Duke University Medical Center
Nuclear Medicine
As a senior in college I worked as a volunteer software engineer at Duke Medical Center under the supervision of Drs. Goodrich and
Harris.
I wrote software in Nutran on a Nuclear Data 12-bit computer using a teletype machine and paper tape.
Major accomplishments included:
Coronary Artery Disease:
We developed a method for screening candidates for coronary angiography by imaging the heart using Thallium-201.
I wrote software to automatically acquire image data, analyze the image, remove background noise, and clearly present a picture of the
heart muscle tissue.
This became a diagnostic clinical procedure.
Nuclear Medicine Imaging:
In just a few months, I wrote software to acquire and process images for diagnostic imaging of
the brain, lungs, and kidneys using the Nuclear Data imaging system.
This leading-edge technology became a part of the suite of clinical diagnostic nuclear imaging
studies.
Upon completion of this work, I was appointed a staff position as research associate.
Radiology
I provided technical support for all departments within radiology.
I conducted research to improve the image quality of Single Photon Emission Computed
Tomography under the supervision of Dr. Ronald Jaszczak.
My research further supported diagnostic imaging in other departments.
A few highlights are presented below.
Nuclear Cardiology
The mission of this research was to develop a mathematical model of blood flow through the
central circulatory system.
Determined cardiac parameters non-invasively using nuclear imaging techniques.
Data acquisition involved detection of gamma rays emitted from a source within the human body.
Radiopharmaceuticals labeled with low energy radioisotopes were intravenously injected in the
external jugular vein.
The radiopharmaceutical tracer follows a venous pathway to the right heard, through the pulmonary
system and the left heard, then continues through the peripheral circulation.
A scintillation camera was used to image the flow of the tracer and to provide a time sequence of
image data to a computer.
A set of mass balance equations were written to model the radiotracer concentration in therms of
flow rate and volume.
The simulated data was fit to experimental data, varying the cardiac parameters to obtain a
minimum of the sum of the squares of the differences between the simulated and actual data.
The results of the model were compared with other methods of measurement.
This technique was commercialized and is now part of the data processing for modern nuclear cardiac stress tests.
Emission Computed Tomography
The mission of this research was to develop an improved method for data reduction, analysis, and
presentation of Single-Photon Emission Computed Tomographic (SPECT) images.
A Monte Carlo computer simulation to study the emission and transport of gamma
radiation through tissue phantoms.
Developed computed tomographic geometry particular to the SPECT simulation
application of variance reduction techniques.
Developed an energy dependent weighting function that reduces the degrading effects of
scatter and attenuation while decreasing the measurement time for a given diagnostic study.
Useful to illustrate and evaluate other methods of improving image quality.
The image reconstruction and processing techniques are incorporated in modern diagnostic emission tomographic systems.
| 
