Accuracy Estimation for Medical Image Registration Using Regression Forests

Hessam Sokooti; Gorkem Saygili; Ben Glocker; Boudewijn Lelieveldt; Marius Staring

doi:10.1007/978-3-319-46726-9_13

Accuracy Estimation for Medical Image Registration Using Regression Forests

Hessam Sokooti, Gorkem Saygili, Ben Glocker, Boudewijn Lelieveldt, Marius Staring

Pattern Recognition and Bioinformatics

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

23 Citations (Scopus)

Abstract

This paper reports a new automatic algorithm to estimate the misregistration in a quantitative manner. A random regression forest is constructed, predicting the local registration error. The forest is built using local and modality independent features related to the registration precision, the transformation model and intensity-based similarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans. The results show that the mean absolute error of regression is 0.72 ± 0.96 mm and the accuracy of classification in three classes (correct, poor and wrong registration) is 93.4 %, comparing favorably to a competing method. In conclusion, a method was proposed that for the first time shows the feasibility of automatic registration assessment by means of regression, and promising results were obtained.

Original language	English
Title of host publication	Medical Image Computing and Computer-Assisted Intervention MICCAI 2016
Subtitle of host publication	19th International Conference, Proceedings part 3
Editors	S. Ourselin, L. Joskowicz, M. Sabuncu , G. Unal, W. Wells
Place of Publication	Cham
Publisher	Springer
Pages	107-115
Number of pages	9
ISBN (Electronic)	978-3-319-46726-9
ISBN (Print)	978-3-319-46725-2
DOIs	https://doi.org/10.1007/978-3-319-46726-9_13
Publication status	Published - 2016
Event	MICCAI 2016 - Medical Image Computing and Computer-Assisted Intervention: 19th International Conference - Athens, Greece Duration: 16 Oct 2016 → 21 Oct 2016

Publication series

Name	Lecture Notes in Computer Science
Volume	9902
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	MICCAI 2016 - Medical Image Computing and Computer-Assisted Intervention
Abbreviated title	MICCAI 2016
Country/Territory	Greece
City	Athens
Period	16/10/16 → 21/10/16

Keywords

Image registration
Registration accuracy
Uncertainty estimation
Regression forests

Access to Document

10.1007/978-3-319-46726-9_13

Cite this

Sokooti, H., Saygili, G., Glocker, B., Lelieveldt, B., & Staring, M. (2016). Accuracy Estimation for Medical Image Registration Using Regression Forests. In S. Ourselin, L. Joskowicz, M. Sabuncu , G. Unal, & W. Wells (Eds.), Medical Image Computing and Computer-Assisted Intervention MICCAI 2016: 19th International Conference, Proceedings part 3 (pp. 107-115). (Lecture Notes in Computer Science; Vol. 9902). Springer. https://doi.org/10.1007/978-3-319-46726-9_13

Sokooti, Hessam ; Saygili, Gorkem ; Glocker, Ben et al. / Accuracy Estimation for Medical Image Registration Using Regression Forests. Medical Image Computing and Computer-Assisted Intervention MICCAI 2016: 19th International Conference, Proceedings part 3. editor / S. Ourselin ; L. Joskowicz ; M. Sabuncu ; G. Unal ; W. Wells. Cham : Springer, 2016. pp. 107-115 (Lecture Notes in Computer Science).

@inproceedings{9fa960e00c6847b58a6e6f4cfbbc16de,

title = "Accuracy Estimation for Medical Image Registration Using Regression Forests",

abstract = "This paper reports a new automatic algorithm to estimate the misregistration in a quantitative manner. A random regression forest is constructed, predicting the local registration error. The forest is built using local and modality independent features related to the registration precision, the transformation model and intensity-based similarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans. The results show that the mean absolute error of regression is 0.72 ± 0.96 mm and the accuracy of classification in three classes (correct, poor and wrong registration) is 93.4 %, comparing favorably to a competing method. In conclusion, a method was proposed that for the first time shows the feasibility of automatic registration assessment by means of regression, and promising results were obtained.",

keywords = "Image registration, Registration accuracy, Uncertainty estimation, Regression forests",

author = "Hessam Sokooti and Gorkem Saygili and Ben Glocker and Boudewijn Lelieveldt and Marius Staring",

year = "2016",

doi = "10.1007/978-3-319-46726-9_13",

language = "English",

isbn = "978-3-319-46725-2",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "107--115",

editor = "{ Ourselin}, S. and L. Joskowicz and {Sabuncu }, M. and G. Unal and W. Wells",

booktitle = "Medical Image Computing and Computer-Assisted Intervention MICCAI 2016",

note = "MICCAI 2016 - Medical Image Computing and Computer-Assisted Intervention : 19th International Conference, MICCAI 2016 ; Conference date: 16-10-2016 Through 21-10-2016",

