Abstract
A University of Central Florida researcher has developed a tissue image database system that enables users to search for and match images based on the presence of a target molecule, internal anatomical structure, or both. The patented system provides powerful tools that can help to expand and support medical and scientific research and informational exchange. For example, a user may be able to immediately determine whether a drug candidate affects a region or certain cells of the brain and from that information can predict its efficacy.
Current image databases employ mining tools that mainly use text-based ontologies and metadata. In contrast, the UCF invention can search the database using query images. The search engine may automatically extract visual characteristics of images presented as queries, search image databases, and list candidate images with characteristics similar to the original queries. This system can perform reverse searches, namely mining data associated with the image using visual features between the query and original image.
Technical Details
The UCF invention comprises a user-accessible tissue image database system and methods for using it. An example system setup can include a computer unit with at least one processing module, at least two computer-readable program code modules, and multiple digital images of tissue samples indicating a target molecule. It may also take the form of hardware or a combination of software and hardware with a storage medium, such as hard disks, CD-ROMs, DVDs, optical storage devices or magnetic storage devices. The system can include images of internal tissue obtained by magnetic resonance imaging or X-ray in PGM format.
Partnering Opportunity
The research team is seeking partners for licensing and/or research collaboration.
Stage of Development
Prototype in development.
Benefit
Search the database not only using metadata of images but also the image itselfMarket Application
Clinical diagnosis and drug screening
Brochure