ANHIR: Automatic Non-Rigid Histological Image Registration Challenge

Jiří Borovec; Jan Kybic; Ignacio Arganda-Carreras; Dmitry V. Sorokin; Gloria Bueno; Alexander V. Khvostikov; Spyridon Bakas; Eric I.Chao Chang; Stefan Heldmann; Kimmo Kartasalo; Leena Latonen; Johannes Lotz; Michelle Noga; Sarthak Pati; Kumaradevan Punithakumar; Pekka Ruusuvuori; Andrzej Skalski; Nazanin Tahmasebi; Masi Valkonen; Ludovic Venet; Yizhe Wang; Nick Weiss; Marek Wodzinski; Yu Xiang; Yan Xu; Yan Yan; Paul Yushkevich; Shengyu Zhao; Arrate Munõz-Barrutia

doi:10.1109/TMI.2020.2986331

ANHIR: Automatic Non-Rigid Histological Image Registration Challenge

Jiří Borovec, Jan Kybic, Ignacio Arganda-Carreras, Dmitry V. Sorokin, Gloria Bueno, Alexander V. Khvostikov, Spyridon Bakas, Eric I.Chao Chang, Stefan Heldmann, Kimmo Kartasalo, Leena Latonen, Johannes Lotz, Michelle Noga, Sarthak Pati, Kumaradevan Punithakumar, Pekka Ruusuvuori, Andrzej Skalski, Nazanin Tahmasebi, Masi Valkonen, Ludovic VenetYizhe Wang, Nick Weiss, Marek Wodzinski, Yu Xiang, Yan Xu, Yan Yan, Paul Yushkevich, Shengyu Zhao, Arrate Munõz-Barrutia

Research output: Contribution to journal › Article › Scientific › peer-review

94 Citations (Scopus)

Abstract

Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds of microscopy histology images in a fair and independentmanner. We have assembled 8 datasets, containing 355 images with 18 different stains, resulting in 481 image pairs to be registered. Registration accuracy was evaluated using manually placed landmarks. In total, 256 teams registered for the challenge, 10 submitted the results, and 6 participated in the workshop. Here, we present the results of 7 well-performing methods from the challenge together with 6 well-known existing methods. The best methods used coarse but robust initial alignment, followed by nonrigid registration, used multiresolution, and were carefully tuned for the data at hand. They outperformed off-the-shelf methods, mostly by being more robust. The best methodscould successfully register over 98% of all landmarks and theirmean landmark registration accuracy (TRE)was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.

Original language	English
Pages (from-to)	3042-3052
Number of pages	11
Journal	IEEE Transactions on Medical Imaging
Volume	39
Issue number	10
DOIs	https://doi.org/10.1109/TMI.2020.2986331
Publication status	Published - 2020
Publication type	A1 Journal article-refereed

Funding

Manuscript received November 22, 2019; revised March 21, 2020; accepted April 2, 2020. Date of publication April 7, 2020; date of current version September 30, 2020. The work of Jan Kybic was supported in part by the OP VVV Funded under Project CZ.02.1.01/0.0/0.0/16_ 019/0000765 and in part by the Research Center for Informatics and the Czech Science Foundation under Project 17-15361S. The work of Dmitry V. Sorokin and Alexander V. Khvostikov was supported by the Russian Science Foundation under Grant 17-11-01279. The work of Spyridon Bakas, Sarthak Pati, Ludovic Venet, and Paul Yushkevich was supported in part by the National Institutes of Health under Grant NIH/NCI/ITCR:U24-CA189523, Grant NIH/NIBIB:R01EB017255, Grant NIH/NIA:R01AG056014, and Grant NIH/NIA:P30AG010124. The work of Pekka Ruusuvuori was supported by the Academy of Finland under Project 313921 and Project 314558. The work of Michelle Noga, Kuma-radevan Punithakumar, and Nazanin Tahmasebi was supported by the WestGrid and Compute Canada. The work of Marek Wodzinski was supported by the Preludium Project funded by the National Science Centre in Poland under Grant UMO-2018/29/N/ST6/00143. The work of Arrate Muñoz-Barrutia was supported in part by the Spanish Ministry of Economy and Competitiveness under Grant TEC2015-73064-EXP, Grant TEC2016-78052-R, and Grant RTC-2017-6600-1 and in part by the 2017 Leonardo Grant for Researchers and Cultural Creators, BBVA Foundation. (Corresponding author: Jan Kybic.) Please see the Acknowledgment section of this article for the author affiliations.

