Pro-Tech Services, Inc. proposes to develop an innovative open software framework for configuring, deploying and discovering heterogeneous data-sharing communities for biomedical research, with the sleep medicine/Polysomnography (PSG) community as its first application. Biomed-P2P peer-to-peer data-sharing technology allows any biomedical researcher connected to the Internet to share a data resource and search for other resources using only their web-browser and the SBIR developed Biomed-P2P client software. This framework enables flexible, descriptive metadata (information describing the content) to layer on top of the biomedical data files that are distributed on existing peer-to-peer networks (such as the Gnutella Open P2P platform), and can be used to share a variety of heterogeneous biomedical resources. Phase I of the proposed project will focus on the feasibility of implementing a platform-independent Biomed- P2P data sharing framework based on a distributed computing model in which biomedical researchers share computer resources and exchange services directly, without a centralized data storage site. Biomed-P2P will let biomedical researchers define templates or schemas for the type of data files they are sharing and facilitate sharing the schema within the network community just like any other file. This will allow the expansion of peer-to-peer networks with new data file types and new schemas without any centralized server or authority. Furthermore, since Biomed-P2P plans to use XML to describe the shared objects, this will allow sharing any kind of network accessible files such as PSG studies, scoring results files, PSG reports, images and text. Phase II will complete the design work and result in advanced prototypes and software tools that fully integrate into a sleep laboratory's existing sleep diagnostic PSG products. Phase II will also finish the design and implementation of software modules that will expand usage of the Biomed-P2P to other biomedical research fields such as Epilepsy research, Neuroscience, Human Genome project and others.
