VerSe : A Vertebrae labelling and segmentation benchmark for multi-detector CT images
Copyright © 2021 Elsevier B.V. All rights reserved..
Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision support systems for diagnosis, surgery planning, and population-based analysis of spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms tackling the labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the results of this evaluation and further investigate the performance variation at the vertebra level, scan level, and different fields of view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top-performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The VerSe content and code can be accessed at: https://github.com/anjany/verse.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2021 |
---|---|
Erschienen: |
2021 |
Enthalten in: |
Zur Gesamtaufnahme - volume:73 |
---|---|
Enthalten in: |
Medical image analysis - 73(2021) vom: 01. Okt., Seite 102166 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Sekuboyina, Anjany [VerfasserIn] |
---|
Links: |
---|
Themen: |
Journal Article |
---|
Anmerkungen: |
Date Completed 23.09.2021 Date Revised 11.11.2022 published: Print-Electronic Citation Status MEDLINE |
---|
doi: |
10.1016/j.media.2021.102166 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM328846163 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM328846163 | ||
003 | DE-627 | ||
005 | 20231225203825.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231225s2021 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1016/j.media.2021.102166 |2 doi | |
028 | 5 | 2 | |a pubmed24n1096.xml |
035 | |a (DE-627)NLM328846163 | ||
035 | |a (NLM)34340104 | ||
035 | |a (PII)S1361-8415(21)00212-7 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Sekuboyina, Anjany |e verfasserin |4 aut | |
245 | 1 | 0 | |a VerSe |b A Vertebrae labelling and segmentation benchmark for multi-detector CT images |
264 | 1 | |c 2021 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 23.09.2021 | ||
500 | |a Date Revised 11.11.2022 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status MEDLINE | ||
520 | |a Copyright © 2021 Elsevier B.V. All rights reserved. | ||
520 | |a Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision support systems for diagnosis, surgery planning, and population-based analysis of spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms tackling the labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the results of this evaluation and further investigate the performance variation at the vertebra level, scan level, and different fields of view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top-performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The VerSe content and code can be accessed at: https://github.com/anjany/verse | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
650 | 4 | |a Labelling | |
650 | 4 | |a Segmentation | |
650 | 4 | |a Spine | |
650 | 4 | |a Vertebrae | |
700 | 1 | |a Husseini, Malek E |e verfasserin |4 aut | |
700 | 1 | |a Bayat, Amirhossein |e verfasserin |4 aut | |
700 | 1 | |a Löffler, Maximilian |e verfasserin |4 aut | |
700 | 1 | |a Liebl, Hans |e verfasserin |4 aut | |
700 | 1 | |a Li, Hongwei |e verfasserin |4 aut | |
700 | 1 | |a Tetteh, Giles |e verfasserin |4 aut | |
700 | 1 | |a Kukačka, Jan |e verfasserin |4 aut | |
700 | 1 | |a Payer, Christian |e verfasserin |4 aut | |
700 | 1 | |a Štern, Darko |e verfasserin |4 aut | |
700 | 1 | |a Urschler, Martin |e verfasserin |4 aut | |
