scholarly journals A computational optimization approach for the automatic generation of Gamma Knife radiosurgery treatment plans

2021 ◽  
Author(s):  
Matthew R Walker ◽  
Mehrdad Malekmohammadi ◽  
Catherine Coolens ◽  
Normand Laperriere ◽  
Robert Heaton ◽  
...  
Keyword(s):  
Stochastic Optimization ◽  
Gamma Knife ◽  
Optimization Problems ◽  
Organs At Risk ◽  
Gamma Knife Radiosurgery ◽  
Automatic Generation ◽  
Optimization Approach ◽  
Combinatorial Optimization Problems ◽  
Automated Method ◽  
Treatment Plans

Gamma knife (GK) radiosurgery is a non-invasive treatment modality which allows single fraction delivery of focused radiation to one or more brain targets. Treatment planning mostly involves manual placement and shaping of shots to conform the prescribed dose to a surgical target. This process can be time consuming and labour intensive. An automated method is needed to determine the optimum combination of treatment parameters to decrease planning time and chance for operator-related error. Recent advancements in hardware platforms which employ parallel computational methods with stochastic optimization schemes are well suited to solving such combinatorial optimization problems efficiently. We present a method of generating optimized GK radiosurgery treatment plans using these techniques, which we name ROCKET (Radiosurgical Optimization Configuration Kit for Enhanced Treatments). Our approach consists of two phases in which shot isocenter positions are generated based on target geometry, followed by optimization of sector collimator parameters. Using this method, complex treatment plans can be generated, on average, in less than one minute, a substantial decrease relative to manual planning. Our results also demonstrate improved selectivity and treatment safety through decreased exposure to nearby organs-at-risk (OARs), compared to manual reference plans with matched coverage. Stochastic optimization is therefore shown to be a robust and efficient clinical tool for the automatic generation of GK radiosurgery treatment plans.

An Agent Based Optimization Approach to Manufacturing Process Planning

2001 ◽  
Author(s):  
Saurabh Deshpande ◽  
Jonathan Cagan
Keyword(s):  
Stochastic Optimization ◽  
Process Planning ◽  
Manufacturing Process ◽  
Optimization Problems ◽  
Search Space ◽  
Optimization Techniques ◽  
Process Models ◽  
Optimization Approach ◽  
Agent Based ◽  
Manufacturing Process Planning

Abstract Many optimization problems, such as manufacturing process planning optimization, are difficult problems due to the large number of potential configurations (process sequences) and associated (process) parameters. In addition, the search space is highly discontinuous and multi-modal. This paper introduces an agent based optimization algorithm that combines stochastic optimization techniques with knowledge based search. The motivation is that such a merging takes advantage of the benefits of stochastic optimization and accelerates the search process using domain knowledge. The result of applying this algorithm to computerized manufacturing process models is presented.

Tension-Efficient Cable-Driven Robot Design Using Particle Swarm Optimization

2015 ◽  
Author(s):  
Joshua T. Bryson ◽  
Sunil K. Agrawal
Keyword(s):  
Particle Swarm Optimization ◽  
Stochastic Optimization ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Robot Design ◽  
Optimization Approach ◽  
Swarm Optimization ◽  
Robot Leg ◽  
Optimization Methodology ◽  
Specific Implementation

Cable-driven robots have advantages which make them attractive solutions for a variety of tasks, however, the unidirectional nature of cable actuators complicates the design and often results in multiply redundant cable architectures which increase cost and robot complexity. This paper presents a stochastic optimization approach to the problem of designing a cable routing for a cable-driven manipulator to provide the desired robot workspace while minimizing the cable tensions required to perform a desired task. Two cable routing design variants are developed for a robot leg through the application of a stochastic optimization methodology called Particle Swarm Optimization. The PSO methodology is summarized, followed by a description of the specific implementation of the methodology to the particular problem of optimizing the cable routing of a robot leg. An objective function is developed to capture all pertinent design criteria in a quantitative evaluation of each particular set of cable parameters. Finally, a description of the PSO execution is presented and the results of the two optimization problems are presented and discussed.

scholarly journals Standardization of terminology in stereotactic radiosurgery: Report from the Standardization Committee of the International Leksell Gamma Knife Society

2014 ◽  
Vol 121 (Suppl_2) ◽  
pp. 2-15 ◽  
Author(s):  
Michael Torrens ◽  
Caroline Chung ◽  
Hyun-Tai Chung ◽  
Patrick Hanssens ◽  
David Jaffray ◽  
...  
Keyword(s):  
Stereotactic Radiosurgery ◽  
Gamma Knife ◽  
Organs At Risk ◽  
Gamma Knife Radiosurgery ◽  
Target Volume ◽  
Conformity Index ◽  
Society Meeting ◽  
Coverage Ratio ◽  
Dose Volume ◽  
Volume Limit

