ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering

Hanna-Leena Halme, Antti Korvenoja, Eero Salli

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

Automated methods for segmentation of ischemic stroke lesions could significantly reduce the workload of radiologists and speed up the beginning of patient treatment. In this paper, we present a method for subacute ischemic stroke lesion segmentation from multispectral magnetic resonance images (MRI). The method involves classification of voxels with a Random Forest algorithm and subsequent classification refinement with contextual clustering. In addition, we utilize the training data to build statistical group-specific templates and use them for calculation of individual voxel-wise differences from the global mean. Our method achieved a Dice coefficient of 0.61 for the leave-one-out cross-validated training data and 0.47 for the testing data of the ISLES challenge 2015.
Original languageEnglish
Title of host publicationBrainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries : First International Workshop, Brainles 2015, Held in Conjunction with MICCAI 2015
EditorsAlessandro Crimi, Bjoern Menze, Oskar Maier, Mauricio Reyes, Heinz Handels
Number of pages11
Volume9556
Place of PublicationSveitsi
PublisherSpringer
Publication date19 Mar 2016
Pages211-221
ISBN (Print)978-3-319-30857-9
ISBN (Electronic)978-3-319-30858-6
DOIs
Publication statusPublished - 19 Mar 2016
MoE publication typeA4 Article in conference proceedings
EventBrainlesion 2015 - Munich, Germany
Duration: 5 Oct 20155 Oct 2015
Conference number: 1

Publication series

NameLecture Notes in Computer Science
Volume9556
ISSN (Print)0302-9743

Fields of Science

  • 3126 Surgery, anesthesiology, intensive care, radiology

Cite this

Halme, H-L., Korvenoja, A., & Salli, E. (2016). ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering. In A. Crimi, B. Menze, O. Maier, M. Reyes, & H. Handels (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: First International Workshop, Brainles 2015, Held in Conjunction with MICCAI 2015 (Vol. 9556, pp. 211-221). (Lecture Notes in Computer Science; Vol. 9556). Sveitsi: Springer. https://doi.org/10.1007/978-3-319-30858-6_18
Halme, Hanna-Leena ; Korvenoja, Antti ; Salli, Eero. / ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: First International Workshop, Brainles 2015, Held in Conjunction with MICCAI 2015. editor / Alessandro Crimi ; Bjoern Menze ; Oskar Maier ; Mauricio Reyes ; Heinz Handels. Vol. 9556 Sveitsi : Springer, 2016. pp. 211-221 (Lecture Notes in Computer Science).
@inproceedings{123f3a364a3d44c893de57744c5b6540,
title = "ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering",
abstract = "Automated methods for segmentation of ischemic stroke lesions could significantly reduce the workload of radiologists and speed up the beginning of patient treatment. In this paper, we present a method for subacute ischemic stroke lesion segmentation from multispectral magnetic resonance images (MRI). The method involves classification of voxels with a Random Forest algorithm and subsequent classification refinement with contextual clustering. In addition, we utilize the training data to build statistical group-specific templates and use them for calculation of individual voxel-wise differences from the global mean. Our method achieved a Dice coefficient of 0.61 for the leave-one-out cross-validated training data and 0.47 for the testing data of the ISLES challenge 2015.",
keywords = "3126 Surgery, anesthesiology, intensive care, radiology",
author = "Hanna-Leena Halme and Antti Korvenoja and Eero Salli",
note = "Volume: Proceeding volume: 9556",
year = "2016",
month = "3",
day = "19",
doi = "10.1007/978-3-319-30858-6_18",
language = "English",
isbn = "978-3-319-30857-9",
volume = "9556",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "211--221",
editor = "Alessandro Crimi and Bjoern Menze and Oskar Maier and Mauricio Reyes and Heinz Handels",
booktitle = "Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries",
address = "United States",

}

Halme, H-L, Korvenoja, A & Salli, E 2016, ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering. in A Crimi, B Menze, O Maier, M Reyes & H Handels (eds), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: First International Workshop, Brainles 2015, Held in Conjunction with MICCAI 2015. vol. 9556, Lecture Notes in Computer Science, vol. 9556, Springer, Sveitsi, pp. 211-221, Brainlesion 2015, Munich, Germany, 05/10/2015. https://doi.org/10.1007/978-3-319-30858-6_18

ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering. / Halme, Hanna-Leena; Korvenoja, Antti; Salli, Eero.

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: First International Workshop, Brainles 2015, Held in Conjunction with MICCAI 2015. ed. / Alessandro Crimi; Bjoern Menze; Oskar Maier; Mauricio Reyes; Heinz Handels. Vol. 9556 Sveitsi : Springer, 2016. p. 211-221 (Lecture Notes in Computer Science; Vol. 9556).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

TY - GEN

T1 - ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering

AU - Halme, Hanna-Leena

AU - Korvenoja, Antti

AU - Salli, Eero

N1 - Volume: Proceeding volume: 9556

PY - 2016/3/19

Y1 - 2016/3/19

N2 - Automated methods for segmentation of ischemic stroke lesions could significantly reduce the workload of radiologists and speed up the beginning of patient treatment. In this paper, we present a method for subacute ischemic stroke lesion segmentation from multispectral magnetic resonance images (MRI). The method involves classification of voxels with a Random Forest algorithm and subsequent classification refinement with contextual clustering. In addition, we utilize the training data to build statistical group-specific templates and use them for calculation of individual voxel-wise differences from the global mean. Our method achieved a Dice coefficient of 0.61 for the leave-one-out cross-validated training data and 0.47 for the testing data of the ISLES challenge 2015.

AB - Automated methods for segmentation of ischemic stroke lesions could significantly reduce the workload of radiologists and speed up the beginning of patient treatment. In this paper, we present a method for subacute ischemic stroke lesion segmentation from multispectral magnetic resonance images (MRI). The method involves classification of voxels with a Random Forest algorithm and subsequent classification refinement with contextual clustering. In addition, we utilize the training data to build statistical group-specific templates and use them for calculation of individual voxel-wise differences from the global mean. Our method achieved a Dice coefficient of 0.61 for the leave-one-out cross-validated training data and 0.47 for the testing data of the ISLES challenge 2015.

KW - 3126 Surgery, anesthesiology, intensive care, radiology

U2 - 10.1007/978-3-319-30858-6_18

DO - 10.1007/978-3-319-30858-6_18

M3 - Conference contribution

SN - 978-3-319-30857-9

VL - 9556

T3 - Lecture Notes in Computer Science

SP - 211

EP - 221

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

A2 - Crimi, Alessandro

A2 - Menze, Bjoern

A2 - Maier, Oskar

A2 - Reyes, Mauricio

A2 - Handels, Heinz

PB - Springer

CY - Sveitsi

ER -

Halme H-L, Korvenoja A, Salli E. ISLES (SISS) Challenge 2015: Segmentation of Stroke Lesions Using Spatial Normalization, Random Forest Classification and Contextual Clustering. In Crimi A, Menze B, Maier O, Reyes M, Handels H, editors, Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: First International Workshop, Brainles 2015, Held in Conjunction with MICCAI 2015. Vol. 9556. Sveitsi: Springer. 2016. p. 211-221. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-30858-6_18