Method for Graphic Parallel Computing of an Imaging Device

The invention allows to develop and/or modify an existing software using an innovative approach that can process a large amount of data in a shorter time, up to 100 times less than systems actually in use.

Title of the patent Method for the interactive parallel processing of data on a cluster with the graphic input/output on a visualization device
Thematic area Electrical, electronic and information engineering
Ownership Alma Mater Studiorum – Università di Bologna, National Institute of Nuclear Physics (INFN)
Inventors Rosa Brancaccio, Franco Casali, Maria Pia Morigi, Matteo Bettuzzi, Giuseppe Levi
Protection Europe, USA
Licensing status Available for licensing agreement
Keywords parallel computing, graphic user interface, speed, real time, user-friendly systems, software optimization, hardware integration
Filed on Nov 16, 2012

Long calculation times, due to the processing of a significant amount of data, means waiting to make important decisions. Too long waiting times de facto prevent the development of technological products requiring instead more details and better features.

At present, the common solution adopted to reduce computation time is to assume systems based on parallel multiprocessors (less expensive and less efficient) or graphics processors GPU (more expensive, better performing but need to rewrite the software).

This invention allows to develop and/or modify an existing software for the processing of a large amount of data in simultaneously parallel and graphic environment.

By using this method, the software will provide data into one-hundredth of the duration initially foreseen, by adopting  a low-cost hardware system. The method allows to create user-friendly graphical interfaces and to hide the complexity of parallel computing , by making the software automatic and easy to use.

It is also possible to integrate, at low cost, the electronic management and synchronization of the instrumentation.

The new method allows to increase the calculation speed  by at least 33 times, thus improving the performance of any tool that works on a significant amount of data (such as a high resolution medical tomography). Main advantages are impressive calculation speedup at low cost and full compatibility with graphics and management of the instrumentation.

Example

Assuming that 5 minutes are required to process medical CT data,  by applying the new method the same result would be obtained in 9 seconds (even down to 3 seconds) with a saving time of two weeks and 600 patients  more that may access the exam per year. Otherwise, it could be possible to maintain a processing time of 5 minutes but increase, the accuracy of the system by up to 33 times.