% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Wagner:282463,
author = {S. J. Wagner and D. Reisenbüchler and N. P. West and J. M.
Niehues and J. Zhu and S. Foersch and G. P. Veldhuizen and
P. Quirke and H. I. Grabsch and P. A. van den Brandt and G.
G. A. Hutchins and S. D. Richman and T. Yuan$^*$ and R.
Langer and J. C. A. Jenniskens and K. Offermans and W.
Mueller and R. Gray and S. B. Gruber and J. K. Greenson and
G. Rennert and J. D. Bonner and D. Schmolze and J.
Jonnagaddala and N. J. Hawkins and R. L. Ward and D. Morton
and M. Seymour and L. Magill and M. Nowak and J. Hay and V.
H. Koelzer and D. N. Church and C. Matek and C. Geppert and
C. Peng and C. Zhi and X. Ouyang and J. A. James and M. B.
Loughrey and M. Salto-Tellez and H. Brenner$^*$ and M.
Hoffmeister$^*$ and D. Truhn and J. A. Schnabel and M.
Boxberg and T. Peng and J. N. Kather},
collaboration = {T. consortium},
othercontributors = {D. Church and E. Domingo and J. Edwards and B. Glimelius
and I. Gogenur and A. Harkin and J. Hay and T. Iveson and E.
Jaeger and C. Kelly and R. Kerr and N. Maka and H. Morgan
and K. Oien and C. Orange and C. Palles and C. Roxburgh and
O. Sansom and M. Saunders and I. Tomlinson},
title = {{T}ransformer-based biomarker prediction from colorectal
cancer histology: {A} large-scale multicentric study.},
journal = {Cancer cell},
volume = {49},
number = {1},
issn = {1535-6108},
address = {New York, NY},
publisher = {Elsevier},
reportid = {DKFZ-2023-01778},
pages = {1650-1661.e4},
year = {2023},
note = {2023 Sep 11;41(9):1650-1661.e4},
abstract = {Deep learning (DL) can accelerate the prediction of
prognostic biomarkers from routine pathology slides in
colorectal cancer (CRC). However, current approaches rely on
convolutional neural networks (CNNs) and have mostly been
validated on small patient cohorts. Here, we develop a new
transformer-based pipeline for end-to-end biomarker
prediction from pathology slides by combining a pre-trained
transformer encoder with a transformer network for patch
aggregation. Our transformer-based approach substantially
improves the performance, generalizability, data efficiency,
and interpretability as compared with current
state-of-the-art algorithms. After training and evaluating
on a large multicenter cohort of over 13,000 patients from
16 colorectal cancer cohorts, we achieve a sensitivity of
0.99 with a negative predictive value of over 0.99 for
prediction of microsatellite instability (MSI) on surgical
resection specimens. We demonstrate that resection
specimen-only training reaches clinical-grade performance on
endoscopic biopsy tissue, solving a long-standing diagnostic
problem.},
keywords = {artificial intelligence (Other) / biomarker (Other) /
colorectal cancer (Other) / deep learning (Other) /
microsatellite instability (Other) / multiple instance
learning (Other) / transformer (Other)},
cin = {C070 / C120 / HD01},
ddc = {610},
cid = {I:(DE-He78)C070-20160331 / I:(DE-He78)C120-20160331 /
I:(DE-He78)HD01-20160331},
pnm = {313 - Krebsrisikofaktoren und Prävention (POF4-313)},
pid = {G:(DE-HGF)POF4-313},
typ = {PUB:(DE-HGF)16},
pubmed = {pmid:37652006},
doi = {10.1016/j.ccell.2023.08.002},
url = {https://inrepo02.dkfz.de/record/282463},
}