The primary objective of this project is to develop new software tools that utilize laser graphics to help the clinical investigator perceive significant information in clinical research datasets. The Phase I activities, performed in collaboration with investigators at the Harvard School of Public Health, identified a specific novel software concept, a """"""""MultiGraph Visualization Tool"""""""", that appears to be distinctly useful in exploring multivariate data to detect differences between subpopulations in cancer studies. The proposed Phase II work will use the statistical graphics concepts and software architecture from Phase I to develop a full software prototype which will be evaluated and refined through interaction with investigators from several clinical research centers. At completion, the prototype will become the basis of a privately-funded Phase III commercialization effort to make this capability broadly available to the clinical research community. The major technical innovation in this project is expected to be the development of visualization software oriented towards laser printer presentations of large volumes of clinical research data. The major health-related contribution is hoped to be the generation of clinical insights through the increased ability of the investigator, assisted by the new software, to notice important patterns in clinical data.
