Picture of person typing on laptop with programming code visible on the laptop screen

World class computing and information science research at Strathclyde...

The Strathprints institutional repository is a digital archive of University of Strathclyde's Open Access research outputs. Strathprints provides access to thousands of Open Access research papers by University of Strathclyde researchers, including by researchers from the Department of Computer & Information Sciences involved in mathematically structured programming, similarity and metric search, computer security, software systems, combinatronics and digital health.

The Department also includes the iSchool Research Group, which performs leading research into socio-technical phenomena and topics such as information retrieval and information seeking behaviour.

Explore

High-level synthesis for medical image processing on Systems on Chip : a case study

Robinson, Fraser D and Crockett, Louise H and Nailon, William H and Stewart, Robert W (2016) High-level synthesis for medical image processing on Systems on Chip : a case study. In: 2016 26th International Conference on Field Programmable Logic and Applications (FPL), 2016. EPFL, Lausanne. ISBN 9781509008513

[img]
Preview
Text (Robinson-etal-FPL2016-High-level-synthesis-for-medical-image-processing-on-systems-on-chip)
Robinson_etal_FPL2016_High_level_synthesis_for_medical_image_processing_on_systems_on_chip.pdf - Accepted Author Manuscript

Download (97kB) | Preview

Abstract

Adaptive radiotherapy is a technique intended to increase the accuracy of radiotherapy. Currently, it is not clinically feasible due to the time required to process the images of patient anatomy. Hardware acceleration of image processing algorithms may allow them to be carried out in a clinically acceptable timeframe. This paper presents the experiences encountered using high-level synthesis tools to design an accelerated segmentation algorithm for computed tomography images targeted for implementation on a System on Chip. Hardware coprocessors and their interfaces for optimal threshold generation and 3D mean filter algorithms were synthesised from C++ functions. Hardware acceleration significantly outperformed the software only implementation. The high-level synthesis tools allowed the rapid exploration of different design options. However, hardware design knowledge was still necessary in order to interpret the results effectively.