FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386192371> ?p ?o ?g. }

• W4386192371 endingPage "107391" @default.
• W4386192371 startingPage "107391" @default.
• W4386192371 abstract "Deep learning (DL)-based denoising of low-dose positron emission tomography (LDPET) and low-dose computed tomography (LDCT) has been widely explored. However, previous methods have focused only on single modality denoising, neglecting the possibility of simultaneously denoising LDPET and LDCT using only one neural network, i.e., joint LDPET/LDCT denoising. Moreover, DL-based denoising methods generally require plenty of well-aligned LD-normal-dose (LD-ND) sample pairs, which can be difficult to obtain. To this end, we propose a self-supervised two-stage training framework named MAsk-then-Cycle (MAC), to achieve self-supervised joint LDPET/LDCT denoising. The first stage of MAC is masked autoencoder (MAE)-based pre-training and the second stage is self-supervised denoising training. Specifically, we propose a self-supervised denoising strategy named cycle self-recombination (CSR), which enables denoising without well-aligned sample pairs. Unlike other methods that treat noise as a homogeneous whole, CSR disentangles noise into signal-dependent and independent noises. This is more in line with the actual imaging process and allows for flexible recombination of noises and signals to generate new samples. These new samples contain implicit constraints that can improve the network's denoising ability. Based on these constraints, we design multiple loss functions to enable self-supervised training. Then we design a CSR-based denoising network to achieve joint 3D LDPET/LDCT denoising. Existing self-supervised methods generally lack pixel-level constraints on networks, which can easily lead to additional artifacts. Before denoising training, we perform MAE-based pre-training to indirectly impose pixel-level constraints on networks. Experiments on an LDPET/LDCT dataset demonstrate its superiority over existing methods. Our method is the first self-supervised joint LDPET/LDCT denoising method. It does not require any prior assumptions and is therefore more robust." @default.
• W4386192371 created "2023-08-27" @default.
• W4386192371 creator A5021073698 @default.
• W4386192371 creator A5026000486 @default.
• W4386192371 creator A5033311734 @default.
• W4386192371 creator A5036398299 @default.
• W4386192371 creator A5053995020 @default.
• W4386192371 creator A5085263462 @default.
• W4386192371 date "2023-10-01" @default.
• W4386192371 modified "2023-09-29" @default.
• W4386192371 title "Self-supervised deep learning for joint 3D low-dose PET/CT image denoising" @default.
• W4386192371 cites W1964231221 @default.
• W4386192371 cites W1983702282 @default.
• W4386192371 cites W1998929291 @default.
• W4386192371 cites W2194775991 @default.
• W4386192371 cites W2319126251 @default.
• W4386192371 cites W2385563966 @default.
• W4386192371 cites W2574952845 @default.
• W4386192371 cites W2584483805 @default.
• W4386192371 cites W2593414223 @default.
• W4386192371 cites W2617128058 @default.
• W4386192371 cites W2743780012 @default.
• W4386192371 cites W2748739903 @default.
• W4386192371 cites W2789588857 @default.
• W4386192371 cites W2798538010 @default.
• W4386192371 cites W2890139949 @default.
• W4386192371 cites W2962793481 @default.
• W4386192371 cites W2963446712 @default.
• W4386192371 cites W3000495410 @default.
• W4386192371 cites W3002137088 @default.
• W4386192371 cites W3022695694 @default.
• W4386192371 cites W3035355889 @default.
• W4386192371 cites W3041500444 @default.
• W4386192371 cites W3103261259 @default.
• W4386192371 cites W3110460200 @default.
• W4386192371 cites W3169902780 @default.
• W4386192371 cites W3172797549 @default.
• W4386192371 cites W3180514382 @default.
• W4386192371 cites W3185100494 @default.
• W4386192371 cites W3193594030 @default.
• W4386192371 cites W3205522968 @default.
• W4386192371 cites W3214080919 @default.
• W4386192371 cites W4210547008 @default.
• W4386192371 cites W4214604401 @default.
• W4386192371 cites W4225162166 @default.
• W4386192371 cites W4225243441 @default.
• W4386192371 cites W4284894230 @default.
• W4386192371 cites W4292136262 @default.
• W4386192371 cites W4292183902 @default.
• W4386192371 cites W4306158369 @default.
• W4386192371 cites W4307211638 @default.
• W4386192371 cites W4312209940 @default.
• W4386192371 cites W4312624488 @default.
• W4386192371 cites W4315702334 @default.
• W4386192371 cites W4319068678 @default.
• W4386192371 cites W4322625633 @default.
• W4386192371 cites W4361861727 @default.
• W4386192371 cites W4366748056 @default.
• W4386192371 cites W4376619205 @default.
• W4386192371 doi "https://doi.org/10.1016/j.compbiomed.2023.107391" @default.
• W4386192371 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37717529" @default.
• W4386192371 hasPublicationYear "2023" @default.
• W4386192371 type Work @default.
• W4386192371 citedByCount "0" @default.
• W4386192371 crossrefType "journal-article" @default.
• W4386192371 hasAuthorship W4386192371A5021073698 @default.
• W4386192371 hasAuthorship W4386192371A5026000486 @default.
• W4386192371 hasAuthorship W4386192371A5033311734 @default.
• W4386192371 hasAuthorship W4386192371A5036398299 @default.
• W4386192371 hasAuthorship W4386192371A5053995020 @default.
• W4386192371 hasAuthorship W4386192371A5085263462 @default.
• W4386192371 hasConcept C101453961 @default.
• W4386192371 hasConcept C115961682 @default.
• W4386192371 hasConcept C119857082 @default.
• W4386192371 hasConcept C136389625 @default.
• W4386192371 hasConcept C153180895 @default.
• W4386192371 hasConcept C154945302 @default.
• W4386192371 hasConcept C163294075 @default.
• W4386192371 hasConcept C202474056 @default.
• W4386192371 hasConcept C23431618 @default.
• W4386192371 hasConcept C2781238097 @default.
• W4386192371 hasConcept C2983327147 @default.
• W4386192371 hasConcept C30814859 @default.
• W4386192371 hasConcept C41008148 @default.
• W4386192371 hasConcept C50644808 @default.
• W4386192371 hasConcept C99498987 @default.
• W4386192371 hasConceptScore W4386192371C101453961 @default.
• W4386192371 hasConceptScore W4386192371C115961682 @default.
• W4386192371 hasConceptScore W4386192371C119857082 @default.
• W4386192371 hasConceptScore W4386192371C136389625 @default.
• W4386192371 hasConceptScore W4386192371C153180895 @default.
• W4386192371 hasConceptScore W4386192371C154945302 @default.
• W4386192371 hasConceptScore W4386192371C163294075 @default.
• W4386192371 hasConceptScore W4386192371C202474056 @default.
• W4386192371 hasConceptScore W4386192371C23431618 @default.
• W4386192371 hasConceptScore W4386192371C2781238097 @default.
• W4386192371 hasConceptScore W4386192371C2983327147 @default.
• W4386192371 hasConceptScore W4386192371C30814859 @default.

• next