{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Haertter D"],"funding":["Studienstiftung des Deutschen Volkes","Soft Matter Center, Duke University","National Institutes of Health","NIAMS NIH HHS","NIH HHS","NIGMS NIH HHS"],"pagination":["dev200621"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC9686994"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["149(21)"],"pubmed_abstract":["The efficient extraction of image data from curved tissue sheets embedded in volumetric imaging data remains a serious and unsolved problem in quantitative studies of embryogenesis. Here, we present DeepProjection (DP), a trainable projection algorithm based on deep learning. This algorithm is trained on user-generated training data to locally classify 3D stack content, and to rapidly and robustly predict binary masks containing the target content, e.g. tissue boundaries, while masking highly fluorescent out-of-plane artifacts. A projection of the masked 3D stack then yields background-free 2D images with undistorted fluorescence intensity values. The binary masks can further be applied to other fluorescent channels or to extract local tissue curvature. DP is designed as a first processing"],"journal":["Development (Cambridge, England)"],"pubmed_title":["DeepProjection: specific and robust projection of curved 2D tissue sheets from 3D microscopy using deep learning."],"pmcid":["PMC9686994"],"funding_grant_id":["R35GM127059","R35 GM127059","R01 AR076342","R01-AR076342"],"pubmed_authors":["Haertter D","Crawford JM","Fogerson SM","Di Talia S","Schmidt CF","Ramkumar N","Wang X","Kiehart DP","Poss KD"],"additional_accession":[]},"is_claimable":false,"name":"DeepProjection: specific and robust projection of curved 2D tissue sheets from 3D microscopy using deep learning.","description":"The efficient extraction of image data from curved tissue sheets embedded in volumetric imaging data remains a serious and unsolved problem in quantitative studies of embryogenesis. Here, we present DeepProjection (DP), a trainable projection algorithm based on deep learning. This algorithm is trained on user-generated training data to locally classify 3D stack content, and to rapidly and robustly predict binary masks containing the target content, e.g. tissue boundaries, while masking highly fluorescent out-of-plane artifacts. A projection of the masked 3D stack then yields background-free 2D images with undistorted fluorescence intensity values. The binary masks can further be applied to other fluorescent channels or to extract local tissue curvature. DP is designed as a first processing","dates":{"release":"2022-01-01T00:00:00Z","publication":"2022 Nov","modification":"2026-05-28T01:38:29.42Z","creation":"2025-04-07T07:47:22.607Z"},"accession":"S-EPMC9686994","cross_references":{"pubmed":["36178108"],"doi":["10.1242/dev.200621"]}}