ObjectThis report has been prepared to ensure more uniform reporting of Gamma Knife radiosurgery treatment parameters by identifying areas of controversy, confusion, or imprecision in terminology and recommending standards.MethodsSeveral working group discussions supplemented by clarification via email allowed the elaboration of a series of provisional recommendations. These were also discussed in open session at the 16th International Leksell Gamma Knife Society Meeting in Sydney, Australia, in March 2012 and approved subject to certain revisions and the performance of an Internet vote for approval from the whole Society. This ballot was undertaken in September 2012.ResultsThe recommendations in relation to volumes are that Gross Target Volume (GTV) should replace Target Volume (TV); Prescription Isodose Volume (PIV) should generally be used; the term Treated Target Volume (TTV) should replace TVPIV, GTV in PIV, and so forth; and the Volume of Accepted Tolerance Dose (VATD) should be used in place of irradiated volume. For dose prescription and measurement, the prescription dose should be supplemented by the Absorbed Dose, or DV% (for example, D95%), the maximum and minimum dose should be related to a specific tissue volume (for example, D2% or preferably D1 mm3), and the median dose (D50%) should be recorded routinely. The Integral Dose becomes the Total Absorbed Energy (TAE). In the assessment of planning quality, the use of the Target Coverage Ratio (TTV/ GTV), Paddick Conformity Index (PCI = TTV2/[GTV · PIV]), New Conformity Index (NCI = [GTV · PIV]/TTV2), Selectivity Index (TTV/PIV), Homogeneity Index (HI = [D2% –D98%]/D50%), and Gradient Index (GI = PIV0.5/PIV) are reemphasized. In relation to the dose to Organs at Risk (OARs), the emphasis is on dose volume recording of the VATD or the dose/volume limit (for example, V10) in most cases, with the additional use of a Maximum Dose to a small volume (such as 1 mm3) and/or a Point Dose and Mean Point Dose in certain circumstances, particularly when referring to serial organs. The recommendations were accepted by the International Leksell Gamma Knife Society by a vote of 92% to 8%.ConclusionsAn agreed-upon and uniform terminology and subsequent standardization of certain methods and procedures will advance the clinical science of stereotactic radiosurgery.

An Agent Based Optimization Approach to Manufacturing Process Planning

2004 ◽  
Vol 126 (1) ◽  
pp. 46-55 ◽  
Author(s):  
Saurabh Deshpande ◽  
Jonathan Cagan
Keyword(s):  
Stochastic Optimization ◽  
Process Planning ◽  
Manufacturing Process ◽  
Optimization Problems ◽  
Search Space ◽  
Optimization Techniques ◽  
Process Models ◽  
Optimization Approach ◽  
Agent Based ◽  
Manufacturing Process Planning

Many optimization problems, such as manufacturing process planning optimization, are difficult problems due to the large number of potential configurations (process sequences) and associated (process) parameters. In addition, the search space is highly discontinuous and multi-modal. This paper introduces an agent based optimization algorithm that combines stochastic optimization techniques with knowledge based search. The motivation is that such a merging takes advantage of the benefits of stochastic optimization and accelerates the search process using domain knowledge. The result of applying this algorithm to computerized manufacturing process models is presented.

scholarly journals Aggregation–Decomposition-Based Multi-Agent Reinforcement Learning for Multi-Reservoir Operations Optimization

Water ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 2688 ◽  
Author(s):  
Milad Hooshyar ◽  
S. Jamshid Mousavi ◽  
Masoud Mahootchi ◽  
Kumaraswamy Ponnambalam
Keyword(s):  
Reinforcement Learning ◽  
Stochastic Optimization ◽  
Optimization Problems ◽  
Transition Probability ◽  
Optimization Under Uncertainty ◽  
Stochastic Dynamic ◽  
Optimization Approach ◽  
Reservoir Operations ◽  
Multi Agent ◽  
Operations Optimization

Stochastic dynamic programming (SDP) is a widely-used method for reservoir operations optimization under uncertainty but suffers from the dual curses of dimensionality and modeling. Reinforcement learning (RL), a simulation-based stochastic optimization approach, can nullify the curse of modeling that arises from the need for calculating a very large transition probability matrix. RL mitigates the curse of the dimensionality problem, but cannot solve it completely as it remains computationally intensive in complex multi-reservoir systems. This paper presents a multi-agent RL approach combined with an aggregation/decomposition (AD-RL) method for reducing the curse of dimensionality in multi-reservoir operation optimization problems. In this model, each reservoir is individually managed by a specific operator (agent) while co-operating with other agents systematically on finding a near-optimal operating policy for the whole system. Each agent makes a decision (release) based on its current state and the feedback it receives from the states of all upstream and downstream reservoirs. The method, along with an efficient artificial neural network-based robust procedure for the task of tuning Q-learning parameters, has been applied to a real-world five-reservoir problem, i.e., the Parambikulam–Aliyar Project (PAP) in India. We demonstrate that the proposed AD-RL approach helps to derive operating policies that are better than or comparable with the policies obtained by other stochastic optimization methods with less computational burden.

scholarly journals Enhanced Genetic Algorithm Optimization Models for Vehicular Routing Problems

2019 ◽  
Vol 9 (2) ◽  
pp. 3140-3149
Keyword(s):  
Genetic Algorithm ◽  
Optimization Problems ◽  
Optimization Models ◽  
Optimization Approach ◽  
Algorithm Optimization ◽  
Routing Problem ◽  
Performance Factors ◽  
Fine Grained ◽  
Combinatorial Optimization Problems ◽  
Routing Strategy

The Vehicle Routing Problem (VRP) is one of the most studied combinatorial optimization problems because of its practical relevance and complexity. Though there are several techniques have been proposed to solve the VRPs and its variants effectively, each technique has its own tradeoff values in terms of the performance factors. From this perspective, the work presented in this paper proposed an intelligent routing strategy for VRP based on distance values between the cities. The proposed strategy uses an enhanced model of Genetic Algorithm to find the optimal tour paths among the cities under distance based optimized tour path estimation scenarios. For distance-based optimization approach, experiments were performed on the standard benchmark TSP instances obtained from TSPLIB. A set of fine-grained result analyses demonstrated that the proposed model of routing strategies performed comparatively better w.r.t. the existing relevant approaches. By considering this problem as the base, a distinct model was developed as a set of assistive modules for Genetic Algorithms (GA), which are aimed at improving the overall efficiency of the typical GA, particularly for optimization problems. The capability of the proposed optimization models for VRP is demonstrated at various levels, particularly at the population initialization stage, using a set of well-defined experiments.

Sign in / Sign up

Export Citation Format

Share Document