Hariharan Ravishankar1, Igor Barani2, Sheshadri Thiruvenkadam1, Marc Mabray2, KS Shriram1, SoonMee Cha2, Rakesh Mullick3, and Suresh Emmanuel Joel3
1General Electric Global Research, Bangalore, India, 2University of California at San Francisco, San Francisco, CA, United States, 3Diagnostic Imaging and Biomedical Technologies, General Electric Global Research, Bangalore, India
In this work, we address the problem of registration of brain images to atlases in presence of large, space-occupying tumor pathologies. We provide an integrated registration framework to address two types of tumors - a) infiltrating tumors - which penetrate the surrounding tissues b) extrinsic tumors - which grow and compress the surrounding tissues. We propose a mathematical model to shrink extrinsic tumors before registration, thereby reversing the compression effects.