Filter Results

Clinical Studies

Contact Us for the Latest Status

  • Evaluation of Organ at Risk Segmentation for Prostate Radiotherapy Utilizing a UNET Variant Architecture (OAR-AI) Scottsdale/Phoenix, Ariz., Jacksonville, Fla., Rochester, Minn.

    The purpose of this study is to determine:  if the artificial intelligence (AI)-generated results are less arduous than manual tracing by radiation oncologist, and the non-inferiority of the quality of AI vs. manual tracing.  These aims will be evaluated in a clinical environment to investigate the impact of an AI algorithm on the clinical workflow. 

    Radiotherapy treatment planning requires precise calculations of radiation exposure, not only for the target volumes (tissue containing malignancy), but of nearby organs-at-risk (OARs), in which the exposure needs to be minimized. Manual segmentation of these organs is a time-consuming task with high interobserver variability. Producing these segmentations automatically will reduce the time required for treatment planning and improve the interobserver variability.
.