[en] A method of computing optimised intensity-modulated beam profiles has been further developed and used to generate highly conformal radiotherapy dose distributions, the features of which are strongly dependent on the tuning built into the algorithm. The optimisation aims to achieve a specified dose prescription. The effect of stratifying the intensity-modulated beam profiles into a number of finite intensity increments is that, provided the number of intensity strata is not too small, highly conformal dose distributions can be achieved with a number of fields (e.g. 15) which is not excessively large. This number, however, depends on the exact shape of the planning target volume (PTV) and its disposition with respect to juxtaposed organs at risk (OARs). These intensity-modulated profiles can therefore be delivered either by apparatus for 'Tomotherapy' or by using the multileaf collimator at each gantry orientation to deliver a sequence of fixed fields with different field sizes, constructing the beam profile via finite increments of beam intensity. When the PTV and OARs overlap, due to including a finite margin on the clinical target volume to account for tissue movement, the dose delivered to the overlap region provides a limit on what can be achieved with conformal therapy

[en] A model for computing tumour control probability (TCP) was studied which embodies a dependence on the size V of the irradiated volume, the density p of clonogenic cells and on the dose D received, together with the α parameter of the Linear Quadratic model of cell kill. There is still uncertainty about its radiobiological parameters and this work aimed to establish parameters for four types of tumour. The model was fitted to 4 published datasets of observed clinical TCP (V, D, clin). For each of the 4 tumour types these data comprise observations for n pairs of (V, D) with D uniform. The model for predicted TCP (V,D,pred), including the effects of interpatient heterogeneity in radiosensitivity α is given by where k is (i) 1 for melanoma (M), (ii) 1.2 for breast (B) and squamous cell carcinoma of the upper respiratory and digestive tracts (S) and (iii)k = 1 + D/300 for nodal control in the upper respiratory and digestive tracts (N). α(i) is a value selected (for each contribution to the average over the population, labelled by i) from the Gaussian distribution with mean α₀ and standard deviation σ_α. K = 40000 samples across the population were averaged. The cost function was minimised by a sequential search of (α₀,σ_α) space. The optimum pair of (α₀,σ_α) values depends on the assumed value of p. The best fitted values with p = 10⁷ are given

[en] This presentation starts with implementation details of an inverse planning optimisation algorithm we have developed to optimise radiation beam-weights for radiotherapy treatment plans. The algorithm is based on fast-simulated-annealing, utilising a cost-function containing both linear and quadratic terms. General conclusions concerning the efficacy, usability and practicality of the algorithm will be presented. We have applied the algorithm extensively to the problem of optimising prostate treatment plans. The results of these investigations will be summarised and it will be shown that the algorithm can reproduce, and improve on, the results of an experienced human planner. During our investigations it was found that significant lowering of the dose to that part of the rectum outside the PTV produced only minor changes in normal-tissue-complication-probability (NTCP). This prompted an investigation into the 'overlap effect'; i.e. that conformal radiotherapy of the prostate is inherently limited by the size of the overlap region of the PTV and rectum. The NTCP of the rectum is due almost entirely to the high dose rectal tissue in the overlap region. The results of this investigation will be presented. The conclusion is that until the margin accounting for set-up errors that is added to the CTV can be reduced, or new methods are developed to deliver controlled non-uniform dose to the PTV, conformal therapy of the prostate is inherently limited by the overlap effect

[en] A radiotherapy treatment plan optimisation algorithm has been applied to 48 prostate plans and the results compared with those of an experienced human planner. Twelve patients were used in the study, and a 3, 4, 6 and 8 field plan (with standard coplanar beam angles for each plan type) were optimised by both the human planner and the optimisation algorithm. The human planner 'optimised' the plan by conventional forward planning techniques. The optimisation algorithm was based on fast-simulated-annealing. 'Importance factors' assigned to different regions of the patient provide a method for controlling the algorithm, and it was found that the same values gave good results for almost all plans. The plans were compared on the basis of dose statistics and normal-tissue-complication-probability (NTCP) and tumour-control-probability (TCP). The results show that the optimisation algorithm yielded results that were at least as good as the human planner for all plan types, and on the whole slightly better. A study of the beam-weights chosen by the optimisation algorithm and the planner will be presented. The optimisation algorithm showed greater variation, in response to individual patient geometry. For simple (e.g. 3 field) plans it was found to consistently achieve slightly higher TCP and lower NTCP values. For more complicated (e.g. 8 fields) plans the optimisation also achieved slightly better results with generally less numbers of beams. The optimisation time was always ≤5 minutes; a factor of up to 20 times faster than the human planner

