Automated treatment planning for a dedicated multi‐source intra‐cranial radiosurgery treatment unit accounting for overlapping structures and dose homogeneity

Ghobadi, Kimia; Ghaffari, Hamid R.; Aleman, Dionne M.; Jaffray, David A.; Ruschin, Mark

doi:10.1118/1.4817555

Cited by 13 publications

(16 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is well established that the predicted DVH technique works well with parallel organs like rectum, bladder, parotid, etc., where the mean dose to a specific organ must be predicted based on the overlap volume histogram (OVH) . However, with extensive literature search we could not find any study where a predictive dose volume technique being applied effectively for a population of large number of serial organs encountered in conventional brain radiotherapy or linear accelerator‐based brain stereotaxy . Perhaps predictive DVH technique does not work efficiently with large number of serial organs like the optic nerve, optic chiasm, brainstem, cochlea, etc., where the dose in a small volume (1%, 0.2 cc, 0.5 cc) must be predicted .…”

Section: Discussionmentioning

confidence: 99%

“…Most of the researchers and all commercial KBP algorithms use predictive dose volume technique . The issue with the predictive dose volume method is the bandwidth of the predicted DVH is too wide at the mid portion of the curve, sometimes as high as 20% error in the dose and/or volume.…”

Section: Discussionmentioning

confidence: 99%

“…All the KBP techniques described so far by numerous research groups are complex in mathematics. For example, implementation of UCSD KBP algorithm required proficiency in the use of ANN and MatLab programming, which may not be frequent among the clinical users.…”

Section: Discussionmentioning

confidence: 99%

“…Optimal treatment plans can be generated by grossly reducing any subjectivity introduced by the treatment planner. Recently, knowledge‐based planning (KBP) helped the treatment planner standardize the treatment plans for different sites like brain, esophagus, prostate, head and neck, spine SBRT, and pancreas …”

Section: Introductionmentioning

confidence: 99%

“…The predicted DVH technique is the one described by many authors and used in all commercially available planning systems. It uses database DVH averaging approach . The assumption is that any DVH can be represented as a combination of the average DVH over a population plus a sum of weighted principal components and a residual.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Standardization of volumetric modulated arc therapy‐based frameless stereotactic technique using a multidimensional ensemble‐aided knowledge‐based planning

et al. 2019

View full text Add to dashboard Cite

Purpose Aim of this article is to describe a new knowledge‐based planning (KBP) methodology using volumetric modulated arc therapy (VMAT) for stereotactic radiosurgery (SRS) and radiotherapy (SRT) assisted by an ensemble mapping technique for use in a Monte Carlo planning system. Methods Libraries of 121 stereotactic patients were assembled on the basis of eight different parameters (a) tumor laterality, (b) whether planning target volume (PTV) dose coverage challenged by the presence of the organ at risk (OAR), (c) prescription dose and number of fractions, (d) number of PTVs, (e) tumor volume, (f) shortest distance between OAR and PTV (edge to edge distance, or EED), (g) center to center distance between OARs and PTV (CCD), and (h) lateral dimension of external contour (brain). For new patients, the most appropriate library plan was selected on the basis of the above categorization. A KBP plan was created based on this selected library plan with all parameters unchanged keeping the isocenter at the center of PTV. Using the same beam configuration, another independent treatment plan (IP) was generated by an experienced dosimetrist for comparison. IP and KBP were compared for 76 new patients. Results Of 197 patients (121 library and 76 new), 103 (52.3%) were placed in the OAR‐challenged category and 94 were placed in the OAR unchallenged category. The ensemble mapping technique shows that, for an OAR‐challenged patient, picking up the library plan is appropriate. IP was marginally better than KBP in PTV coverage and dose conformity (PCI). Library plans, IP, and KBP offer a mean PCI of 0.77 ± 0.2, 0.79 ± 0.2, and 0.78 ± 0.4, mean PTV‐V99% of 97.3 ± 22.0%, 98.9 ± 14.1%, and 98.2 ± 13.2%, and mean MU of 2403.8 ± 2403.8, 2344.0 ± 2423.6, and 2473.6 ± 2296.8, respectively. Statistically significant differences were observed in the planning time between the IP and KBP plans for both OAR‐challenged (P < 0.001) and ‐unchallenged (P < 0.002) categories. Comparison of optimization and dose calculation time showed a much lower average planning time of 111.0 ± 84.1 min for KBP as against 248.2 ± 96.6 min for IP. Conclusion Validation results for KBP plans indicate the multidimensional ensemble mapping mechanism can accurately pick up the most appropriate library plan. KBP plans, although slightly inferior in their dosimetric quality, fulfill all the required clinical conditions and dose constraints. KBP plans save considerable planning time and are nearly independent of the skill and knowledge of the treatment planner. KBP works well with a Monte Carlo planning system like Monaco.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Standardization of volumetric modulated arc therapy‐based frameless stereotactic technique using a multidimensional ensemble‐aided knowledge‐based planning

