ESTRO 2024 - Abstract Book

S5963

RTT - Treatment planning, OAR and target definitions

ESTRO 2024

Fig.1

In most cases, radiographer volumes were smaller than clinician volumes, with only one case where both radiographers contoured larger by 10-22%. p-values for absolute volumes were: CTVp 0.027 and 0.013, CTVpsv 0.005 and 0.009, CTVsv <0.001 and 0.010 (radiographer 1 and 2 respectively). The radiographers’ DICE scores across CTVp, CTVpsv and CTVsv were good (mean 0.84-0.86, range 0.78 to 0.90) comparing to the clinical oncologist volumes. DICE scores between radiographers were similarly high; mean 0.85 0.86, range 0.78-0.90 (Fig. 2).

Fig. 2

The mean distance to agreements were concise with mean values as follows for CTVp (1.34mm, 1.40mm), CTVpsv (1.33mm, 1.32mm) and CTVsv (1.29mm, 1.23mm), with values ranging from 0.85mm to 3.16mm.

Two cases were replanned; one with DICE scores closest to both radiographers’ mean values, and one where radiographer volumes were larger than the clinical oncologist’s. All OAR constraints were met, and clinically acceptable plans were achieved. For the DICE mean matched plan, coverage to the original doctor volumes was within minor deviation tolerance at 91.8%, despite the radiographer volumes being smaller.

Conclusion: