RAPT: A Parallel Radiotherapy Treatment Planning Code


Gill, A, Surridge, M, Scielzo, G, Felici, R, Modesti, M and Sardu, G (1996) RAPT: A Parallel Radiotherapy Treatment Planning Code. High Performance Computing and Networking, Brussels, Belgium, 15 - 19 Apr 1996. Springer, 183-193.

Download

Full text not available from this repository.

Description/Abstract

We describe the RAPT prototype software package for stereotactic radiotherapy treatment planning, which has been developed as part of the ESPRIT EUROPORT project for creating parallel high-performance computing (HPC) versions of commercially significant codes. This paper focuses on the parallel high-performance simulation kernel, and describes its implementation, and performance measurements for both accuracy and computation time. These show that a fully 3-D Monte-Carlo simulation, taking account of tissue inhomogeneities can be performed in around 1 hour an acceptable delay for clinical use. Every year 700,000 Europeans die of cancer, and 1.2 million contract the disease. About a quarter of the new cases each year receive radiosurgery, but 20 per cent of these treatments fail either because the tumours are not precisely located, or the irradiation does not reach every part of them. By making available an accurate simulation on a platform which is affordable by a typical, medium-sized hospital radiotherapy department, RAPT has the potential to improve treatment in many of these cases, and so to make a direct impact on patient care.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Additional Information: Lecture Notes in Computer Science Event Dates: 15-19 April 1996
ISSNs: 3540611428
Divisions: Faculty of Physical Sciences and Engineering > Electronics and Computer Science
Faculty of Physical Sciences and Engineering > Electronics and Computer Science > IT Innovation Centre
ePrint ID: 252523
Date Deposited: 21 Mar 2003
Last Modified: 27 Mar 2014 19:54
Publisher: Springer
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/252523

Actions (login required)

View Item View Item