Computing Science

Best paper award recognizes MRI advances

August 05, 2011
Print

PhD student Brian Booth (pictured above) and supervisor Ghassan Hamarneh won the Best Paper Award at the 2011 IEEE International Conference on Healthcare Informatics, Imaging and Systems Biology (HISB) in San Jose, California.

Booth and Hamarneh are part of SFU's Medical Image Analysis Lab, where they work on clinical healthcare issues by advancing imaging technology. Their paper, Consistent Information Content Estimation for Diffusion Tensor MR Images, introduces a new way to measure entropy, or the amount of information available in data, for Diffusion Tensor MRI medical images. The importance of measuring entropy (data) in medical imaging is vast as it helps to register images and segment parts of the anatomy.

Diffusion Tensor MRI (DT-MRI) is one of the most recent medical imaging techniques and allows clinicians to capture more detailed connectivity information in the brain's white matter. Unlike traditional images that have a grayscale or colour (red, blue, and green) value at each pixel, DT-MRI contains, at each pixel, a more complicated structure: a second order tensor.

In their award-winning paper, Booth and Hamarneh proposed the first type of measurement of entropy for DT-MRI data that makes use of the complete second order tensor without loss of information. Their improved entropy measure can then be used to improve the accuracy of existing algorithms that rely on entropy to align images or perform object detection.

Booth holds an NSERC doctoral scholarship and won the Sir James Lougheed Award of Distinction in 2012. He is also the first SFU student to win an IODE War Memorial Scholarship, one of only six PhD students out of 111 students across Canada to be recognized with the award.

The focus of the HISB conference is to foster an integrated approach to healthcare research among people working in healthcare informatics, imaging and systems biology. The integration of these fields is expected to lead to a better understanding of diseases.