Daily patient geometry correction: application of NAL and eNAL protocols

Purpose: To test the NAL and eNAL correction protocols using daily patient setup displacements.

Methods and material: In total, the analysis was performed for 749 and 797 kV CBCT images for gynecological and prostate patients, respectively, each of 30 patients. After the planning procedure, patients were set up on the treatment table in the treatment position every day. The on-line correction protocol was applied. KV CBCT images were acquired by means of x-ray lamp mounted orthogonally on Linac. Patient setup displacement was assigned. NAL and eNAL corrections protocols were simulated using daily data from online corrections for these two groups of patients. The overall systematic error and random error were calculated for each direction.

Results: For the prostate group, the random errors for daily Raw data (no correction) in LAT, LONG, and VERT directions were 2.0 mm, 1.6 mm, and 3.2 mm, respectively. For NAL and eNAL protocols, they were in the range of 1.8 mm to 3.2 mm. For the gynecological group, the random errors were: for daily Raw data 2.2 mm, 1.7 mm, and 3.2 mm, respectively. For NAL and eNAL protocols, they were in the range of 2.0 to 3.4 mm.

For the prostate group, values of systematic errors 1.8 mm, 1.8 mm, and 3.3 mm, respectively for Raw data. For NAL and eNAL protocols, these values were less than 1.8 mm. For the gynecological group, the systematic errors were 2.6 mm, 2.3 mm, and 2.8 mm, respectively, for Raw data. For NAL ana eNAL protocols less than 1.8 mm.

For the gynecological group, for Raw data, 45% of the total displacement vectors exceeded 5 mm, whereas only 25% did after the NAL procedure and 29% after the eNAL procedure. For the prostate group, for Raw data, 34% of the total displacement vectors exceeded 5 mm, whereas only 22% did after NAL procedure and 28% after eNAL procedure Conclusions: For gynecological and prostate cancer patients, the NAL and eNAL correction protocols can be safely applied to substantially reduce setup errors.