SPIE Medical Imaging: Live Demonstrations

Tuesday 13 February 2018, 5:00 PM - 7:00 PM, Salon D/E

Sedeen: An extensible viewer for digital pathology

Anne Martel, Sunnybrook Research Institute, Toronto, Metin Gurcan, Wake Forest, NC, Anant Madabhushi, Case Western Reserve University, OH

 With the advent of whole slide digital scanners, histopathology slides can be digitized into very high-resolution digital images, realizing a new "big data" stream that can potentially rival "omics data" in size and complexity. Digital pathology enabled companion diagnostic tests could yield predictions of cancer risk and aggressiveness in a manner similar to molecular diagnostic tests and there is great interest in using new image analysis techniques to create these tests. In order to collect the training data needed to develop new algorithms and then carry out the clinical validation needed to test them, a software platform that addresses the needs of pathologists, cancer researchers and algorithm developers is needed; we aim to provide such a platform under our NIH ITCR funded program to build a Pathology Image Informatics Platform (PIIP:www.pathiip.org).

We are expanding the capabilities of Sedeen, a freely available digital pathology viewer that is capable of displaying whole slide images in a wide variety of formats. In addition to the built in annotation and image registration tools, a software developer’s kit allows researchers to extend the analysis capabilities of the viewer and the PIIP team have already created plugins for several existing analysis algorithms including stain quantification and out of focus detection. The aim of this demonstration will be to introduce the platform to researchers working in the area of digital pathology analysis, demonstrate how to co-register an MR image with a WSI and to solicit feedback on what additional functionality is required.