A mixture of experts committee machine to design compensators for intensity modulated radiation therapy

J.H. Goodband; Olivier C.LK. Haas; John A. Mills

doi:10.1016/j.patcog.2006.03.018

A mixture of experts committee machine to design compensators for intensity modulated radiation therapy

J.H. Goodband, Olivier C.LK. Haas, John A. Mills

Research output: Contribution to journal › Article › peer-review

15 Citations (Scopus)

Abstract

This paper presents a new algorithm to produce a near optimal mixture of experts model (MEM) architecture for a continuous mapping. The MEM is applied to a new method incorporating photon scatter for designing compensators for intensity modulated radiation therapy. The algorithm utilizes the fuzzy C-means clustering algorithm to partition data before training commences. A reduction in the size of training sets also allows the Levenberg–Marquardt algorithm to be implemented. As a result, both training time and validation error are reduced. A 71% reduction in prediction error compared with that of a single neural network is achieved.

Original language	English
Pages (from-to)	1704-1714
Journal	Pattern Recognition
Volume	39
Issue number	9
DOIs	https://doi.org/10.1016/j.patcog.2006.03.018
Publication status	Published - Sept 2006

Bibliographical note

This paper is not yet available on the repository

Funder

Engineering and Physical Sciences Research Council, Industrial Case Training award no. 02303507

Keywords

Committee machines
Neural networks
Fuzzy C-means
Compensators
Radiation therapy

Access to Document

10.1016/j.patcog.2006.03.018

MAESTRO: Methods and Advanced Equipment for Simulation and Treatment in Radiation Oncology
haas, O., Burnham, K., Skworcow, P. & Sahih, A.
1/05/04 → 30/04/09
Project: Research

Cite this

@article{58c9fa931f9848de9cdbb960ba0dc1b3,

title = "A mixture of experts committee machine to design compensators for intensity modulated radiation therapy",

abstract = "This paper presents a new algorithm to produce a near optimal mixture of experts model (MEM) architecture for a continuous mapping. The MEM is applied to a new method incorporating photon scatter for designing compensators for intensity modulated radiation therapy. The algorithm utilizes the fuzzy C-means clustering algorithm to partition data before training commences. A reduction in the size of training sets also allows the Levenberg–Marquardt algorithm to be implemented. As a result, both training time and validation error are reduced. A 71% reduction in prediction error compared with that of a single neural network is achieved.",

keywords = "Committee machines, Neural networks, Fuzzy C-means, Compensators, Radiation therapy",

author = "J.H. Goodband and Haas, {Olivier C.LK.} and Mills, {John A.}",

note = "This paper is not yet available on the repository",

year = "2006",

month = sep,

doi = "10.1016/j.patcog.2006.03.018",

language = "English",

volume = "39",

pages = "1704--1714",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier",

number = "9",

}

TY - JOUR

T1 - A mixture of experts committee machine to design compensators for intensity modulated radiation therapy

AU - Goodband, J.H.

AU - Haas, Olivier C.LK.

AU - Mills, John A.

N1 - This paper is not yet available on the repository

PY - 2006/9

Y1 - 2006/9

N2 - This paper presents a new algorithm to produce a near optimal mixture of experts model (MEM) architecture for a continuous mapping. The MEM is applied to a new method incorporating photon scatter for designing compensators for intensity modulated radiation therapy. The algorithm utilizes the fuzzy C-means clustering algorithm to partition data before training commences. A reduction in the size of training sets also allows the Levenberg–Marquardt algorithm to be implemented. As a result, both training time and validation error are reduced. A 71% reduction in prediction error compared with that of a single neural network is achieved.

AB - This paper presents a new algorithm to produce a near optimal mixture of experts model (MEM) architecture for a continuous mapping. The MEM is applied to a new method incorporating photon scatter for designing compensators for intensity modulated radiation therapy. The algorithm utilizes the fuzzy C-means clustering algorithm to partition data before training commences. A reduction in the size of training sets also allows the Levenberg–Marquardt algorithm to be implemented. As a result, both training time and validation error are reduced. A 71% reduction in prediction error compared with that of a single neural network is achieved.

KW - Committee machines

KW - Neural networks

KW - Fuzzy C-means

KW - Compensators

KW - Radiation therapy

U2 - 10.1016/j.patcog.2006.03.018

DO - 10.1016/j.patcog.2006.03.018

M3 - Article

SN - 0031-3203

VL - 39

SP - 1704

EP - 1714

JO - Pattern Recognition

JF - Pattern Recognition

IS - 9

ER -

A mixture of experts committee machine to design compensators for intensity modulated radiation therapy

Abstract

Bibliographical note

Funder

Keywords

Access to Document

Fingerprint

Projects

MAESTRO: Methods and Advanced Equipment for Simulation and Treatment in Radiation Oncology

Cite this