The Dwork laboratory at CU Anschutz focuses on small data exploitation for computational imaging and radiation oncology. Applications include medical image reconstruction, image quantitation, radiation treatment planning, and inexpensive medical devices. The aim of our work is to increase speed and robustness for timely, reliable, clinically applicable results. The speed is increased by constraining the solution with known physics, thorough algorithm development, and effective use of the computer architecture. Robustness is achieved with theoretical guarantees and empirical confirmation.
The principal investigator of the lab is Nicholas Dwork.
Our goal is to get the image to the clinician for diagnosis and staging as quickly as possible. To this end, we are developing several technologies that permit faster scanning:
By combining these technologies, we hope to increase the utility of the MR imaging system for small children, the placenta, and fetuses.
MR spectroscopy provides molecular information that is spatially localized. By injecting a contrast agent, we can see where it goes and what metabolites are created. This provides an unprecedented capability for non-invasively diagnosing and staging cancer and metabolic disorders. The spectroscopy adds a dimension to the space of information that must be understood, but the contrast agents either provide a low signal-to-noise ratio or are limited in time. The Dwork lab uses the structure of the imaging system and the metabolites to provide clinicians with information rich images.
Radiation therapy is a treatment for cancer where energy is deposited in the body to cause tumor necrosis. The goal of radiation therapy is to spare any organs-at-risk while depositing high energy to the target volume. The Dwork Lab is working to create effective treatment plans even when the patient's anatomy changes during treatment while also reducing the overall time of treatment for improved comfort and effectiveness.