Abstract: In Defence of mathematical models for deep learning based registration

Lasse Hansen; Maximilian Blendowski; Mattias P. Heinrich

doi:10.1007/978-3-658-29267-6_6

Abstract: In Defence of mathematical models for deep learning based registration

Lasse Hansen^*, Maximilian Blendowski, Mattias P. Heinrich

^*Corresponding author for this work

Institute of Medical Informatics

Abstract

Deep learning based methods have not reached clinically acceptable results for common medical registration tasks that could be adequately solved using conventional methods. The slower progress compared to image segmentation is due to the lower availability of expert correspondences and the very large learnable parameter space for naive deep learning solutions. We strongly believe that it is necessary and beneficial to integrate conventional optimisation strategies as differentiable modules into deep learning based registration.

Original language	English
Title of host publication	Bildverarbeitung für die Medizin 2020
Editors	Thomas Tolxdorff, Thomas M. Deserno, Heinz Handels, Andreas Maier, Klaus H. Maier-Hein, Christoph Palm
Number of pages	1
Publisher	Springer Vieweg, Wiesbaden
Publication date	12.02.2020
Pages	32-32
ISBN (Print)	978-3-658-29266-9
ISBN (Electronic)	978-3-658-29267-6
DOIs	https://doi.org/10.1007/978-3-658-29267-6_6
Publication status	Published - 12.02.2020
Event	Bildverarbeitung für die Medizin 2020 - International workshop on Algorithmen - Systeme - Anwendungen - Berlin, Germany Duration: 15.03.2020 → 17.03.2020 Conference number: 237969

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.1007/978-3-658-29267-6_6

Cite this

Hansen, L., Blendowski, M., & Heinrich, M. P. (2020). Abstract: In Defence of mathematical models for deep learning based registration. In T. Tolxdorff, T. M. Deserno, H. Handels, A. Maier, K. H. Maier-Hein, & C. Palm (Eds.), Bildverarbeitung für die Medizin 2020 (pp. 32-32). (Informatik aktuell). Springer Vieweg, Wiesbaden. https://doi.org/10.1007/978-3-658-29267-6_6

Hansen, Lasse ; Blendowski, Maximilian ; Heinrich, Mattias P. / Abstract: In Defence of mathematical models for deep learning based registration. Bildverarbeitung für die Medizin 2020. editor / Thomas Tolxdorff ; Thomas M. Deserno ; Heinz Handels ; Andreas Maier ; Klaus H. Maier-Hein ; Christoph Palm. Springer Vieweg, Wiesbaden, 2020. pp. 32-32 (Informatik aktuell).

@inproceedings{6044002b32fd4f7c864be086bf067d77,

title = "Abstract: In Defence of mathematical models for deep learning based registration",

abstract = "Deep learning based methods have not reached clinically acceptable results for common medical registration tasks that could be adequately solved using conventional methods. The slower progress compared to image segmentation is due to the lower availability of expert correspondences and the very large learnable parameter space for naive deep learning solutions. We strongly believe that it is necessary and beneficial to integrate conventional optimisation strategies as differentiable modules into deep learning based registration.",

author = "Lasse Hansen and Maximilian Blendowski and Heinrich, \{Mattias P.\}",

note = "Publisher Copyright: {\textcopyright} Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2020. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; Bildverarbeitung f{\"u}r die Medizin 2020 - International workshop on Algorithmen - Systeme - Anwendungen<br/> ; Conference date: 15-03-2020 Through 17-03-2020",

year = "2020",

month = feb,

day = "12",

doi = "10.1007/978-3-658-29267-6\_6",

language = "English",

isbn = "978-3-658-29266-9",

series = "Informatik aktuell",

publisher = "Springer Vieweg, Wiesbaden",

pages = "32--32",

editor = "Tolxdorff, \{Thomas \} and \{M. Deserno\}, \{Thomas \} and Handels, \{Heinz \} and Maier, \{Andreas \} and \{H. Maier-Hein\}, \{Klaus \} and Palm, \{Christoph \}",

booktitle = "Bildverarbeitung f{\"u}r die Medizin 2020",

}

Hansen, L, Blendowski, M & Heinrich, MP 2020, Abstract: In Defence of mathematical models for deep learning based registration. in T Tolxdorff, T M. Deserno, H Handels, A Maier, K H. Maier-Hein & C Palm (eds), Bildverarbeitung für die Medizin 2020. Informatik aktuell, Springer Vieweg, Wiesbaden, pp. 32-32, Bildverarbeitung für die Medizin 2020 - International workshop on Algorithmen - Systeme - Anwendungen
, Berlin, Berlin, Germany, 15.03.20. https://doi.org/10.1007/978-3-658-29267-6_6

Abstract: In Defence of mathematical models for deep learning based registration. / Hansen, Lasse; Blendowski, Maximilian; Heinrich, Mattias P.
Bildverarbeitung für die Medizin 2020. ed. / Thomas Tolxdorff; Thomas M. Deserno; Heinz Handels; Andreas Maier; Klaus H. Maier-Hein; Christoph Palm. Springer Vieweg, Wiesbaden, 2020. p. 32-32 (Informatik aktuell).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - Abstract: In Defence of mathematical models for deep learning based registration

AU - Hansen, Lasse

AU - Blendowski, Maximilian

AU - Heinrich, Mattias P.

N1 - Conference code: 237969

PY - 2020/2/12

Y1 - 2020/2/12

N2 - Deep learning based methods have not reached clinically acceptable results for common medical registration tasks that could be adequately solved using conventional methods. The slower progress compared to image segmentation is due to the lower availability of expert correspondences and the very large learnable parameter space for naive deep learning solutions. We strongly believe that it is necessary and beneficial to integrate conventional optimisation strategies as differentiable modules into deep learning based registration.

AB - Deep learning based methods have not reached clinically acceptable results for common medical registration tasks that could be adequately solved using conventional methods. The slower progress compared to image segmentation is due to the lower availability of expert correspondences and the very large learnable parameter space for naive deep learning solutions. We strongly believe that it is necessary and beneficial to integrate conventional optimisation strategies as differentiable modules into deep learning based registration.

UR - https://www.scopus.com/pages/publications/85083074551

U2 - 10.1007/978-3-658-29267-6_6

DO - 10.1007/978-3-658-29267-6_6

M3 - Conference contribution

SN - 978-3-658-29266-9

T3 - Informatik aktuell

SP - 32

EP - 32

BT - Bildverarbeitung für die Medizin 2020

A2 - Tolxdorff, Thomas

A2 - M. Deserno, Thomas

A2 - Handels, Heinz

A2 - Maier, Andreas

A2 - H. Maier-Hein, Klaus

A2 - Palm, Christoph

PB - Springer Vieweg, Wiesbaden

T2 - Bildverarbeitung für die Medizin 2020 - International workshop on Algorithmen - Systeme - Anwendungen<br/>

Y2 - 15 March 2020 through 17 March 2020

ER -

Abstract: In Defence of mathematical models for deep learning based registration

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this