 map;
- transforming, using the computing device, each MR image into a pseudo-CT map using a DL-TESLA model based on the FA1TE1 map, the FA2TE2 map, and the R1 map;
- and
- transforming, using the computing device, the pseudo-CT map into the linear attenuation coefficient map using piecewise linear scaling.
Dependencies are inferred from the claim text.