Segmentation for Digital Pathology Studies
Molecular Pathology Image Segmentation
[ USPTO Patent 7,949,181 ] [ ISBI 2008 ] [ MIAAB 2007 ]
A tissue microarray with known disease stages can be used to enable efficient pathology slide image analysis. We focus on an intuitive approach for segmenting such images, using the Hierarchical Self-Organizing Map (HSOM). Our approach introduces the use of unsupervised clustering using both color and texture features, and the use of unsupervised color merging outside of the HSOM framework. An application to segmentation for prostrate cancer staging is presented.
multimodality image registration
Deformable Registration with Spatially Varying Degrees of Freedom (DOF)
[ ISBI 2008 ]
Intrasubject deformable registration applications use a high DOF deformation to accurately align soft tissue. However, smoothness constraints applied to the deformation may distort the more rigid tissue types. We present a technique that aligns tissue types based on spatially varying DOF.
Anatomically guided image registration
[ USPTO Application 20070280556 ] [ AIPR 2006 ]
Simple anatomically guided approach for whole body image registration offers enhanced alignment of large coverage inter-scan studies. In this experiment, we provide anatomy specific transformations to capture their independent motions.
Validation of Registration Methods
Correlating Perfusion/Diffusion to Anatomy
[ ASNR 2007 ]
This was a clinical study designed to assess the performance of our multimodality registration algorithms in registering functional (PET/DWI) images with corresponding anatomical images (MRI). Quantitative assessment was done using point correspondences.