Matches in SemOpenAlex for { <https://semopenalex.org/work/W4380051497> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W4380051497 abstract "The exorbitant cost of accurately annotating the large-scale serial scanning electron microscope (SEM) images as the ground truth for training has always been a great challenge for brain map reconstruction by deep learning methods in neural connectome studies. The representation ability of the model is strongly correlated with the number of such high-quality labels. Recently, the masked autoencoder (MAE) has been shown to effectively pre-train Vision Transformers (ViT) to improve their representational capabilities.In this paper, we investigated a self-pre-training paradigm for serial SEM images with MAE to implement downstream segmentation tasks. We randomly masked voxels in three-dimensional brain image patches and trained an autoencoder to reconstruct the neuronal structures.We tested different pre-training and fine-tuning configurations on three different serial SEM datasets of mouse brains, including two public ones, SNEMI3D and MitoEM-R, and one acquired in our lab. A series of masking ratios were examined and the optimal ratio for pre-training efficiency was spotted for 3D segmentation. The MAE pre-training strategy significantly outperformed the supervised learning from scratch. Our work shows that the general framework of can be a unified approach for effective learning of the representation of heterogeneous neural structural features in serial SEM images to greatly facilitate brain connectome reconstruction." @default.
- W4380051497 created "2023-06-10" @default.
- W4380051497 creator A5020691605 @default.
- W4380051497 creator A5030105361 @default.
- W4380051497 creator A5059791894 @default.
- W4380051497 creator A5067339749 @default.
- W4380051497 date "2023-06-08" @default.
- W4380051497 modified "2023-09-25" @default.
- W4380051497 title "Learning the heterogeneous representation of brain's structure from serial SEM images using a masked autoencoder" @default.
- W4380051497 cites W1513082520 @default.
- W4380051497 cites W1608731182 @default.
- W4380051497 cites W2033403400 @default.
- W4380051497 cites W2080858319 @default.
- W4380051497 cites W2995808743 @default.
- W4380051497 cites W2997574889 @default.
- W4380051497 cites W3089720110 @default.
- W4380051497 cites W3106295246 @default.
- W4380051497 cites W3109531692 @default.
- W4380051497 cites W3159799042 @default.
- W4380051497 cites W3170863103 @default.
- W4380051497 doi "https://doi.org/10.3389/fninf.2023.1118419" @default.
- W4380051497 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37360945" @default.
- W4380051497 hasPublicationYear "2023" @default.
- W4380051497 type Work @default.
- W4380051497 citedByCount "0" @default.
- W4380051497 crossrefType "journal-article" @default.
- W4380051497 hasAuthorship W4380051497A5020691605 @default.
- W4380051497 hasAuthorship W4380051497A5030105361 @default.
- W4380051497 hasAuthorship W4380051497A5059791894 @default.
- W4380051497 hasAuthorship W4380051497A5067339749 @default.
- W4380051497 hasBestOaLocation W43800514971 @default.
- W4380051497 hasConcept C101738243 @default.
- W4380051497 hasConcept C142362112 @default.
- W4380051497 hasConcept C153180895 @default.
- W4380051497 hasConcept C153349607 @default.
- W4380051497 hasConcept C154945302 @default.
- W4380051497 hasConcept C15744967 @default.
- W4380051497 hasConcept C169760540 @default.
- W4380051497 hasConcept C17744445 @default.
- W4380051497 hasConcept C199539241 @default.
- W4380051497 hasConcept C2776359362 @default.
- W4380051497 hasConcept C2777402240 @default.
- W4380051497 hasConcept C3018011982 @default.
- W4380051497 hasConcept C31972630 @default.
- W4380051497 hasConcept C41008148 @default.
- W4380051497 hasConcept C45715564 @default.
- W4380051497 hasConcept C50644808 @default.
- W4380051497 hasConcept C54170458 @default.
- W4380051497 hasConcept C89600930 @default.
- W4380051497 hasConcept C94625758 @default.
- W4380051497 hasConceptScore W4380051497C101738243 @default.
- W4380051497 hasConceptScore W4380051497C142362112 @default.
- W4380051497 hasConceptScore W4380051497C153180895 @default.
- W4380051497 hasConceptScore W4380051497C153349607 @default.
- W4380051497 hasConceptScore W4380051497C154945302 @default.
- W4380051497 hasConceptScore W4380051497C15744967 @default.
- W4380051497 hasConceptScore W4380051497C169760540 @default.
- W4380051497 hasConceptScore W4380051497C17744445 @default.
- W4380051497 hasConceptScore W4380051497C199539241 @default.
- W4380051497 hasConceptScore W4380051497C2776359362 @default.
- W4380051497 hasConceptScore W4380051497C2777402240 @default.
- W4380051497 hasConceptScore W4380051497C3018011982 @default.
- W4380051497 hasConceptScore W4380051497C31972630 @default.
- W4380051497 hasConceptScore W4380051497C41008148 @default.
- W4380051497 hasConceptScore W4380051497C45715564 @default.
- W4380051497 hasConceptScore W4380051497C50644808 @default.
- W4380051497 hasConceptScore W4380051497C54170458 @default.
- W4380051497 hasConceptScore W4380051497C89600930 @default.
- W4380051497 hasConceptScore W4380051497C94625758 @default.
- W4380051497 hasLocation W43800514971 @default.
- W4380051497 hasLocation W43800514972 @default.
- W4380051497 hasLocation W43800514973 @default.
- W4380051497 hasOpenAccess W4380051497 @default.
- W4380051497 hasPrimaryLocation W43800514971 @default.
- W4380051497 hasRelatedWork W1669643531 @default.
- W4380051497 hasRelatedWork W2005437358 @default.
- W4380051497 hasRelatedWork W2008656436 @default.
- W4380051497 hasRelatedWork W2023558673 @default.
- W4380051497 hasRelatedWork W2134924024 @default.
- W4380051497 hasRelatedWork W2292254049 @default.
- W4380051497 hasRelatedWork W2517104666 @default.
- W4380051497 hasRelatedWork W2897995864 @default.
- W4380051497 hasRelatedWork W2998168123 @default.
- W4380051497 hasRelatedWork W4287995534 @default.
- W4380051497 hasVolume "17" @default.
- W4380051497 isParatext "false" @default.
- W4380051497 isRetracted "false" @default.
- W4380051497 workType "article" @default.