Keywords

Image registration
Microscopy

Publication forum classification

Publication forum level 2

ASJC Scopus subject areas

Software
Radiological and Ultrasound Technology
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/TMI.2020.2986331

https://europepmc.org/articles/pmc7584382?pdf=renderLicence: Unspecified

Cite this

Borovec, J., Kybic, J., Arganda-Carreras, I., Sorokin, D. V., Bueno, G., Khvostikov, A. V., Bakas, S., Chang, E. I. C., Heldmann, S., Kartasalo, K., Latonen, L., Lotz, J., Noga, M., Pati, S., Punithakumar, K., Ruusuvuori, P., Skalski, A., Tahmasebi, N., Valkonen, M., ... Munõz-Barrutia, A. (2020). ANHIR: Automatic Non-Rigid Histological Image Registration Challenge. IEEE Transactions on Medical Imaging, 39(10), 3042-3052. https://doi.org/10.1109/TMI.2020.2986331

@article{54dc1a239e844b31967a15cf5d8159a9,

title = "ANHIR: Automatic Non-Rigid Histological Image Registration Challenge",

abstract = "Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds of microscopy histology images in a fair and independentmanner. We have assembled 8 datasets, containing 355 images with 18 different stains, resulting in 481 image pairs to be registered. Registration accuracy was evaluated using manually placed landmarks. In total, 256 teams registered for the challenge, 10 submitted the results, and 6 participated in the workshop. Here, we present the results of 7 well-performing methods from the challenge together with 6 well-known existing methods. The best methods used coarse but robust initial alignment, followed by nonrigid registration, used multiresolution, and were carefully tuned for the data at hand. They outperformed off-the-shelf methods, mostly by being more robust. The best methodscould successfully register over 98% of all landmarks and theirmean landmark registration accuracy (TRE)was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.",

keywords = "Image registration, Microscopy",

author = "Ji{\v r}{\'i} Borovec and Jan Kybic and Ignacio Arganda-Carreras and Sorokin, {Dmitry V.} and Gloria Bueno and Khvostikov, {Alexander V.} and Spyridon Bakas and Chang, {Eric I.Chao} and Stefan Heldmann and Kimmo Kartasalo and Leena Latonen and Johannes Lotz and Michelle Noga and Sarthak Pati and Kumaradevan Punithakumar and Pekka Ruusuvuori and Andrzej Skalski and Nazanin Tahmasebi and Masi Valkonen and Ludovic Venet and Yizhe Wang and Nick Weiss and Marek Wodzinski and Yu Xiang and Yan Xu and Yan Yan and Paul Yushkevich and Shengyu Zhao and Arrate Mun{\~o}z-Barrutia",

note = "Funding Information: Manuscript received November 22, 2019; revised March 21, 2020; accepted April 2, 2020. Date of publication April 7, 2020; date of current version September 30, 2020. The work of Jan Kybic was supported in part by the OP VVV Funded under Project CZ.02.1.01/0.0/0.0/16_ 019/0000765 and in part by the Research Center for Informatics and the Czech Science Foundation under Project 17-15361S. The work of Dmitry V. Sorokin and Alexander V. Khvostikov was supported by the Russian Science Foundation under Grant 17-11-01279. The work of Spyridon Bakas, Sarthak Pati, Ludovic Venet, and Paul Yushkevich was supported in part by the National Institutes of Health under Grant NIH/NCI/ITCR:U24-CA189523, Grant NIH/NIBIB:R01EB017255, Grant NIH/NIA:R01AG056014, and Grant NIH/NIA:P30AG010124. The work of Pekka Ruusuvuori was supported by the Academy of Finland under Project 313921 and Project 314558. The work of Michelle Noga, Kuma-radevan Punithakumar, and Nazanin Tahmasebi was supported by the WestGrid and Compute Canada. The work of Marek Wodzinski was supported by the Preludium Project funded by the National Science Centre in Poland under Grant UMO-2018/29/N/ST6/00143. The work of Arrate Mu{\~n}oz-Barrutia was supported in part by the Spanish Ministry of Economy and Competitiveness under Grant TEC2015-73064-EXP, Grant TEC2016-78052-R, and Grant RTC-2017-6600-1 and in part by the 2017 Leonardo Grant for Researchers and Cultural Creators, BBVA Foundation. (Corresponding author: Jan Kybic.) Please see the Acknowledgment section of this article for the author affiliations. Publisher Copyright: {\textcopyright} 2020 IEEE.",

year = "2020",

doi = "10.1109/TMI.2020.2986331",

language = "English",

volume = "39",

pages = "3042--3052",

journal = "IEEE Transactions on Medical Imaging",

issn = "0278-0062",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "10",

}

Borovec, J, Kybic, J, Arganda-Carreras, I, Sorokin, DV, Bueno, G, Khvostikov, AV, Bakas, S, Chang, EIC, Heldmann, S, Kartasalo, K, Latonen, L, Lotz, J, Noga, M, Pati, S, Punithakumar, K, Ruusuvuori, P, Skalski, A, Tahmasebi, N, Valkonen, M, Venet, L, Wang, Y, Weiss, N, Wodzinski, M, Xiang, Y, Xu, Y, Yan, Y, Yushkevich, P, Zhao, S & Munõz-Barrutia, A 2020, 'ANHIR: Automatic Non-Rigid Histological Image Registration Challenge', IEEE Transactions on Medical Imaging, vol. 39, no. 10, pp. 3042-3052. https://doi.org/10.1109/TMI.2020.2986331

TY - JOUR

T1 - ANHIR

T2 - Automatic Non-Rigid Histological Image Registration Challenge

AU - Borovec, Jiří

AU - Kybic, Jan

AU - Arganda-Carreras, Ignacio

AU - Sorokin, Dmitry V.

