guest ::
| Home > Publications database > Embarrassingly Simple Scribble Supervision for 3D Medical Segmentation > print |
| Home > Publications database > Embarrassingly Simple Scribble Supervision for 3D Medical Segmentation > print |
| 001 | 294877 | ||
| 005 | 20241210182915.0 | ||
| 024 | 7 | _ | |a 10.48550/ARXIV.2403.12834 |2 doi |
| 037 | _ | _ | |a DKFZ-2024-02587 |
| 100 | 1 | _ | |a Gotkowski, Karol |0 P:(DE-He78)47726a6f46da663d4584c1c7254a285a |b 0 |e First author |u dkfz |
| 245 | _ | _ | |a Embarrassingly Simple Scribble Supervision for 3D Medical Segmentation |
| 260 | _ | _ | |c 2024 |b arXiv |
| 336 | 7 | _ | |a Preprint |b preprint |m preprint |0 PUB:(DE-HGF)25 |s 1733825332_31422 |2 PUB:(DE-HGF) |
| 336 | 7 | _ | |a WORKING_PAPER |2 ORCID |
| 336 | 7 | _ | |a Electronic Article |0 28 |2 EndNote |
| 336 | 7 | _ | |a preprint |2 DRIVER |
| 336 | 7 | _ | |a ARTICLE |2 BibTeX |
| 336 | 7 | _ | |a Output Types/Working Paper |2 DataCite |
| 520 | _ | _ | |a Traditionally, segmentation algorithms require dense annotations for training, demanding significant annotation efforts, particularly within the 3D medical imaging field. Scribble-supervised learning emerges as a possible solution to this challenge, promising a reduction in annotation efforts when creating large-scale datasets. Recently, a plethora of methods for optimized learning from scribbles have been proposed, but have so far failed to position scribble annotation as a beneficial alternative. We relate this shortcoming to two major issues: 1) the complex nature of many methods which deeply ties them to the underlying segmentation model, thus preventing a migration to more powerful state-of-the-art models as the field progresses and 2) the lack of a systematic evaluation to validate consistent performance across the broader medical domain, resulting in a lack of trust when applying these methods to new segmentation problems. To address these issues, we propose a comprehensive scribble supervision benchmark consisting of seven datasets covering a diverse set of anatomies and pathologies imaged with varying modalities. We furthermore propose the systematic use of partial losses, i.e. losses that are only computed on annotated voxels. Contrary to most existing methods, these losses can be seamlessly integrated into state-of-the-art segmentation methods, enabling them to learn from scribble annotations while preserving their original loss formulations. Our evaluation using nnU-Net reveals that while most existing methods suffer from a lack of generalization, the proposed approach consistently delivers state-of-the-art performance. Thanks to its simplicity, our approach presents an embarrassingly simple yet effective solution to the challenges of scribble supervision. Source code as well as our extensive scribble benchmarking suite will be made publicly available upon publication. |
| 536 | _ | _ | |a 315 - Bildgebung und Radioonkologie (POF4-315) |0 G:(DE-HGF)POF4-315 |c POF4-315 |f POF IV |x 0 |
| 588 | _ | _ | |a Dataset connected to DataCite |
| 650 | _ | 7 | |a Computer Vision and Pattern Recognition (cs.CV) |2 Other |
| 650 | _ | 7 | |a FOS: Computer and information sciences |2 Other |
| 700 | 1 | _ | |a Lüth, Carsten |0 P:(DE-He78)6a78e3a44a8038881d941fb467eb4e19 |b 1 |u dkfz |
| 700 | 1 | _ | |a Jäger, Paul F. |0 P:(DE-He78)04a0b5a49db132d8f00cee326cb743ca |b 2 |u dkfz |
| 700 | 1 | _ | |a Ziegler, Sebastian |0 P:(DE-He78)ce72738384a651b46bb9b3f92dde8230 |b 3 |u dkfz |
| 700 | 1 | _ | |a Krämer, Lars |0 P:(DE-He78)9491eaed26df05482b258b0cac49375e |b 4 |u dkfz |
| 700 | 1 | _ | |a Denner, Stefan |0 P:(DE-He78)e35f6a9bd89b1c66d107df8a2325a758 |b 5 |u dkfz |
| 700 | 1 | _ | |a Xiao, Shuhan |0 P:(DE-He78)d2bf7126723ea8f6005ba141ea3c3e2c |b 6 |u dkfz |
| 700 | 1 | _ | |a Disch, Nico |0 P:(DE-He78)6aeec5fe74dfe54da185a85629cfbf3a |b 7 |u dkfz |
| 700 | 1 | _ | |a Maier-Hein, Klaus H. |0 P:(DE-He78)33c74005e1ce56f7025c4f6be15321b3 |b 8 |u dkfz |
| 700 | 1 | _ | |a Isensee, Fabian |0 P:(DE-He78)7ea9af59d03ec7deb982a0e0562358fa |b 9 |e Last author |u dkfz |
| 773 | _ | _ | |a 10.48550/ARXIV.2403.12834 |
| 909 | C | O | |o oai:inrepo02.dkfz.de:294877 |p VDB |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 0 |6 P:(DE-He78)47726a6f46da663d4584c1c7254a285a |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 1 |6 P:(DE-He78)6a78e3a44a8038881d941fb467eb4e19 |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 2 |6 P:(DE-He78)04a0b5a49db132d8f00cee326cb743ca |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 3 |6 P:(DE-He78)ce72738384a651b46bb9b3f92dde8230 |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 4 |6 P:(DE-He78)9491eaed26df05482b258b0cac49375e |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 5 |6 P:(DE-He78)e35f6a9bd89b1c66d107df8a2325a758 |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 6 |6 P:(DE-He78)d2bf7126723ea8f6005ba141ea3c3e2c |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 7 |6 P:(DE-He78)6aeec5fe74dfe54da185a85629cfbf3a |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 8 |6 P:(DE-He78)33c74005e1ce56f7025c4f6be15321b3 |
| 910 | 1 | _ | |a Deutsches Krebsforschungszentrum |0 I:(DE-588b)2036810-0 |k DKFZ |b 9 |6 P:(DE-He78)7ea9af59d03ec7deb982a0e0562358fa |
| 913 | 1 | _ | |a DE-HGF |b Gesundheit |l Krebsforschung |1 G:(DE-HGF)POF4-310 |0 G:(DE-HGF)POF4-315 |3 G:(DE-HGF)POF4 |2 G:(DE-HGF)POF4-300 |4 G:(DE-HGF)POF |v Bildgebung und Radioonkologie |x 0 |
| 914 | 1 | _ | |y 2024 |
| 920 | 1 | _ | |0 I:(DE-He78)E230-20160331 |k E230 |l E230 Medizinische Bildverarbeitung |x 0 |
| 920 | 1 | _ | |0 I:(DE-He78)E290-20160331 |k E290 |l NWG Interaktives maschinelles Lernen |x 1 |
| 980 | _ | _ | |a preprint |
| 980 | _ | _ | |a VDB |
| 980 | _ | _ | |a I:(DE-He78)E230-20160331 |
| 980 | _ | _ | |a I:(DE-He78)E290-20160331 |
| 980 | _ | _ | |a UNRESTRICTED |
| Library | Collection | CLSMajor | CLSMinor | Language | Author |
|---|