[en] A CT scanner comprises a rotatable gantry carrying an X-ray source which provides a fan-shaped beam of X-rays and an X-ray detector which scans the X-ray intensity along a line in the plane of the beam. The gantry is rotatable about an axis relative to a couch which supports the patient under investigation. The X-ray detector comprises two concentric X-ray masks exposed around a scintillator tube, the two masks have helical X-ray transmissive apertures, the helices being of opposite senses so that rotation of the masks in opposite directions causes the exposed area of the scintillator tube to move along its length. (author)

[en] A method for computing optimised wedge angles will be presented along with a performance evaluation over 12 patients with early prostate cancer. The method (an optimisation algorithm) was applied to standard 3 field treatment plans for each patient. The optimisation algorithm was based on simulated anealing using an efficient dose based cost function. The algorithm has been run in three PLAN MODES: (1) where the wedge angles were fixed by the human planner and only the beam-weights were optimised; (2) where both the wedge angles and beam-weights were optimised; and (3) where both the wedge angles and beam-weights were optimised and a non-uniform dose was prescribed to the PTV. In the latter PLAN MODE, a uniform 100% dose was prescribed to all of the PTV except for that region that overlaps with the rectum where a lower (e.g. 90%) dose was prescribed. The resulting optimised plans have been compared with those of the human planner who found beam-weights by conventional forward planning. Plans were compared on the basis of dose statistics, normal-tissue-complication-probability (NTCP) and tumour-control-probability (TCP). The results show that all 3 PLAN MODES produced plans with slightly higher TCP for the same rectal NTCP, than the human planner. The best results were observed for PLAN MODE 3, where a non-uniform PTV dose was prescribed. An average increase in TCP of 0.73% (± 0.20 95% confidence interval) is predicted. Probably the most significant benefit of the algorithm in the prostate setting is the time saved (about a factor of 10) in computing optimised beam-weights and wedge angles for this simple plan

[en] A radiotherapy treatment plan optimisation algorithm has been applied to 48 prostate plans and the results compared with those of an experienced human planner. Twelve patients were used in the study, and 3-, 4-, 6- and 8-field plans (with standard coplanar beam angles for each plan type) were optimised by both the human planner and the optimisation algorithm. The human planner 'optimised' the plan by conventional forward planning techniques. The optimisation algorithm was based on fast simulated annealing using a cost-function designed to achieve a homogenous dose in the 'planning-target-volume' and to minimise the integral dose to the organs at risk. 'Importance factors' assigned to different regions of the patient provide a method for controlling the algorithm, and it was found that the same values gave good results for almost all plans. A study of the convergence of the algorithm is presented and optimal convergence parameters are determined. The plans were compared on the basis of both dose statistics and 'normal-tissue-complication-probability' (NTCP) and 'tumour-control-probability' (TCP). The results of the comparison study show that the optimisation algorithm yielded results that were at least as good as the human planner for all plan types, and on the whole slightly better. A study of the beam-weights chosen by the optimisation algorithm and the planner revealed differences that increased with the number of beams in the plan. The planner was found to make small perturbations about a conceived optimal beam-weight set. The optimisation algorithm showed much greater variation, in response to individual patient geometry, frequently deselecting certain beams altogether from the plan. The algorithm is shown to be a useful tool for radiotherapy treatment planning. For simple (e.g., three-field) plans it was found to consistently achieve slightly higher TCP and lower NTCP values. For more complicated (e.g., eight-field) plans the optimisation also achieved slightly better results with generally less numbers of beams, unfavourable beams being deselected from the plan. Probably the greatest benefit is the reduced time taken by the optimisation to compute optimised beam-weights. This time was always ≤ 5 min; a factor of up to 20-times faster than the human planner

[en] A treatment plan optimisation algorithm has been applied to 12 patients with early prostate cancer in order to determine the optimum beam-weights and wedge angles for a standard conformal three-field treatment technique. The optimisation algorithm was based on fast-simulated-annealing using a cost function designed to achieve a uniform dose in the planning-target-volume (PTV) and to minimise the integral doses to the organs-at-risk. The algorithm has been applied to standard conformal three-field plans created by an experienced human planner, and run in three PLAN MODES: (1) where the wedge angles were fixed by the human planner and only the beam-weights were optimised; (2) where both the wedge angles and beam-weights were optimised; and (3) where both the wedge angles and beam-weights were optimised and a non-uniform dose was prescribed to the PTV. In the latter PLAN MODE, a uniform 100% dose was prescribed to all of the PTV except for that region that overlaps with the rectum where a lower (e.g., 90%) dose was prescribed. The resulting optimised plans have been compared with those of the human planner who found beam-weights by conventional forward planning techniques. Plans were compared on the basis of dose statistics, normal-tissue-complication-probability (NTCP) and tumour-control-probability (TCP). The results of the comparison showed that all three PLAN MODES produced plans with slightly higher TCP for the same rectal NTCP, than the human planner. The best results were observed for PLAN MODE 3, where an average increase in TCP of 0.73% (± 0.20, 95% confidence interval) was predicted by the biological models. This increase arises from a beneficial dose gradient which is produced across the tumour. Although the TCP gain is small it comes with no increase in treatment complexity, and could translate into increased cures given the large numbers of patients being referred. A study of the beam-weights and wedge angles chosen by the optimisation algorithm revealed significant inter-patient variability the implications of which are examined. Probably the most significant benefit of the algorithm is the time saved (about a factor of 10) in computing optimised beam-weights and wedge angles for this simple plan

