IBM Research Computer Science Blog  

IBM computer scientists have been at the forefront of scientific and technological innovation across a broad range of research areas.  They have made pioneering contributions in artificial intelligence, high-speed processor design, computer architecture, natural language processing, programming languages, optimizing compilers, operating systems, storage systems, computer-supported cooperative work, databases, speech recognition, integer programming, and service-oriented architectures, to name a few.


2015 SIIM Innovator Award    (up to IBM Research Computer Science Blog)

Colin Compas, Suman Sedai, David Beymer, Tanveer Syed-Mahmood, and Prasanth Prasanna of IBM Reseach - Almaden and IBM Research - Australia won the Society for Imaging Informatics in Medicine's 2015 Innovator Award for their poster "Annotation Assembler – A Web Tool for Applying Semi-Automatic Annotations to Images", which was presented at the SIIM Annual meeting.

The authors describe their system in the "case presentation" as: "We have developed a web-based tool that allows for the association of semantic image findings with an image. These findings can be anything that could be found in a radiology report, patient history, or image meta-data. Using dictionary based concept extraction our tool is able to automatically identify relevant information in the text that is then verified by the clinician and associated with the given image. The backbone of this system is a hierarchical organizational structure that allows for information about a patient to be stored at different levels. A patient’s gender, for example, is applied to all images contained within that patient’s record, while age is specific for the time of image acquisition. Diseases may be chronic and associated with many imaging studies or acute and associated only with a single image. The main advantages of this system are: 1) Automatic extraction of concepts from text associated with image, 2) Robust search over all information associated with images, 3) Hierarchical storage system that allows for information to be stored at different levels of the patient interaction, and 4) A web-based interface that allows for use of local or remote image collections. A screenshot of the system has been included."

posted by Brent Hailpern on Thu, 23 Jul 2015 14:15:05 -0400