I’m an Experimental Psychologist 👓 by training, and I use brain imaging to answer questions about the human mind. I'm also a software developer 💻 with a passion for creating the next generation of scientific tools to study the brain.
Software 💻
Bio 👤
Brain extraction (aka. skull stripping) is part of many neuroimaging workflows. Sometimes, we may work with images that are abnormal such as data from a stroke patient. Brain extraction methods can be less reliable in these situations due to large lesions. By replacing the missing tissue in the image we can get an accurate brain extracted image. Once extracted from the skull, to view the brain with the lesion we then simply remove the tissue we filled in.
{work in progress}
The process of brain normalization attempts to warp an individual's brain image to match a template image. Similar to some popular "face swapping" mobile phone apps. However, images from stroke patients have lesions which make the brain look abnormal due to differences in anatomy and image intensities. This presents a problem when we wish to match a lesioned image to a normal template. I'm currently working on a project where I compare some popular normalization techniques combined with different approaches to dealing with lesioned images. The goal is to numerically, and statistically find the "best in class" method for normalizing stroke neuroimaging data. The results will directly affect many ongoing projects, as well as future projects since stroke is a leading cause of adult disability in the world.
{this is a work in progress}