Premium
TU‐AB‐BRA‐08: Slice by Slice Approach to Quantifying Inter‐Fractional Organ Motion
Author(s) -
Mazur T,
Cardenas M,
Li H,
Mutic S,
Bradley J,
Tsien C,
Green O
Publication year - 2016
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4957418
Subject(s) - centroid , margin (machine learning) , boundary (topology) , mathematics , orientation (vector space) , contouring , geometry , intersection (aeronautics) , esophagus , nuclear medicine , algorithm , computer science , mathematical analysis , anatomy , medicine , computer graphics (images) , machine learning , engineering , aerospace engineering
Purpose: To demonstrate the feasibility of a rapid computational algorithm for generating esophageal margin expansions. Methods: Inter‐fraction MRI allows visualization of relative variations of the esophagus and other critical structures when compared to planning CT. We have developed a rapid computational method for quantifying this motion. CT simulation and on‐treatment MR data sets of seven patients were obtained using a 0.3 T MR‐guided RT system. Rigid registration guided by bony anatomy from the cricoid to the gastro‐esophageal junction was used for transferring CT contours to MR volumes. CT‐ and MR‐based contours were re‐sampled to form boundaries – BCT and BMR respectively – that densely sample the image grid. Centroids CCT and CMR were calculated slice‐by‐slice from BCT and BMR respectively. For each CT boundary point BCT(i) within an area enclosed by the MR boundary points, the intersection BMR(j) was calculated between the MR boundary and a line defined by BCT(i) and CCT. The length of the vector defined by BCT(i) and BMR(j) provides a margin expansion within a given slice for that vector orientation. This procedure is repeated for every slice to characterize margin expansions along the esophagus. Results: Expansions were determined for in‐plane quadrants along the length of the esophagus. With the analysis relying on simple geometry, the non‐negligible time constraint in the procedure is manual contouring. Given the soft tissue contrast on MR volumes, systematic margin uncertainties are ultimately determined by in‐plane spatial resolution (1.5×1.5 mm2 in most cases). Conclusion: We demonstrate a rapid method for assessing inter‐fractional variation for the sake of suggesting thorough treatment margins (applied to the esophagus). Ongoing work aims to assess intra‐fractional esophageal motion.