English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Deep image prior (DIP) has recently attracted attention owing to itsunsupervised positron emission tomography (PET) image reconstruction, whichdoes not require any prior training dataset. In this paper, we present thefirst attempt to implement an end-to-end DIP-based fully 3D PET imagereconstruction method that incorporates a forward-projection model into a lossfunction. To implement a practical fully 3D PET image reconstruction, whichcould not be performed due to a graphics processing unit memory limitation, wemodify the DIP optimization to block-iteration and sequentially learn anordered sequence of block sinograms. Furthermore, the relative differencepenalty (RDP) term was added to the loss function to enhance the quantitativePET image accuracy. We evaluated our proposed method using Monte Carlosimulation with [$^{18}$F]FDG PET data of a human brain and a preclinical studyon monkey brain [$^{18}$F]FDG PET data. The proposed method was compared withthe maximum-likelihood expectation maximization (EM), maximum-a-posterior EMwith RDP, and hybrid DIP-based PET reconstruction methods. The simulationresults showed that the proposed method improved the PET image quality byreducing statistical noise and preserved a contrast of brain structures andinserted tumor compared with other algorithms. In the preclinical experiment,finer structures and better contrast recovery were obtained by the proposedmethod. This indicated that the proposed method can produce high-quality imageswithout a prior training dataset. Thus, the proposed method is a key enablingtechnology for the straightforward and practical implementation of end-to-endDIP-based fully 3D PET image reconstruction.

Fully 3D Implementation of the End-to-end Deep Image Prior-based PET Image Reconstruction Using Block Iterative Algorithm