We can continue to generalize the concept of distance to the distribution of the 3D dose, because the distribution of the dose can also be considered as a surface. For two distributions of 3D D1 and D2 doses (standardized doses are always used, unless otherwise stated), the interval from one point to D1 (with dose of points D1 (x1, y1, z1)) to D2: The Truth and Performance Level Estimation Algorithm (STAPLE) uses a probability map to create a „better fit” from a collection of contours (Figure 1) . The STAPLE algorithm has created a standard of reference from all the clinician`s manual outlines (`STAPLE`). The modified manual, SPICE and SPICE contours were compared to STAPLE by: cube seam coefficient (DSC) and average/maximum distance from chord (DTA). The DSC is a statistical measure of spatial overlap between two structures. It is defined as 2x total volume/volume and normalizes the cut degree from 0 (no overlap) to 1 (perfect ride), with a good match defined as >0.7-0.8 [41,51.52]. The DTA is a geometric parameter that measures the shortest voxel distance from one structure to another, ideal for 0 mm. The mated structures (ear glands, submandibular and cochlear glands) were considered together. For salivary glands and submandibular glands, SPICE produced three contours (`1`, `2` or `3`), each based on different „soil truth data” . Comparisons between these and STAPLE for all 10 patients were made to determine the most accurate for subsequent application and evaluation. The study was carried out with appropriate local research and development approval. There are three sources of uncertainty (difference) in comparing a dose distribution in the treatment system with actual measurements: (1) Uncertainty in the configuration of the ghost; (2) uncertainty (quantum noise) in the dosimetry system (for example. B film in our study); and (3) uncertainty due to discrepancies between the photon beam modeled in the processing system and the photon beam delivered.
The uncertainty of adjustment at the x-y level results in spatial shift (global) and the uncertainty of adjustment in the Z direction results in a dose difference (global), the level provided by the treatment system being different from the measurement plan. The quantum noise of the dosimetria system can be considered a random shift of the dose axis and therefore has a global effect. As noted in Low and Dempsey3 and presented in table TABLEAU II, the measured amounts (z.B average marginal distance and D90) are changed at low noise levels (2%) this is not essential, but if the sound level is high enough, for example. B 5%, if only 68% of pixels should fall into the ±5%, even without spatial or dose displacement. In the latter case, certain quantities (for example. B D99) are no longer suitable for evaluation. The uncertainty of primary interest in imRT quality assurance could be both global (changes in speed/symmetry/exit) and local (MLC/pin) and the change could be in both space (pines/MLC) and dose axles (flat/symmetry/output-change). The uncertainty of the installation and the noise of the dosimetrie system26 should, as far as possible, be inferred from the measured overall uncertainty and the inadequacy of the model could then be grouped with other delivery uncertainties. B for example configuration uncertainties, and the use of processing margins.