Home » Technology » Scientific Computing » The BioPSE Project

The BioPSE Project

© Adam Hughes

As we've seen in the past, the development of a problem-solving environment, or PSE, has been discussed and attempted in many fields of scientific study by countless disparate groups. Such an effort requires diligent planning and a thorough understanding of the problems commonly encountered in a simulation setting involving the field of interest. Seldom has such an undertaking yielded such a self-consistent and viable design as is presented in the BioPSE (Bioelectric Problem Solving Environment) project at the National Institutes for Health (NIH).

Realizing that the computational study of bioelectric field phenomena has traditionally involved the use of many different pieces of software to address different aspects of a given problem, the NIH developers set out to address the problem of continuity within a simulation study. One of the major issues present in this manner of computational work is the overhead involved in maintaining input decks for each piece of software, translating data between formats and analyzing the widely varying types of data obtained as results.

The BioPSE project attempts to overcome much of this wasted effort on the part of the scientist by keeping the data off the computer disk and internalizing it during the life of the simulation. This means that data flows from one step of a study to the next without human intervention for translation purposes. This, in itself, is a major improvement over the old conglomeration method, but it leads to an even more sophisticated idea.

Perhaps BioPSE's most exciting feature is that it allows the scientist to guide his simulation at any time during its execution. This ability to provide direction on-the-fly has been dubbed "computational steering," and it is potentially very powerful. This feature would allow the scientist to direct the output of one segment of the simulation as input for any number of possible calculations further downstream, thus enhancing the overall versatility of any simulation and also the quality of the insights it provides.

In addition to the impressive features listed above, BioPSE allows users access to the building blocks of the PSE, giving them the power to add in their own applications under the BioPSE umbrella. In addition, there are tools available to manage this functionality.

All in all, it appears that BioPSE should prove to be a very valuable tool for the computational bioelectric field scientist.

The copyright of the article The BioPSE Project in Scientific Computing is owned by Adam Hughes. Permission to republish The BioPSE Project in print or online must be granted by the author in writing.