(Sunday, 7/29/2018) 5:05 PM - 6:00 PM
Room: Room 209
Purpose: The purpose of this research is to develop effective data integrity models for contoured anatomy in a radiotherapy workflow for both real-time and retrospective analysis.
Methods: Within this study, two classes of contour integrity models were developed: data driven models and contiguousness models. The data driven models aim to highlight contours which deviate from a gross set of contours from similar disease sites and encompass the following regions of interest (ROI): bladder, femoral heads, spinal cord, and rectum. The contiguousness models, which individually analyze the geometry of contours to detect possible errors, are applied across many different ROI’s and are divided into two metrics: Extent and Region Growing over volume.
Results: We found that 70% of detected bladder contours were verified as suspicious. The spinal cord and rectum models verified that 73% and 80% of contours were suspicious, respectively. The contiguousness models were the most accurate models with a 0% false positive rate. 100% of the non-contiguous contours detected by the Extents model were verified as suspicious, however the Region Growing model detected two to five suspicious contours which the Extents model did not in most regions of interest. The Region Growing model produced zero false negatives in all regions of interest other than prostate. The contiguousness via extent model took an average of 0.2 seconds per contour while the runtime of the region growing method was dependent on the number of voxels in the contour.
Conclusion: Both contiguousness models are suited for real time use in clinical radiotherapy while the data driven models are also suited for quality assurance in treatment planning. Contour integrity systems should become integrated into the clinical planning to process to ensure only verified contours are treated, minimizing the risk of erroneous dosing to critical anatomy and improving the safety of radiation therapy.
Keywords
Not Applicable / None Entered.
Taxonomy
IM/TH- image segmentation: General (Most aspects)
Contact Email
