Creating Biomedical Technologies to Improve Health

2018 BESIP Project

Physics Group, Department of Radiology and Imaging Sciences
Mentor Name: 
Roberto Maass-Moreno, Ph.D
Mentor Email: 
Mentor Telephone: 
(301) 402-8983

Laboratory and Project Description

Breakthroughs in computer technology and imaging physics have led to spectacular advances in medical imaging. The resulting new imaging tools have played no small part in many of the improvements in diagnosis and treatment of common and complex diseases - particularly cancer.
Our summer interns have contributed to the progress in this area by designing new image segmentation methods, methods for accurate radiation dose estimation using Monte Carlo simulations, new image reconstruction methods and image enhancement and analysis procedures. Some of our previous interns have presented and received awards for their work at international conferences and some have found their projects exciting enough to put medical imaging in their career plans. If you have computer, digital image analysis, radio-biology, bioengineering or nuclear physics skills/interests then our projects may be of interest to you. Experience in script languages such as Matlab, IDL and the like would be advantageous but not essential. In our Division we are involved mostly with Positron Emission Tomography (PET), Single Photon Emission Tomography (SPECT) and multi-modality PET/CT, PET/MRI and SPECT/CT scanners.

This summer, our projects will involve the study of methods to characterize the quantitative content of PET diagnostic images.   The theory indicates that calibrations and scatter plus attenuation corrections for PET images should provide accurate quantifiable images.  However, in an effort to appeal to the visual quality and human detectability of lesions, processing is applied to the images that may compromise its quantitative value.   In addition, dose and time constraints may also impact quantification.  The question we want to answer is whether there is a reproducible and efficient way to assess the effect of filters or other resolution and contrast enhancement procedures on the quantitative quality of a diagnostic PET image.  Having such quantitative measures would allow a more rational design of reconstruction, acquisition and processing protocols.