[en] Background and purpose: An optimization algorithm has been developed to determine the best beam-arrangement for a small number of intensity-modulated radiotherapy (IMRT) fields. The algorithm is designed to avoid, if possible, beam-orientations that pass through organs-at-risk (OARs) with low radiation tolerance. Materials and methods: An independent, fast IMRT algorithm based on the Bortfeld algorithm was developed to determine the profile of the intensity-modulated beams (IMBs) for each beam-arrangement and a fast-simulated-annealing algorithm finds the 'optimal' beam-arrangement. The final beam-arrangement was transferred to the CORVUS (NOMOS Corporation) treatment planning system, and the IMBs were re-optimized for comparison with a standard nine-field, equi-spaced arrangement. The algorithm has been initially tested on a single example patient, with a parotid gland carcinoma. Results: The nine-field, IMRT plan for an example patient with a parotid gland tumour significantly reduced the dose to the cochlea compared with the conformal radiotherapy plan. In addition, the planning-target-volume (PTV) homogeneity was improved, but the plan produced a higher dose to the contralateral parotid (73% of the OAR received more than 6 Gy). The beam-orientation optimization algorithm produced a three-field plan that greatly reduced the dose to the contralateral parotid (maximum dose of 2 Gy), whilst maintaining the PTV dose homogeneity and the reduced cochlear dose of the nine-field plan. Some changes in the dose to the other OARs, namely the brain and the oral cavity, were seen, but were deemed not to be clinically significant. Conclusions: In conclusion, IMB-orientation optimization for head and neck treatment sites can produce improvements in treatment plans with only a few fields

[en] Background and purpose: A comparison between three classes of intensity-modulated delivery techniques was undertaken to examine the dosimetric consequences of using a multileaf collimator (MLC) reshaped on each imaged fraction as opposed to compensators designed on the first day of treatment potentially giving a treatment technique whose accuracy is thus degraded by movement. Materials and methods: The effects of inter-fractional patient movement for a cohort of six breast patients were studied. Five treatment techniques were evaluated, two using a compensator, two using multiple static fields (MSF) and one using a dynamic multileaf collimator (DMLC). The compensated techniques consisted of (i) the use of compensators designed on day 1 only and used each fraction thereafter and (ii) the use of a compensator redesigned for each imaged fraction. The two MSF techniques were (i) a four-field-component design and (ii) a method where the fluence interval between the MLC field components was set so they were equivalent to the compensator ('quantized' MSF-MLC). The final technique investigated was the DMLC. Plans were produced for each of the five methods and a paired t-test was used to assess the reduction in the breast volume outside the dose range 95-105% between sets of pairs of techniques. An on-line correction strategy was simulated to determine the number of treatments that required intervention. The action levels were calculated using the difference between the volume outside the dose range 95-105% calculated for treatments where the DMLC was designed on day 1 only and for each imaged fraction. Differences of greater than 2%, greater than 5% and greater than 10% were investigated. Results: Thirty-five plans were evaluated for each technique. Results showed that a statistically significant mean reduction in the volume of the breast outside the dose range 95-105% could be achieved if the compensators were designed on each imaged fraction rather than on day 1 only (P=0.0045). When the comparison was made between the 'quantized' MSF-MLC and the technique where the compensators were designed on day 1 only, a statistically significant mean reduction in the volume of the breast tissue outside the dose range 95-105% was not achieved (P=0.21). Comparison of the DMLC technique to the technique where the compensators were designed on day 1 only results in a statistically significant mean reduction in the volume outside the dose range 95-105% (P=0.024). This corresponds to a mean reduction in the volume outside 95-105% dose of 1.94%. The 2% action level showed the greatest reduction in the volume outside 95-105% dose and intervention was only required in approximately one-third of the treatments investigated. Conclusions: Redesigning MSFs for each imaged fraction did not provide a statistically significant mean reduction in the volume outside the dose range 95-105%. However, using the DMLC technique creates a statistically significant mean reduction in the volume outside the dose range 95-105%