Using Surface-Based Imaging for Patient Identification
B Sintay*, D Wiant, Cone Health Cancer Center, GREENSBORO, NCSU-E-J-73 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall
Purpose: A feasibility study to determine if daily surface-based imaging can be used for patient identification with reliable accuracy.
Methods: Three-dimensional surface images were acquired using AlignRT (VisionRT Ltd., London, UK) prior to daily radiotherapy treatment for 12 breast cancer patients. These were compared retrospectively to body contours generated from computed tomography simulation (CT sim) scans. Two types of comparisons were made. The first compared the surface of each patient for all fractions of treatment. The second compared a random incorrect patient body contour for each patient in the study. The 'surface statistic' tool reveals the percentage of surface points within a threshold distance of 3mm.
Results: Comparison of the patients' daily AlignRT images to the CT sim body contour showed similarity of the surfaces of 81.0% +/- 6.6% with a total range of 61.5-95.3%. This same comparison of the wrong patient's CT sim body contour to a random fraction of another patient's AlignRT image revealed similarity of the surfaces of 27.9% +/- 10.2% with a range of 6.7-58.3%. A threshold of approximately 60% separates misidentification from correct identification in our limited study set. There was no overlap between the two groups in our test. However, we recognize that the two groups come close to the threshold value and could overlap in a larger study set.
Conclusions: Using daily surface-based imaging to identify patients is feasible with AlignRT using a surface statistic threshold of approximately 60%. This has the possibility of improving radiotherapy safety using existing image-guided technology. Further study is needed to better understand the sensitivity and specificity of this tool for a larger patient population.