Prediction of Tumour Motion using Interacting Multiple Model Filter

Devi Putra; LOlivier C.L. Haas; John A. Mills; Keith J. Burnham

Prediction of Tumour Motion using Interacting Multiple Model Filter

Devi Putra, LOlivier C.L. Haas, John A. Mills, Keith J. Burnham

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding › peer-review

142 Downloads (Pure)

Abstract

Accurate prediction of tumour motion - over a prescribed time - is essential for enabling adaptive radiotherapy. The prediction time horizon is determined by measurement processing time, predictor algorithm processing time and the time-to-adapt radiation delivery. A trade off between the predictor algorithm complexity and the required prediction time horizon, therefore, has to be made. This paper proposes an interacting multiple model (IMM) filter and two Kalman filters to predict 0.2 s ahead respiratory tumour motions. The performance of the filters is evaluated using 333 traces of 4 minutes respiratory motions for 24 adult patients. The average RMSE of the IMM filter and the best Kalman filter with 5Hz measurements rate are 0.98 mm and 1.1 mm, which are improvements of 38% and 30% compared to use of measurements only.

Original language	English
Title of host publication	Advances in Medical, Signal and Information Processing, 2006. MEDSIP 2006. IET 3rd International Conference On
Publisher	IET
Pages	1-4
Number of pages	4
ISBN (Print)	978-0-86341-658-3
Publication status	Published - Jul 2006
Event	3rd IET International Conference on Advances in Medical, Signal and Information Processing - Glasgow, United Kingdom Duration: 17 Jul 2006 → 19 Jul 2006

Conference

Conference	3rd IET International Conference on Advances in Medical, Signal and Information Processing
Country/Territory	United Kingdom
City	Glasgow
Period	17/07/06 → 19/07/06

Bibliographical note

e Framework 6 European integrated project Methods and Advanced Equipment for Simulation and Treatment in Radiation Oncology (MAESTRO) CE LSHC CT 2004 503564.

Keywords

tumour motion prediction methods
Kalman
filter
interacting multiple model filter
medical systems

Access to Document

Haas4combFinal published version, 4.62 MB

http://www.theiet.org

Cite this

Putra, D, Haas, LOCL, Mills, JA & Burnham, KJ 2006, Prediction of Tumour Motion using Interacting Multiple Model Filter. in Advances in Medical, Signal and Information Processing, 2006. MEDSIP 2006. IET 3rd International Conference On. IET, pp. 1-4, 3rd IET International Conference on Advances in Medical, Signal and Information Processing, Glasgow, United Kingdom, 17/07/06. <http://www.theiet.org>

@inproceedings{4d031851726949f7810ab179c9c0c7db,

title = "Prediction of Tumour Motion using Interacting Multiple Model Filter",

abstract = "Accurate prediction of tumour motion - over a prescribed time - is essential for enabling adaptive radiotherapy. The prediction time horizon is determined by measurement processing time, predictor algorithm processing time and the time-to-adapt radiation delivery. A trade off between the predictor algorithm complexity and the required prediction time horizon, therefore, has to be made. This paper proposes an interacting multiple model (IMM) filter and two Kalman filters to predict 0.2 s ahead respiratory tumour motions. The performance of the filters is evaluated using 333 traces of 4 minutes respiratory motions for 24 adult patients. The average RMSE of the IMM filter and the best Kalman filter with 5Hz measurements rate are 0.98 mm and 1.1 mm, which are improvements of 38% and 30% compared to use of measurements only.",

keywords = "tumour motion prediction methods, Kalman, filter, interacting multiple model filter, medical systems",

author = "Devi Putra and Haas, {LOlivier C.L.} and Mills, {John A.} and Burnham, {Keith J.}",

note = "e Framework 6 European integrated project Methods and Advanced Equipment for Simulation and Treatment in Radiation Oncology (MAESTRO) CE LSHC CT 2004 503564.; 3rd IET International Conference on Advances in Medical, Signal and Information Processing ; Conference date: 17-07-2006 Through 19-07-2006",

year = "2006",

month = jul,

language = "English",

isbn = "978-0-86341-658-3",

pages = "1--4",

booktitle = "Advances in Medical, Signal and Information Processing, 2006. MEDSIP 2006. IET 3rd International Conference On",

publisher = "IET",

address = "United Kingdom",

}

TY - GEN

T1 - Prediction of Tumour Motion using Interacting Multiple Model Filter

AU - Putra, Devi

AU - Haas, LOlivier C.L.

AU - Mills, John A.

AU - Burnham, Keith J.

N1 - e Framework 6 European integrated project Methods and Advanced Equipment for Simulation and Treatment in Radiation Oncology (MAESTRO) CE LSHC CT 2004 503564.

PY - 2006/7

Y1 - 2006/7

N2 - Accurate prediction of tumour motion - over a prescribed time - is essential for enabling adaptive radiotherapy. The prediction time horizon is determined by measurement processing time, predictor algorithm processing time and the time-to-adapt radiation delivery. A trade off between the predictor algorithm complexity and the required prediction time horizon, therefore, has to be made. This paper proposes an interacting multiple model (IMM) filter and two Kalman filters to predict 0.2 s ahead respiratory tumour motions. The performance of the filters is evaluated using 333 traces of 4 minutes respiratory motions for 24 adult patients. The average RMSE of the IMM filter and the best Kalman filter with 5Hz measurements rate are 0.98 mm and 1.1 mm, which are improvements of 38% and 30% compared to use of measurements only.

AB - Accurate prediction of tumour motion - over a prescribed time - is essential for enabling adaptive radiotherapy. The prediction time horizon is determined by measurement processing time, predictor algorithm processing time and the time-to-adapt radiation delivery. A trade off between the predictor algorithm complexity and the required prediction time horizon, therefore, has to be made. This paper proposes an interacting multiple model (IMM) filter and two Kalman filters to predict 0.2 s ahead respiratory tumour motions. The performance of the filters is evaluated using 333 traces of 4 minutes respiratory motions for 24 adult patients. The average RMSE of the IMM filter and the best Kalman filter with 5Hz measurements rate are 0.98 mm and 1.1 mm, which are improvements of 38% and 30% compared to use of measurements only.

KW - tumour motion prediction methods

KW - Kalman

KW - filter

KW - interacting multiple model filter

KW - medical systems

M3 - Conference proceeding

SN - 978-0-86341-658-3

SP - 1

EP - 4

BT - Advances in Medical, Signal and Information Processing, 2006. MEDSIP 2006. IET 3rd International Conference On

PB - IET

T2 - 3rd IET International Conference on Advances in Medical, Signal and Information Processing

Y2 - 17 July 2006 through 19 July 2006

ER -

Prediction of Tumour Motion using Interacting Multiple Model Filter

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this