AU - Bueno, Gloria

AU - Khvostikov, Alexander V.

AU - Bakas, Spyridon

AU - Chang, Eric I.Chao

AU - Heldmann, Stefan

AU - Kartasalo, Kimmo

AU - Latonen, Leena

AU - Lotz, Johannes

AU - Noga, Michelle

AU - Pati, Sarthak

AU - Punithakumar, Kumaradevan

AU - Ruusuvuori, Pekka

AU - Skalski, Andrzej

AU - Tahmasebi, Nazanin

AU - Valkonen, Masi

AU - Venet, Ludovic

AU - Wang, Yizhe

AU - Weiss, Nick

AU - Wodzinski, Marek

AU - Xiang, Yu

AU - Xu, Yan

AU - Yan, Yan

AU - Yushkevich, Paul

AU - Zhao, Shengyu

AU - Munõz-Barrutia, Arrate

N1 - Funding Information: Manuscript received November 22, 2019; revised March 21, 2020; accepted April 2, 2020. Date of publication April 7, 2020; date of current version September 30, 2020. The work of Jan Kybic was supported in part by the OP VVV Funded under Project CZ.02.1.01/0.0/0.0/16_ 019/0000765 and in part by the Research Center for Informatics and the Czech Science Foundation under Project 17-15361S. The work of Dmitry V. Sorokin and Alexander V. Khvostikov was supported by the Russian Science Foundation under Grant 17-11-01279. The work of Spyridon Bakas, Sarthak Pati, Ludovic Venet, and Paul Yushkevich was supported in part by the National Institutes of Health under Grant NIH/NCI/ITCR:U24-CA189523, Grant NIH/NIBIB:R01EB017255, Grant NIH/NIA:R01AG056014, and Grant NIH/NIA:P30AG010124. The work of Pekka Ruusuvuori was supported by the Academy of Finland under Project 313921 and Project 314558. The work of Michelle Noga, Kuma-radevan Punithakumar, and Nazanin Tahmasebi was supported by the WestGrid and Compute Canada. The work of Marek Wodzinski was supported by the Preludium Project funded by the National Science Centre in Poland under Grant UMO-2018/29/N/ST6/00143. The work of Arrate Muñoz-Barrutia was supported in part by the Spanish Ministry of Economy and Competitiveness under Grant TEC2015-73064-EXP, Grant TEC2016-78052-R, and Grant RTC-2017-6600-1 and in part by the 2017 Leonardo Grant for Researchers and Cultural Creators, BBVA Foundation. (Corresponding author: Jan Kybic.) Please see the Acknowledgment section of this article for the author affiliations. Publisher Copyright: © 2020 IEEE.

PY - 2020

Y1 - 2020

N2 - Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds of microscopy histology images in a fair and independentmanner. We have assembled 8 datasets, containing 355 images with 18 different stains, resulting in 481 image pairs to be registered. Registration accuracy was evaluated using manually placed landmarks. In total, 256 teams registered for the challenge, 10 submitted the results, and 6 participated in the workshop. Here, we present the results of 7 well-performing methods from the challenge together with 6 well-known existing methods. The best methods used coarse but robust initial alignment, followed by nonrigid registration, used multiresolution, and were carefully tuned for the data at hand. They outperformed off-the-shelf methods, mostly by being more robust. The best methodscould successfully register over 98% of all landmarks and theirmean landmark registration accuracy (TRE)was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.

AB - Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds of microscopy histology images in a fair and independentmanner. We have assembled 8 datasets, containing 355 images with 18 different stains, resulting in 481 image pairs to be registered. Registration accuracy was evaluated using manually placed landmarks. In total, 256 teams registered for the challenge, 10 submitted the results, and 6 participated in the workshop. Here, we present the results of 7 well-performing methods from the challenge together with 6 well-known existing methods. The best methods used coarse but robust initial alignment, followed by nonrigid registration, used multiresolution, and were carefully tuned for the data at hand. They outperformed off-the-shelf methods, mostly by being more robust. The best methodscould successfully register over 98% of all landmarks and theirmean landmark registration accuracy (TRE)was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.

KW - Image registration

KW - Microscopy

U2 - 10.1109/TMI.2020.2986331

DO - 10.1109/TMI.2020.2986331

M3 - Article

C2 - 32275587

AN - SCOPUS:85087014178

SN - 0278-0062

VL - 39

SP - 3042

EP - 3052

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

IS - 10

ER -

ANHIR: Automatic Non-Rigid Histological Image Registration Challenge

Abstract

Funding

Keywords

Publication forum classification

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this