A computerized remote table control for fast on-line patient repositioning: Implementation and clinical feasibility

A computerized remote control for a Siemens ZXT treatment couch was impleme nted and its characteristics were investigated to establish its feasibility for on-line setup corrections, using portal imagine. Communication with th e table was obtained by connecting it via a serial line to a work station. The treatment couch enables "goto" commands in the three main directions an d around the isocenter. The accuracy of the movements after giving such a c ommand was checked and the time for each movement was recorded. First, the movements into a single direction were studied (range of -4 to +4 cm and -4 degrees to +-4 degrees). Each command was repeated four times. Second, the table was moved into the three main directions simultaneously. For this ex periment a clinically relevant three-dimensional (3-D) normal distribution of shifts was used [N=200, standard deviation (SD) 5 mm in the three main d irections]. This latter experiment was done twice: without and with rotatio ns (a distribution with SD 1 degrees). During the first experiment, with sh ifts into one direction, no systematic deviations were found. The overall a ccuracy of the shifts was 0.6 mm (1 SD) in each direction and 0.04 degrees (1 SD) for the rotations. The time required for a translation ranged betwee n 4 and 13 s and for the rotation between 8 and 20 s. The second experiment with the 3-D distribution of setup errors yielded an error in the 3-D vect or length equal to 0.96 mm (1 SD), independent of rotations. Shifts were pe rformed in less than 11 s for 95% of the cases without rotations. When rota tions were also performed, 95% of the movements finished in less than 16 s. In conclusion, the table movements are accurate and enable on-line setup c orrections in daily clinical practice. (C) 2000 American Association of Phy sicists in Medicine. [S0094-2405 (00)00502-2].