et al. 2019

View full text Add to dashboard Cite

show abstract

A linear programming approach to inverse planning in Gamma Knife radiosurgery

et al. 2019

View full text Add to dashboard Cite

PurposeLeksell Gamma Knife® is a stereotactic radiosurgery system that allows fine‐grained control of the delivered dose distribution. We describe a new inverse planning approach that both resolves shortcomings of earlier approaches and unlocks new capabilities.MethodsWe fix the isocenter positions and perform sector‐duration optimization using linear programming, and study the effect of beam‐on time penalization on the trade‐off between beam‐on time and plan quality. We also describe two techniques that reduce the problem size and thus further reduce the solution time: dualization and representative subsampling.ResultsThe beam‐on time penalization reduces the beam‐on time by a factor 2–3 compared with the naïve alternative. Dualization and representative subsampling each leads to optimization time‐savings by a factor 5–20. Overall, we find in a comparison with 75 clinical plans that we can always find plans with similar coverage and better selectivity and beam‐on time. In 44 of these, we can even find a plan that also has better gradient index. On a standard GammaPlan workstation, the optimization times ranged from 2.3 to 26 s with a median time of 5.7 s.ConclusionWe present a combination of techniques that enables sector‐duration optimization in a clinically feasible time frame.

show abstract

Automatic inverse treatment planning of Gamma Knife radiosurgery via deep reinforcement learning

et al. 2022

View full text Add to dashboard Cite

Purpose Several inverse planning algorithms have been developed for Gamma Knife (GK) radiosurgery to determine a large number of plan parameters by solving an optimization problem, which typically consists of multiple objectives. The priorities among these objectives need to be repetitively adjusted to achieve a clinically good plan for each patient. This study aimed to achieve automatic and intelligent priority tuning by developing a deep reinforcement learning (DRL)‐based method to model the tuning behaviors of human planners. Methods We built a priority‐tuning policy network using deep convolutional neural networks. Its input was a vector composed of multiple plan metrics that were used in our institution for GK plan evaluation. The network can determine which tuning action to take based on the observed quality of the intermediate plan. We trained the network using an end‐to‐end DRL framework to approximate the optimal action‐value function. A scoring function was designed to measure the plan quality to calculate the received reward of a tuning action. Results Vestibular schwannoma was chosen as the test bed in this study. The number of training, validation and testing cases were 5, 5, and 16, respectively. For these three datasets, the average scores of the initial plans obtained with the same initial priority set were 3.63 ± 1.34, 3.83 ± 0.86 and 4.20 ± 0.78, respectively, while they were improved to 5.28 ± 0.23, 4.97 ± 0.44 and 5.22 ± 0.26 through manual priority tuning by human expert planners. Our network achieved competitive results with 5.42 ± 0.11, 5.10 ± 0. 42, 5.28 ± 0.20, respectively. Conclusions Our network can generate GK plans of comparable or slightly higher quality than the plans generated by human planners via manual priority tuning for vestibular schwannoma cases. The network can potentially be incorporated into the clinical workflow as planning assistance to improve GK planning efficiency and help to reduce plan quality variation caused by interplanner variability. We also hope that our method can reduce the workload of GK planners and allow them to spend more time on more challenging cases.

show abstract

Automated treatment planning for a dedicated multi‐source intra‐cranial radiosurgery treatment unit accounting for overlapping structures and dose homogeneity

Cited by 13 publications

References 19 publications

Standardization of volumetric modulated arc therapy‐based frameless stereotactic technique using a multidimensional ensemble‐aided knowledge‐based planning

Standardization of volumetric modulated arc therapy‐based frameless stereotactic technique using a multidimensional ensemble‐aided knowledge‐based planning

A linear programming approach to inverse planning in Gamma Knife radiosurgery

Automatic inverse treatment planning of Gamma Knife radiosurgery via deep reinforcement learning

Contact Info

Product

Resources

About