}

Sokooti, H, Saygili, G, Glocker, B, Lelieveldt, B & Staring, M 2016, Accuracy Estimation for Medical Image Registration Using Regression Forests. in S Ourselin, L Joskowicz, M Sabuncu , G Unal & W Wells (eds), Medical Image Computing and Computer-Assisted Intervention MICCAI 2016: 19th International Conference, Proceedings part 3. Lecture Notes in Computer Science, vol. 9902, Springer, Cham, pp. 107-115, MICCAI 2016 - Medical Image Computing and Computer-Assisted Intervention, Athens, Greece, 16/10/16. https://doi.org/10.1007/978-3-319-46726-9_13

Accuracy Estimation for Medical Image Registration Using Regression Forests. / Sokooti, Hessam; Saygili, Gorkem; Glocker, Ben et al.
Medical Image Computing and Computer-Assisted Intervention MICCAI 2016: 19th International Conference, Proceedings part 3. ed. / S. Ourselin; L. Joskowicz; M. Sabuncu ; G. Unal; W. Wells. Cham: Springer, 2016. p. 107-115 (Lecture Notes in Computer Science; Vol. 9902).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Accuracy Estimation for Medical Image Registration Using Regression Forests

AU - Sokooti, Hessam

AU - Saygili, Gorkem

AU - Glocker, Ben

AU - Lelieveldt, Boudewijn

AU - Staring, Marius

PY - 2016

Y1 - 2016

N2 - This paper reports a new automatic algorithm to estimate the misregistration in a quantitative manner. A random regression forest is constructed, predicting the local registration error. The forest is built using local and modality independent features related to the registration precision, the transformation model and intensity-based similarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans. The results show that the mean absolute error of regression is 0.72 ± 0.96 mm and the accuracy of classification in three classes (correct, poor and wrong registration) is 93.4 %, comparing favorably to a competing method. In conclusion, a method was proposed that for the first time shows the feasibility of automatic registration assessment by means of regression, and promising results were obtained.

AB - This paper reports a new automatic algorithm to estimate the misregistration in a quantitative manner. A random regression forest is constructed, predicting the local registration error. The forest is built using local and modality independent features related to the registration precision, the transformation model and intensity-based similarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans. The results show that the mean absolute error of regression is 0.72 ± 0.96 mm and the accuracy of classification in three classes (correct, poor and wrong registration) is 93.4 %, comparing favorably to a competing method. In conclusion, a method was proposed that for the first time shows the feasibility of automatic registration assessment by means of regression, and promising results were obtained.

KW - Image registration

KW - Registration accuracy

KW - Uncertainty estimation

KW - Regression forests

U2 - 10.1007/978-3-319-46726-9_13

DO - 10.1007/978-3-319-46726-9_13

M3 - Conference contribution

SN - 978-3-319-46725-2

T3 - Lecture Notes in Computer Science

SP - 107

EP - 115

BT - Medical Image Computing and Computer-Assisted Intervention MICCAI 2016

A2 - Ourselin, S.

A2 - Joskowicz, L.

A2 - Sabuncu , M.

A2 - Unal, G.

A2 - Wells, W.

PB - Springer

CY - Cham

T2 - MICCAI 2016 - Medical Image Computing and Computer-Assisted Intervention

Y2 - 16 October 2016 through 21 October 2016

ER -

Sokooti H, Saygili G, Glocker B, Lelieveldt B , Staring M. Accuracy Estimation for Medical Image Registration Using Regression Forests. In Ourselin S, Joskowicz L, Sabuncu M, Unal G, Wells W, editors, Medical Image Computing and Computer-Assisted Intervention MICCAI 2016: 19th International Conference, Proceedings part 3. Cham: Springer. 2016. p. 107-115. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-46726-9_13

Accuracy Estimation for Medical Image Registration Using Regression Forests

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this