Predicting stroke lesion and clinical outcome with random forests

Oskar Maier; Heinz Handels

doi:10.1007/978-3-319-55524-9_21

Predicting stroke lesion and clinical outcome with random forests

Oskar Maier^*, Heinz Handels

^*Corresponding author for this work

Institute of Medical Informatics

Abstract

The treatment of ischemic stroke requires fast decisions for which the potentially fatal risks of an intervention have to be weighted against the presumed benefits. Ideally, the treating physician could predict the outcome under different circumstances beforehand and thus make an informed treatment decision. To this end, this article presents two new methods: one for lesion outcome and one for clinical outcome prediction from multispectral magnetic resonance sequences. After extracting tailored image features, a random forest classifier respectively regressor is trained. Both approaches were submitted to the Ischemic Stroke Lesion Segmentation (ISLES) 2017 challenge and obtained a first and third place. The outcome underlines the robustness of our designed features and stresses the approach’s resilience against overfitting when faced with small training datasets.

Original language	English
Title of host publication	BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries
Editors	Alessandro Crimi, Bjoern Menze, Oskar Maier, Mauricio Reyes, Stefan Winzeck, Heinz Handels
Number of pages	12
Volume	10154 LNCS
Publisher	Springer, Cham
Publication date	12.04.2016
Pages	219-230
ISBN (Print)	978-3-319-55523-2
ISBN (Electronic)	978-3-319-55524-9
DOIs	https://doi.org/10.1007/978-3-319-55524-9_21
Publication status	Published - 12.04.2016
Event	19th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2016 - Athens, Greece Duration: 17.10.2016 → 21.10.2016 http://miccai2016.org/en/

UN SDGs

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

Access to Document

10.1007/978-3-319-55524-9_21

Cite this

Maier, O., & Handels, H. (2016). Predicting stroke lesion and clinical outcome with random forests. In A. Crimi, B. Menze, O. Maier, M. Reyes, S. Winzeck, & H. Handels (Eds.), BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries (Vol. 10154 LNCS, pp. 219-230). (Lecture Notes in Computer Science; Vol. 10154 LNCS). Springer, Cham. https://doi.org/10.1007/978-3-319-55524-9_21

Maier, Oskar ; Handels, Heinz. / Predicting stroke lesion and clinical outcome with random forests. BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. editor / Alessandro Crimi ; Bjoern Menze ; Oskar Maier ; Mauricio Reyes ; Stefan Winzeck ; Heinz Handels. Vol. 10154 LNCS Springer, Cham, 2016. pp. 219-230 (Lecture Notes in Computer Science).

@inproceedings{c6a93d2b5c23468b9dfcd18f9456c518,

title = "Predicting stroke lesion and clinical outcome with random forests",

abstract = "The treatment of ischemic stroke requires fast decisions for which the potentially fatal risks of an intervention have to be weighted against the presumed benefits. Ideally, the treating physician could predict the outcome under different circumstances beforehand and thus make an informed treatment decision. To this end, this article presents two new methods: one for lesion outcome and one for clinical outcome prediction from multispectral magnetic resonance sequences. After extracting tailored image features, a random forest classifier respectively regressor is trained. Both approaches were submitted to the Ischemic Stroke Lesion Segmentation (ISLES) 2017 challenge and obtained a first and third place. The outcome underlines the robustness of our designed features and stresses the approach{\textquoteright}s resilience against overfitting when faced with small training datasets.",

author = "Oskar Maier and Heinz Handels",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2016. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.; 19th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2016<br/> ; Conference date: 17-10-2016 Through 21-10-2016",

year = "2016",

month = apr,

day = "12",

doi = "10.1007/978-3-319-55524-9_21",

language = "English",

isbn = "978-3-319-55523-2",

volume = "10154 LNCS",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Cham",

pages = "219--230",

editor = "Crimi, {Alessandro } and Menze, {Bjoern } and Maier, {Oskar } and Reyes, {Mauricio } and Winzeck, {Stefan } and Handels, {Heinz }",

booktitle = "BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries",

url = "http://miccai2016.org/en/",

}

Maier, O & Handels, H 2016, Predicting stroke lesion and clinical outcome with random forests. in A Crimi, B Menze, O Maier, M Reyes, S Winzeck & H Handels (eds), BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. vol. 10154 LNCS, Lecture Notes in Computer Science, vol. 10154 LNCS, Springer, Cham, pp. 219-230, 19th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2016
, Athens, Greece, 17.10.16. https://doi.org/10.1007/978-3-319-55524-9_21

Predicting stroke lesion and clinical outcome with random forests. / Maier, Oskar; Handels, Heinz.
BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. ed. / Alessandro Crimi; Bjoern Menze; Oskar Maier; Mauricio Reyes; Stefan Winzeck; Heinz Handels. Vol. 10154 LNCS Springer, Cham, 2016. p. 219-230 (Lecture Notes in Computer Science; Vol. 10154 LNCS).

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

TY - GEN

T1 - Predicting stroke lesion and clinical outcome with random forests

AU - Maier, Oskar

AU - Handels, Heinz

PY - 2016/4/12

Y1 - 2016/4/12

N2 - The treatment of ischemic stroke requires fast decisions for which the potentially fatal risks of an intervention have to be weighted against the presumed benefits. Ideally, the treating physician could predict the outcome under different circumstances beforehand and thus make an informed treatment decision. To this end, this article presents two new methods: one for lesion outcome and one for clinical outcome prediction from multispectral magnetic resonance sequences. After extracting tailored image features, a random forest classifier respectively regressor is trained. Both approaches were submitted to the Ischemic Stroke Lesion Segmentation (ISLES) 2017 challenge and obtained a first and third place. The outcome underlines the robustness of our designed features and stresses the approach’s resilience against overfitting when faced with small training datasets.

AB - The treatment of ischemic stroke requires fast decisions for which the potentially fatal risks of an intervention have to be weighted against the presumed benefits. Ideally, the treating physician could predict the outcome under different circumstances beforehand and thus make an informed treatment decision. To this end, this article presents two new methods: one for lesion outcome and one for clinical outcome prediction from multispectral magnetic resonance sequences. After extracting tailored image features, a random forest classifier respectively regressor is trained. Both approaches were submitted to the Ischemic Stroke Lesion Segmentation (ISLES) 2017 challenge and obtained a first and third place. The outcome underlines the robustness of our designed features and stresses the approach’s resilience against overfitting when faced with small training datasets.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85018704681&origin=inward&txGid=85e630f4d156ee532ccbb1c51dc705c6

U2 - 10.1007/978-3-319-55524-9_21

DO - 10.1007/978-3-319-55524-9_21

M3 - Conference contribution

SN - 978-3-319-55523-2

VL - 10154 LNCS

T3 - Lecture Notes in Computer Science

SP - 219

EP - 230

BT - BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries

A2 - Crimi, Alessandro

A2 - Menze, Bjoern

A2 - Maier, Oskar

A2 - Reyes, Mauricio

A2 - Winzeck, Stefan

A2 - Handels, Heinz

PB - Springer, Cham

T2 - 19th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2016<br/>

Y2 - 17 October 2016 through 21 October 2016

ER -

Maier O, Handels H. Predicting stroke lesion and clinical outcome with random forests. In Crimi A, Menze B, Maier O, Reyes M, Winzeck S, Handels H, editors, BrainLes 2016: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Vol. 10154 LNCS. Springer, Cham. 2016. p. 219-230. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-55524-9_21

Predicting stroke lesion and clinical outcome with random forests

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this