ANHIR: Automatic Non-Rigid Histological Image Registration Challenge

被引：81

作者：

Borovec, Jiri ^{[1
]}

Kybic, Jan ^{[1
]}

Arganda-Carreras, Ignacio ^{[2
,3
,4
]}

Sorokin, Dmitry V. ^{[5
]}

Bueno, Gloria ^{[6
,7
]}

Khvostikov, Alexander V. ^{[5
]}

Bakas, Spyridon ^{[8
]}

Chang, Eric I-Chao ^{[9
]}

Heldmann, Stefan ^{[10
]}

Kartasalo, Kimmo ^{[11
]}

Latonen, Leena ^{[12
]}

Lotz, Johannes ^{[10
]}

Noga, Michelle ^{[13
,14
]}

Pati, Sarthak ^{[8
]}

Punithakumar, Kumaradevan ^{[13
,14
]}

Ruusuvuori, Pekka ^{[11
,15
]}

Skalski, Andrzej ^{[16
]}

Tahmasebi, Nazanin ^{[13
,14
]}

Valkonen, Masi ^{[11
]}

Venet, Ludovic ^{[8
]}

Wang, Yizhe ^{[17
]}

Weiss, Nick ^{[10
]}

Wodzinski, Marek ^{[16
]}

Xiang, Yu ^{[17
]}

Xu, Yan ^{[18
,19
,20
]}

Yan, Yan ^{[21
]}

Yushkevich, Paul ^{[8
]}

Zhao, Shengyu ^{[22
]}

Munoz-Barrutia, Arrate ^{[23
,24
]}

机构：

[1] Czech Tech Univ, Fac Elect Engn, Prague, Czech Republic

[2] Basque Fdn Sci, Ikerbasque, Bilbao, Spain

[3] Univ Basque Country, Dept Comp Sci & Artificial Intelligence, Donostia San Sebastian, Spain

[4] Donostia Int Phys Ctr, Donostia San Sebastian, Spain

[5] Lomonosov Moscow State Univ, Fac Computat Math & Cybernet, Lab Math Methods Image Proc, Moscow, Russia

[6] Univ Castilla La Mancha, VISILAB Grp, Ciudad Real, Spain

[7] Univ Castilla La Mancha, ETS Ingenieros Ind, Ciudad Real, Spain

[8] Univ Penn, Perelman Sch Med, Philadelphia, PA 19104 USA

[9] Microsoft Res, Beijing, Peoples R China

[10] Fraunhofer MEVIS, Lubeck, Germany

[11] Tampere Univ, Fac Med & Hlth Technol, Tampere, Finland

[12] Univ Eastern Finland, Inst Biomed, Kuopio, Finland

[13] Univ Alberta, Dept Radiol & Diagnost Imaging, Edmonton, AB, Canada

[14] Mazankowski Alberta Heart Inst, Servier Virtual Cardiac Ctr, Edmonton, AB, Canada

[15] Univ Turku, Inst Biomed, Turku, Finland

[16] AGH Univ Sci & Technol, Dept Measurement & Elect, Krakow, Poland

[17] Chengdu Knowledge Vis Sci & Technol Co Ltd, Chengdu, Peoples R China

[18] Microsoft Res, Beijing, Peoples R China

[19] Beihang Univ, Sch Biol Sci & Med Engn, Beijing, Peoples R China

[20] Beihang Univ, Beijing Adv Innovat Ctr Biomed Engn, Beijing, Peoples R China

[21] Univ Elect Sci & Technol China, Chengdu, Peoples R China

[22] Tsinghua Univ, Inst Interdisciplinary Informat Sci, Beijing, Peoples R China

[23] Univ Carlos III Madrid, Dept Bioingn & Ingn Aeroesp, Leganes, Spain

[24] Inst Invest Sanitaria Gregorio Maranon, Madrid, Spain

来源：

IEEE TRANSACTIONS ON MEDICAL IMAGING | 2020年 / 39卷 / 10期

基金：

美国国家卫生研究院; 芬兰科学院; 俄罗斯科学基金会;

关键词：

Image registration; Microscopy; Lung; Magnetic resonance imaging; IEEE Senior Members; Robustness; Image resolution; microscopy; VOLUME RECONSTRUCTION; ELASTIC REGISTRATION; FRAMEWORK; ALGORITHMS; ACCURACY;

D O I：

10.1109/TMI.2020.2986331

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

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 independent manner. 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 non-rigid 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 methods could successfully register over 98% of all landmarks and their mean landmark registration accuracy (TRE) was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.

引用

页码：3042 / 3052

页数：11

共 74 条

[1]

Arganda-Carreras I, 2006, LECT NOTES COMPUT SC, V4241, P85

[2] Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain [J].

Avants, B. B. ;

Epstein, C. L. ;

Grossman, M. ;

Gee, J. C. .

MEDICAL IMAGE ANALYSIS, 2008, 12 (01) :26-41

[3]

Bakas S., 2018, ARXIV181102629

[4]

Borovec J., 2019, BIRL BENCHMARK IMAGE

[5]

Borovec J, 2018, IEEE IMAGE PROC, P3368, DOI 10.1109/ICIP.2018.8451040

[6] Binary Pattern Dictionary Learning for Gene Expression Representation in Drosophila Imaginal Discs [J].

Borovec, Jiri ;

Kybic, Jan .

COMPUTER VISION - ACCV 2016 WORKSHOPS, PT II, 2017, 10117 :555-569

[7]

Borovec J, 2013, I S BIOMED IMAGING, P1034

[8] RESULTS OF A MULTI-INSTITUTION DEFORMABLE REGISTRATION ACCURACY STUDY (MIDRAS) [J].

Brock, Kristy K. .

INTERNATIONAL JOURNAL OF RADIATION ONCOLOGY BIOLOGY PHYSICS, 2010, 76 (02) :583-596

[9] A SURVEY OF IMAGE REGISTRATION TECHNIQUES [J].

BROWN, LG .

COMPUTING SURVEYS, 1992, 24 (04) :325-376

[10] A framework for evaluation of deformable image registration spatial accuracy using large landmark point sets [J].

Castillo, Richard ;

Castillo, Edward ;

Guerra, Rudy ;

Johnson, Valen E. ;

McPhail, Travis ;

Garg, Amit K. ;

Guerrero, Thomas .

PHYSICS IN MEDICINE AND BIOLOGY, 2009, 54 (07) :1849-1870

← 1 2 3 4 5 6 7 8 →