{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/microscopy-cell-segmentation-via","title":"Microscopy Cell Segmentation via Convolutional LSTM Networks","arxiv_id":"1805.11247","date":"2018-05-29","proceeding":null,"authors":["Assaf Arbelle","Tammy Riklin Raviv"],"abstract":"Live cell microscopy sequences exhibit complex spatial structures and\ncomplicated temporal behaviour, making their analysis a challenging task.\nConsidering cell segmentation problem, which plays a significant role in the\nanalysis, the spatial properties of the data can be captured using\nConvolutional Neural Networks (CNNs). Recent approaches show promising\nsegmentation results using convolutional encoder-decoders such as the U-Net.\nNevertheless, these methods are limited by their inability to incorporate\ntemporal information, that can facilitate segmentation of individual touching\ncells or of cells that are partially visible. In order to exploit cell dynamics\nwe propose a novel segmentation architecture which integrates Convolutional\nLong Short Term Memory (C-LSTM) with the U-Net. The network's unique\narchitecture allows it to capture multi-scale, compact, spatio-temporal\nencoding in the C-LSTMs memory units. The method was evaluated on the Cell\nTracking Challenge and achieved state-of-the-art results (1st on Fluo-N2DH-SIM+\nand 2nd on DIC-C2DL-HeLa datasets) The code is freely available at:\nhttps://github.com/arbellea/LSTM-UNet.git","url_abs":"http://arxiv.org/abs/1805.11247v2","url_pdf":"http://arxiv.org/pdf/1805.11247v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"microscopy-cell-segmentation-via","repo_url":"https://github.com/arbellea/LSTM-UNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"microscopy-cell-segmentation-via","repo_url":"https://github.com/alvchn/fcn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"microscopy-cell-segmentation-via","repo_url":"https://github.com/ruveydayilmaz0/bvdm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"cell-segmentation","task_name":"Cell Segmentation"},{"task_slug":"cell-tracking","task_name":"Cell Tracking"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"u-net","method_name":"U-Net"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/cell-segmentation-on-dic-c2dh-hela","task":"Cell Segmentation","dataset":"DIC-C2DH-HeLa","model":"EncLSTM","rank_in_archive_order":1,"of":2,"metrics":{"SEG (~Mean IoU)":"0.793"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-dic-c2dh-hela","task":"Cell Segmentation","dataset":"DIC-C2DH-HeLa","model":"DecLSTM","rank_in_archive_order":2,"of":2,"metrics":{"SEG (~Mean IoU)":"0.511"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-fluo-n2dh-gowt1","task":"Cell Segmentation","dataset":"Fluo-N2DH-GOWT1","model":"DecLSTM","rank_in_archive_order":1,"of":2,"metrics":{"SEG (~Mean IoU)":"0.854"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-fluo-n2dh-gowt1","task":"Cell Segmentation","dataset":"Fluo-N2DH-GOWT1","model":"EncLSTM","rank_in_archive_order":2,"of":2,"metrics":{"SEG (~Mean IoU)":"0.85"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-fluo-n2dh-sim","task":"Cell Segmentation","dataset":"Fluo-N2DH-SIM+","model":"EncLSTM","rank_in_archive_order":1,"of":2,"metrics":{"SEG (~Mean IoU)":"0.811"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-fluo-n2dh-sim","task":"Cell Segmentation","dataset":"Fluo-N2DH-SIM+","model":"DecLSTM","rank_in_archive_order":2,"of":2,"metrics":{"SEG (~Mean IoU)":"0.802"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-fluo-n2dl-hela","task":"Cell Segmentation","dataset":"Fluo-N2DL-HeLa","model":"DecLSTM","rank_in_archive_order":2,"of":3,"metrics":{"SEG (~Mean IoU)":"0.839"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-fluo-n2dl-hela","task":"Cell Segmentation","dataset":"Fluo-N2DL-HeLa","model":"EncLSTM","rank_in_archive_order":3,"of":3,"metrics":{"SEG (~Mean IoU)":"0.811"},"uses_additional_data":false},{"leaderboard":"/sota/cell-segmentation-on-phc-c2dh-u373","task":"Cell Segmentation","dataset":"PhC-C2DH-U373","model":"EncLSTM","rank_in_archive_order":1,"of":1,"metrics":{"SEG (~Mean IoU)":"0.842"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.11247","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}