Update on the latest Innovations from Microsoft Research

I have blogged in the past (actually almost exactly a year ago – oh my, how time flies when you are having fun) about Scientist Innovation, and exciting things we are doing in the Research area.  Well, a year is a long time in the software industry, and so it is time for an update.  Among all the great things that are going on, I wanted to focus on the ones that are the most exciting from my perspective.  First of all, there is Pivot.  I have more thoughts than I can handle in a Blog about the potential applications for Pivot for visualizing data and images.  Be sure to check out this video: Gary Flake discusses Pivot @ TED2010  I am already talking to several customers about how to build Pivot Collections of DICOM images.  This could be huge, as medical diagnostic systems are generating vast amounts of image data that are used in Clinical Trials, and companies are having a really hard time managing all this.  DICOM images are typically stored in PACS systems.  These are relatively old systems built for a specific purpose.  Viewing and manipulating large image sets has not been the intent when designing them.  Another exciting development is the recent announcement of a Silverlight Control for Pivot.  This opens up the potential for almost everyone to access this great technology with a Browser.  The Pivot team is working double overtime to keep up with the exploding demand, and more exciting announcements are coming in the summer.  Here are some really great examples of Pivot in action: http://momcollection.cloudapp.net/   http://netflixpivot.cloudapp.net/   I could not help put on my thinking hat about how all this could work, and one of the ideas I am investigating is how to do even more with images and Pivot.  What if we could store the images in a repository that is ‘semantically aware’, so we could go beyond the limitations of file systems?  As it happens, the folks at Microsoft Research have already built such a repository called Zentity.  They call it a Research-Output Repository, but for me it is simpler to refer to it as a semantically aware repository.  I am starting to think about how to automatically pull out images and their relevant metadata from PACS systems and to create Pivot Collections and to use Zentity as the repository.  More to come!  Microsoft External Research has a whole range of great projects that are highly relevant to the BioPharma community.  See here for a great deck by Alex Wade.  The latest announcement from the Scholarly Communications Team is the Chem4Word add-in.  See here for a 3-minute introductory demo.   Finally, be sure to check out the Clinical Documentation Solution Accelerator (CDSA)  – a truly amazing free application that really shows off the power of the Office platform: http://www.mscui.net/CDSA.htm  The code can be downloaded from here: http://code.msdn.microsoft.com/cdsa

I have also been following closely the evolution of the Microsoft Semantic Engine.  See here for a PDC presentation to learn more.  The presentation deck can be found here.  Here is a great Blog posting about it.  I am really looking forward to seeing this technology make its way as a core element of the Microsoft stack, and to augment our Search and database technologies.  Among other things, the potential to integrate the Semantic Engine with solutions such as MetaPoint for metadata enrichment using semantic approaches is really exciting, and will solve many difficult problems in the Enterprise Information space.