CoNIC Challenge: Pushing the frontiers of nuclear detection, segmentation, classification and counting
Simon Graham
(1)
,
Quoc Dang Vu
(1)
,
Mostafa Jahanifar
(1)
,
Martin Weigert
(2)
,
Uwe Schmidt
(3)
,
Wenhua Zhang
(4)
,
Jun Zhang
(5)
,
Sen Yang
(6)
,
Jinxi Xiang
(7)
,
Xiyue Wang
(6)
,
Josef Lorenz Rumberger
(8, 3, 9)
,
Elias Baumann
(10)
,
Peter Hirsch
(8, 3)
,
Lihao Liu
(11)
,
Chenyang Hong
(12)
,
Angelica Aviles-Rivero
(11)
,
Ayushi Jain
,
Heeyoung Ahn
,
Yiyu Hong
,
Hussam Azzuni
(13)
,
Min Xu
(13)
,
Mohammad Yaqub
(13)
,
Marie-Claire Blache
(14)
,
Benoît Piégu
(14)
,
Bertrand Vernay
(15)
,
Tim Scherr
(16)
,
Moritz Böhland
(16)
,
Katharina Löffler
(16)
,
Jiachen Li
(17)
,
Weiqin Ying
(17)
,
Chixin Wang
(17)
,
David Snead
(18, 1)
,
Shan Raza
(1)
,
Fayyaz Minhas
(1)
,
Nasir Rajpoot
(1, 18)
,
. The Conic Challenge Consortium
1
University of Warwick [Coventry]
2 EPFL - Ecole Polytechnique Fédérale de Lausanne
3 HU Berlin - Humboldt-Universität zu Berlin = Humboldt University of Berlin = Université Humboldt de Berlin
4 HKU - The University of Hong Kong
5 Tencent AI Lab
6 SCU - Sichuan University [Chengdu]
7 THU - Tsinghua University [Beijing]
8 Helmholtz-Gemeinschaft = Helmholtz Association
9 Charité - UniversitätsMedizin = Charité - University Hospital [Berlin]
10 UNIBE - Universität Bern / University of Bern
11 CAM - University of Cambridge [UK]
12 CUHK - The Chinese University of Hong Kong [Hong Kong]
13 MBZUAI - Mohamed bin Zayed University of Artificial Intelligence
14 PRC - Physiologie de la reproduction et des comportements [Nouzilly]
15 IGBMC - Institut de Génétique et de Biologie Moléculaire et Cellulaire
16 KIT - Karlsruhe Institute of Technology
17 SCUT - South China University of Technology [Guangzhou]
18 UHCW - University Hospital Coventry Warwickshire
2 EPFL - Ecole Polytechnique Fédérale de Lausanne
3 HU Berlin - Humboldt-Universität zu Berlin = Humboldt University of Berlin = Université Humboldt de Berlin
4 HKU - The University of Hong Kong
5 Tencent AI Lab
6 SCU - Sichuan University [Chengdu]
7 THU - Tsinghua University [Beijing]
8 Helmholtz-Gemeinschaft = Helmholtz Association
9 Charité - UniversitätsMedizin = Charité - University Hospital [Berlin]
10 UNIBE - Universität Bern / University of Bern
11 CAM - University of Cambridge [UK]
12 CUHK - The Chinese University of Hong Kong [Hong Kong]
13 MBZUAI - Mohamed bin Zayed University of Artificial Intelligence
14 PRC - Physiologie de la reproduction et des comportements [Nouzilly]
15 IGBMC - Institut de Génétique et de Biologie Moléculaire et Cellulaire
16 KIT - Karlsruhe Institute of Technology
17 SCUT - South China University of Technology [Guangzhou]
18 UHCW - University Hospital Coventry Warwickshire
Simon Graham
- Fonction : co premier-auteur
- PersonId : 1339121
- ORCID : 0000-0002-2214-8212
Martin Weigert
- Fonction : Auteur
- PersonId : 1339122
- ORCID : 0000-0002-7780-9057
Xiyue Wang
- Fonction : Auteur
- PersonId : 827799
- ORCID : 0000-0002-3597-9090
Lihao Liu
- Fonction : Auteur
- PersonId : 1339124
- ORCID : 0000-0002-8983-2342
Ayushi Jain
- Fonction : Auteur
Heeyoung Ahn
- Fonction : Auteur
Yiyu Hong
- Fonction : Auteur
Hussam Azzuni
- Fonction : Auteur
- PersonId : 1339125
- ORCID : 0000-0001-7267-8919
Mohammad Yaqub
- Fonction : Auteur
- PersonId : 1339126
- ORCID : 0000-0001-6896-1105
Marie-Claire Blache
- Fonction : Auteur
- PersonId : 744126
- IdHAL : marie-claire-blache
- ORCID : 0000-0003-2748-7018
Benoît Piégu
- Fonction : Auteur
- PersonId : 744876
- IdHAL : benoit-piegu
- ORCID : 0000-0003-4384-1726
Bertrand Vernay
- Fonction : Auteur
- PersonId : 1339127
- ORCID : 0000-0002-7843-3872
Tim Scherr
- Fonction : Auteur
- PersonId : 1339128
- ORCID : 0000-0001-8755-2825
Moritz Böhland
- Fonction : Auteur
- PersonId : 1339129
- ORCID : 0000-0002-9321-8169
Katharina Löffler
- Fonction : Auteur
- PersonId : 1339130
- ORCID : 0000-0001-9769-3517
Shan Raza
- Fonction : Auteur
- PersonId : 1339131
- ORCID : 0000-0002-1097-1738
Nasir Rajpoot
- Fonction : Auteur
- PersonId : 1339132
- ORCID : 0000-0002-4706-1308
. The Conic Challenge Consortium
- Fonction : Auteur
Résumé
Nuclear detection, segmentation and morphometric profiling are essential in helping us further understand the relationship between histology and patient outcome. To drive innovation in this area, we setup a community-wide challenge using the largest available dataset of its kind to assess nuclear segmentation and cellular composition. Our challenge, named CoNIC, stimulated the development of reproducible algorithms for cellular recognition with real-time result inspection on public leaderboards. We conducted an extensive post-challenge analysis based on the top-performing models using 1,658 whole-slide images of colon tissue. With around 700 million detected nuclei per model, associated features were used for dysplasia grading and survival analysis, where we demonstrated that the challenge's improvement over the previous state-of-the-art led to significant boosts in downstream performance. Our findings also suggest that eosinophils and neutrophils play an important role in the tumour microevironment. We release challenge models and WSI-level results to foster the development of further methods for biomarker discovery.