IT News & Events

News about IT at Indiana University and the world

Menu

IU PhD student set to understand the shape of human white matter through high-performance computing

Kitchell’s work, in the laboratory of Dr. Franco Pestilli, focuses on developing methods of examining the morphology (or, shape) of the brain’s white matter tracts.

Did you know that your brain generates about 12-25 watts of electricity, which is enough power to light a low-wattage LED light bulb? The human brain is a fascinating organ which is involved in nearly everything that we do and is the main topic of study for fourth-year IU graduate student Lindsey Kitchell. Human brains are made up of two types of matter: gray and white. The gray matter functions as the computers of the brain and the white matter serves as the cables, or “tracts”, that connect those computers, allowing them to communicate. While the shape of grey matter regions of the brain have been studied extensively, the shape of the white matter tracts have not received much attention in research communities; Lindsey Kitchell, a student in the Department of Psychological and Brain Sciences and the Cognitive Science Program, would like to change that.

Lindsey Kitchell, IU PhD student, with 3D models of the human brain

Kitchell’s work, in the laboratory of Dr. Franco Pestilli, focuses on developing methods of examining the morphology (or, shape) of the brain’s white matter tracts. She is particularly interested in measuring changes in the tracts’ shapes in relation to cognition and neurological disorders, which can cause changes to their micro- and macrostructure. To this end, Kitchell has developed a shape analysis pipeline that quantifies the shape of individual white matter tracts. This shape information, combined with machine learning, can accurately predict if an individual has been diagnosed with bipolar disorder or is a control subject. Kitchell recently presented this work at the Shape in Medical Imaging workshop at the 21st international conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in Granada, Spain. In tandem with this, Kitchell also researches the natural variability of white matter morphology at a population level. She and her collaborators have analyzed data obtained from the white matter tracts of 1,000 healthy human brains, and are working toward determining a clear definition of normal morphological variability and variability caused by diseases.

Kitchell works using the brainlife.io platform, and credits access to IU’s supercomputers Karst and Carbonate used through the platform with improving the pace of her data analysis, significantly. With Karst and Carbonate, she reports being able to “process the over 1,000 brains mentioned above in weeks, while it would have taken months without the supercomputers.” Karst Desktop provides an additional benefit.

I can leave my work up and return to it at any moment from nearly anywhere; it is an easy interface to the supercomputers.

Lindsey Kitchell

Much of Kitchell’s data processing is done in collaboration with brainlife.io, an online neuroimaging platform developed at IU by Kitchell’s mentor, Franco Pestilli. She is a core team-member of brainlife.io, the platform uses cloud and high-performance computing systems (including Karst and Carbonate) flexibly to accelerate neuroscience discovery to promoting open data sharing and making it easier for experts to work with neuroimaging data. Ultimately, Kitchell hopes that her research will open new avenues of investigation by integrating white matter morphology into existing neuroimaging analyses and by providing new morphological biomarkers to offer a broader, clearer picture of a variety of neurological disorders, such as Bipolar disorder, ADHD, Schizophrenia, and Alzheimer’s disease. To read more about working being done by Dr. Pestilli and Ms. Kitchell, see the lab website here.

Kitchell and Pestilli’s work and brainlife.io are supported by grants by the BRAIN Initiative through awards from National Science Foundation (NSF IIS 1636893 and BCS 1734853), as well as IU’s Emerging Areas of Research Program, the College of Arts and Sciences, and the Pervasive Technology Institute.