Pictured above: Ronald Roman, Alexander Kennedy, and Max Xiong work together to analyze an MRI of the brain, using software to build a model of the hippocampus.
This summer, high school students from the Salt Lake area came to the University of Utah to participate in hands-on research in image analysis of the human brain. In conjunction with graduate students and faculty from the School of Computing and the Scientific Computing and Imaging Institute, the students learned how computer science can help neuroscience researchers understand the brain and the disorders that affect it, such as Alzheimer’s disease and autism.
Advances in medical imaging devices, such as magnetic resonance imaging, or MRI, have led to our ability to acquire detailed information about the living human brain, including its anatomical structure, function and connectivity. However, making sense of this complex data is a difficult task, especially in large imaging studies that may include hundreds or even thousands of participants. This is where computer science can play an important role. Image analysis algorithms can automatically quantify properties of the brain, such as the size of brain structures, or the functional activity in different brain regions. This provides neuroscience researchers with insights into how the brain functions and what abnormalities are present in diseased brains.
The students in the program used state-of-the-art image analysis software tools to analyze MRIs from real-world brain imaging studies. For example, the students used ITK SNAP to build a 3-D model of the hippocampus from brain images of patients with Alzheimer’s disease. The students then used statistical analysis to see how the volume of this structure, important in memory function, decreases by using publicly-available MRI data from the Open Access Series of Imaging Studies. In another project, the students explored the functional connectivity between different regions of the brain, using functional MRI, or fMRI. The data comes from the Autism Brain Imaging Data Exchange. Students used machine learning tools to explore differences between functional connectivity in subjects with typically developing brains and subjects with autism spectrum disorder.