700 | 1 | |a Chen, Maodong |e verfasserin |4 aut | |
700 | 1 | |a Cheng, Dalong |e verfasserin |4 aut | |
700 | 1 | |a Lessmann, Nikolas |e verfasserin |4 aut | |
700 | 1 | |a Hu, Yujin |e verfasserin |4 aut | |
700 | 1 | |a Wang, Tianfu |e verfasserin |4 aut | |
700 | 1 | |a Yang, Dong |e verfasserin |4 aut | |
700 | 1 | |a Xu, Daguang |e verfasserin |4 aut | |
700 | 1 | |a Ambellan, Felix |e verfasserin |4 aut | |
700 | 1 | |a Amiranashvili, Tamaz |e verfasserin |4 aut | |
700 | 1 | |a Ehlke, Moritz |e verfasserin |4 aut | |
700 | 1 | |a Lamecker, Hans |e verfasserin |4 aut | |
700 | 1 | |a Lehnert, Sebastian |e verfasserin |4 aut | |
700 | 1 | |a Lirio, Marilia |e verfasserin |4 aut | |
700 | 1 | |a Olaguer, Nicolás Pérez de |e verfasserin |4 aut | |
700 | 1 | |a Ramm, Heiko |e verfasserin |4 aut | |
700 | 1 | |a Sahu, Manish |e verfasserin |4 aut | |
700 | 1 | |a Tack, Alexander |e verfasserin |4 aut | |
700 | 1 | |a Zachow, Stefan |e verfasserin |4 aut | |
700 | 1 | |a Jiang, Tao |e verfasserin |4 aut | |
700 | 1 | |a Ma, Xinjun |e verfasserin |4 aut | |
700 | 1 | |a Angerman, Christoph |e verfasserin |4 aut | |
700 | 1 | |a Wang, Xin |e verfasserin |4 aut | |
700 | 1 | |a Brown, Kevin |e verfasserin |4 aut | |
700 | 1 | |a Kirszenberg, Alexandre |e verfasserin |4 aut | |
700 | 1 | |a Puybareau, Élodie |e verfasserin |4 aut | |
700 | 1 | |a Chen, Di |e verfasserin |4 aut | |
700 | 1 | |a Bai, Yiwei |e verfasserin |4 aut | |
700 | 1 | |a Rapazzo, Brandon H |e verfasserin |4 aut | |
700 | 1 | |a Yeah, Timyoas |e verfasserin |4 aut | |
700 | 1 | |a Zhang, Amber |e verfasserin |4 aut | |
700 | 1 | |a Xu, Shangliang |e verfasserin |4 aut | |
700 | 1 | |a Hou, Feng |e verfasserin |4 aut | |
700 | 1 | |a He, Zhiqiang |e verfasserin |4 aut | |
700 | 1 | |a Zeng, Chan |e verfasserin |4 aut | |
700 | 1 | |a Xiangshang, Zheng |e verfasserin |4 aut | |
700 | 1 | |a Liming, Xu |e verfasserin |4 aut | |
700 | 1 | |a Netherton, Tucker J |e verfasserin |4 aut | |
700 | 1 | |a Mumme, Raymond P |e verfasserin |4 aut | |
700 | 1 | |a Court, Laurence E |e verfasserin |4 aut | |
700 | 1 | |a Huang, Zixun |e verfasserin |4 aut | |
700 | 1 | |a He, Chenhang |e verfasserin |4 aut | |
700 | 1 | |a Wang, Li-Wen |e verfasserin |4 aut | |
700 | 1 | |a Ling, Sai Ho |e verfasserin |4 aut | |
700 | 1 | |a Huỳnh, Lê Duy |e verfasserin |4 aut | |
700 | 1 | |a Boutry, Nicolas |e verfasserin |4 aut | |
700 | 1 | |a Jakubicek, Roman |e verfasserin |4 aut | |
700 | 1 | |a Chmelik, Jiri |e verfasserin |4 aut | |
700 | 1 | |a Mulay, Supriti |e verfasserin |4 aut | |
700 | 1 | |a Sivaprakasam, Mohanasankar |e verfasserin |4 aut | |
700 | 1 | |a Paetzold, Johannes C |e verfasserin |4 aut | |
700 | 1 | |a Shit, Suprosanna |e verfasserin |4 aut | |
700 | 1 | |a Ezhov, Ivan |e verfasserin |4 aut | |
700 | 1 | |a Wiestler, Benedikt |e verfasserin |4 aut | |
700 | 1 | |a Glocker, Ben |e verfasserin |4 aut | |
700 | 1 | |a Valentinitsch, Alexander |e verfasserin |4 aut | |
700 | 1 | |a Rempfler, Markus |e verfasserin |4 aut | |
700 | 1 | |a Menze, Björn H |e verfasserin |4 aut | |
700 | 1 | |a Kirschke, Jan S |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Medical image analysis |d 1996 |g 73(2021) vom: 01. Okt., Seite 102166 |w (DE-627)NLM096527188 |x 1361-8423 |7 nnns |
773 | 1 | 8 | |g volume:73 |g year:2021 |g day:01 |g month:10 |g pages:102166 |
856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.media.2021.102166 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 73 |j 2021 |b 01 |c 10 |h 102166 |