Method and apparatus for performing single gating in positron emission tomograpy sytems
Inventors
Qi, Wenyuan • Yang, Li • Kolthammer, Jeffrey • TSAI, Yu-Jung • ASMA, Evren • Iatrou, Maria • Qi, Jinyi • LI, Tiantian
Assignees
Canon Inc • Canon Medical Systems Corp • University of California San Diego UCSD
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A method for performing single gating in a positron emission tomography (PET) system includes: receiving list-mode data acquired by scanning an imaging object using the PET system, the list-mode data being affected by quasi-periodic motion of the imaging object; producing a plurality of vectors based on the received list-mode data; generating a reference vector based on the produced plurality of vectors; selecting, from the produced plurality of vectors, a set of vectors corresponding to a single gate, based on respective differences compared with the generated reference vector; and generating an image of the imaging object based on the selected set of vectors.
Core Innovation
The described invention processes positron emission tomography (PET) list-mode data affected by quasi-periodic motion, including respiratory motion and cardiac motion, by applying single gating. The method divides the list-mode data into a plurality of sets of data and generates a first plurality of vectors corresponding to different ones of the plurality of sets of data. These vectors represent data-derived representations such as sinograms, downsampled sinograms, latent vectors via autoencoders, or image-derived/downsampled equivalents.
A reference vector corresponding to a target respiration phase is generated based on a second plurality of vectors included in the first plurality of vectors, where each of the second plurality of vectors corresponds to the target respiration phase. The method computes respective differences between the generated reference vector and the first plurality of vectors, and selects, from the plurality of sets of data, a set of data to be used to reconstruct an image based on the differences. The image is then generated based on the selected set of data.
To define and choose the target respiration phase and the amount of selected data, the method uses a data-driven reference-vector approach and reports examples such as deriving respiratory waveform information using principal component analysis (PCA) and independent component analysis (ICA) on mini vectors or from an external device. The selection can include a count percentage ranked according to vector-difference variation, with an optional turning point obtained from a curve relating count percentage to vector-difference variation, and is contrasted with fixed-phase quiescent gating, including end-expiration.
Claims Coverage
The independent claims cover two aspects, a method and an apparatus, each processing PET list-mode data with a reference vector for a target respiration phase and selecting a data set for image reconstruction based on differences between vectors. Across these independent claims, the core inventive features can be expressed using three inventive elements.
Reference-vector single gating for target respiration phase
generating a reference vector corresponding to a target respiration phase included in a cycle of a respiration of the imaging object based on a second plurality of vectors included in the first plurality of vectors, each of the second plurality of vectors corresponding to the target respiration phase
Partition list-mode data into vector-represented sets
dividing the list-mode data into a plurality of sets of data; generating a first plurality of vectors corresponding to a different one of the plurality of sets of data
Select reconstruction data set by reference-vector differences
selecting, from the plurality of sets of data, a set of data to be used to reconstruct an image, based on respective differences between the generated reference vector and the first plurality of vectors
Both independent claims require generating a reference vector for a target respiration phase from vectors tied to that phase, generating vectors for data sets obtained by dividing list-mode data, selecting one data set for image reconstruction based on differences to the reference vector, and generating the image based on the selected set.
Stated Advantages
Phantom image improvements versus ungated and fixed-phase quiescent gating, reported using contrast recovery coefficient (CRC).
Documented Applications
Processing PET list-mode data during scanning of an imaging object affected by quasi-periodic motion, including respiratory motion and cardiac motion, to perform single gating and reconstruct an image.
Interested in licensing this patent?