Laboratory Notes

Computational Anatomy

Polina Golland

Computer Science and Artificial Intelligence Laboratory (CSAIL)

Understanding how the brain develops and how diseases affect its anatomy and function pose intriguing research questions. Magnetic Resonance Imaging (MRI) provides high resolution scans of brain structure. With proper interpretation techniques, images taken over an extended period of time can shed light on how the anatomy of the brain changes in normal development and under the influence of disease. Identifying such trends in turn leads to better understanding of the disorders and to improved diagnosis and treatment. Our research focuses on the computational modeling techniques that enable such descriptions from MRI images.

Today’s image-based brain studies typically compare two or more sets of brain images with a goal of identifying differences between the populations represented by the subjects in the study. For example, the shape of subcortical structures, such as the hippocampus, or the cortical folding patterns in the scans of schizophrenia patients would be compared to a set of brain images of matched normal controls. The analysis starts with a feature extraction step, generating a numerical description of the example images.

A follow up statistical analysis generates a description of the differences between the two groups and an estimate on how well the detected differences will apply to the entire population. In addition to the technical challenges of very complex shapes and a limited number of examples, a large amount of natural variation in the anatomy within the normal population makes the analysis quite difficult, requiring novel methods to incorporate knowledge of the domain into models.

To be useful in a clinical context, the resulting statistical model must be mapped back to the image domain so that it can be related to other anatomical and physiological information. We have developed a sensitivity analysis that enables an explicit visual interpretation of the detected differences in biologically meaningful terms of deformations of the anatomy that would make the two populations indistinguishable from each other.

We work with researchers at Brigham and Women’s Hospital, Mass General Hospital and Washington University, St. Louis, applying these methods for studying the neuroanatomy of schizophrenia and of healthy aging versus the progression of Alzheimer’s disease. We believe that ultimately this research will improve our understanding of how a particular disease affects brain structure, leading to new, more effective therapies.

Pages: 1 2 3 4 5 6

One Response to “Laboratory Notes”

  1. I am impressed with this website , rattling I am a fan .

Leave a Reply

Replies which add to the content of this article in the EECS Newsletter are welcome. The Department reserves the right to moderate all comments. If you would like to provide any updated information please send an email to newsletter@eecs.mit.edu.