Column generation for IMRT cancer therapy optimization with implementable segments

A radiation beam passes through normal tissue to reach tumor. The latest devices for the radiotherapy of cancer provide intensity modulated radiation treatment, or IMRT. This method refines cancer treatment by varying the intensity profile across the face of a radiation beam. Intensity modulation is usually accomplished by partitioning each beam, distinguished by its angle of entry, into an array of smaller sized units, called beamlets, assigned different intensities. Planning treatment calls for an optimization over beamlet intensities to maximize the dose delivered to the targeted tumor while keeping the distribution of dose throughout the various organs within physician prescribed bounds. The choice of beam angles can be entered into the optimization as well. A common method to produce an intensity pattern is to block out different parts of the beam for different amounts of time. This can be done sliding narrow blocks (leafs) of unit width into the beam from either of two opposing sides to create different beam shapes called segments. A sequence of segments with their exposure times is superimposed to yield the dose distribution actually received in the patient. Current two stage treatment is derived in separate steps: optimization over independently considered beamlet intensities, and generation of a sequence of segments to approximate the planned intensity map. The approximation degrades the solution, and the separate search for segments adds to planning time. We present a mixed integer programming alternative employing column generation to optimize dose over segments themselves. Only segments that can be realized with delivery devices are generated, and adjustments made for the effects of block edges, so that the optimized plans are directly implementable. Preliminary testing demonstrates gains in both planning efficiency and quality of the plans produced. A portion of the work of Dr. Langer, Mr. Thai and Dr. Preciado-Walters was supported by National Science Foundation grant ECS-0120145 and National Cancer Institute 1R41CA91688-01.     

A risk-based modeling approach for radiation therapy treatment planning under tumor shrinkage uncertainty

Robust optimization approaches have been widely used to address uncertainties in radiation therapy treatment planning problems. Because of the unknown probability distribution of uncertainties, robust bounds may not be correctly chosen, and a risk of undesirable effects from worst-case realizations may exist. In this study, we developed a risk-based robust approach, embedded within the conditional value-at-risk representation of the dose-volume constraint, to deal with tumor shrinkage uncertainty during radiation therapy. The objective of our proposed model is to reduce dose variability in the worst-case scenarios as well as the total delivered dose to healthy tissues and target dose deviations from the prescribed dose, especially, in underdosed scenarios. We also took advantage of adaptive radiation therapy in our treatment planning approach. This fractionation technique considers the response of the tumor to treatment up to a particular point in time and reoptimizes the treatment plan using an estimate of tumor shrinkage. The benefits of our model were tested in a clinical lung cancer case. Four plans were generated and compared: static, nominal-adaptive, robust-adaptive, and conventional robust (worst-case) optimization. Our results showed that the robust-adaptive model, which is a risk-based model, achieved less dose variability and more control on the worst-case scenarios while delivering the prescribed dose to the tumor target and sparing organs at risk. This model also outperformed other models in terms of tumor dose homogeneity and plan robustness.     

IMRT treatment planning for prostate cancer using prioritized prescription optimization and mean-tail-dose functions

Treatment planning for intensity modulated radiation therapy (IMRT) is challenging due to both the size of the computational problems (thousands of variables and constraints) and the multi-objective, imprecise nature of the goals. We apply hierarchical programming to IMRT treatment planning. In this formulation, treatment planning goals/objectives are ordered in an absolute hierarchy, and the problem is solved from the top-down such that more important goals are optimized in turn. After each objective is optimized, that objective function is converted into a constraint when optimizing lower-priority objectives. We also demonstrate the usefulness of a linear/quadratic formulation, including the use of mean-tail-dose (mean dose to the hottest fraction of a given structure), to facilitate computational efficiency. In contrast to the conventional use of dose-volume constraints (no more than x% volume of a structure should receive more than y dose), the mean-tail-dose formulation ensures convex feasibility spaces and convex objective functions. To widen the search space without seriously degrading higher priority goals, we allowed higher priority constraints to relax or 'slip' a clinically negligible amount during lower priority iterations. This method was developed and tuned for external beam prostate planning and subsequently tested using a suite of 10 patient datasets. In all cases, good dose distributions were generated without individual plan parameter adjustments. It was found that allowance for a small amount of 'slip,' especially in target dose homogeneity, often resulted in improved normal tissue dose burdens. Compared to the conventional IMRT treatment planning objective function formulation using a weighted linear sum of terms representing very different dosimetric goals, this method: (1) is completely automatic, requiring no user intervention, (2) ensures high-priority planning goals are not seriously degraded by lower-priority goals, and (3) ensures that lower priority, yet still important, normal tissue goals are separately pushed as far as possible without seriously impacting higher priority goals.     

The determination of optimal treatment plans for Volumetric Modulated Arc Therapy (VMAT)

The success of radiation therapy depends on the ability to deliver the proper amount of radiation to cancerous cells while protecting healthy tissues. As a natural consequence, any new treatment technology improves quality standards concerning primarily this issue. Similar to the widely used Intensity Modulated Radiation Therapy (IMRT), the radiation resource is outside of the patient’s body and the beam is shaped by a multi-leaf collimator mounted on the linear accelerator’s head during the state-of-the-art Volumetric Modulated Arc Therapy (VMAT) as well. However, unlike IMRT, the gantry of the accelerator may rotate along one or more arcs and deliver radiation continuously. This property makes VMAT powerful in obtaining high conformal plans in terms of dose distribution; but the apertures are interdependent and optimal treatment planning problem cannot be decomposed into simpler independent subproblems as a consequence. In this work, we consider optimal treatment planning problem for VMAT. First, we formulate a mixed-integer linear program minimizing total radiation dose intensity subject to clinical requirements embedded within the constraints. Then, we develop efficient solution procedures combining Benders decomposition with certain acceleration strategies. We investigate their performance on a large set of test instances obtained from an anonymous real prostate cancer data.     

On pole assignment of distributed parameter systems with unbounded input elements

In this paper, the pole assignment problem is considered for a class of distributed parameter systems with unbounded input element and with multiple spectral structure. A formula on the spectrum of the closed loop operator is proved and a formula of pole assignment is obtained. Finally, an example concerning a beam vibration is given. This work is supported by the National Natural Sciences Foundation of China and the National Key Project of China, and partly by the Post Doctoral Science Foundation of China and the Youth Science Foundation of Shanxi.     

Optimal time to change premiums

The claim arrival process to an insurance company is modeled by a compound Poisson process whose intensity and/or jump size distribution changes at an unobservable time with a known distribution. It is in the insurance company’s interest to detect the change time as soon as possible in order to re-evaluate a new fair value for premiums to keep its profit level the same. This is equivalent to a problem in which the intensity and the jump size change at the same time but the intensity changes to a random variable with a know distribution. This problem becomes an optimal stopping problem for a Markovian sufficient statistic. Here, a special case of this problem is solved, in which the rate of the arrivals moves up to one of two possible values, and the Markovian sufficient statistic is two-dimensional. This work was partially supported by the US Army Pantheon Project and National Science Foundation under grant DMS-0604491.     

Optimal design of pitched laminated wood beams

The optimal design of a pitched laminated wood beam is considered. An engineering formulation is given in which the volume of the beam is minimized. The problem is then reformulated and solved as a generalized geometric (signomial) program. Sample designs are presented.This research was partially supported by the Office of Naval Research under Contracts Nos. N00014-75-C-0267 and N00014-75-C-0865; by the US Energy Research and Development Administration Contract No. E(04-3)-326 PA-18; and by the National Science Foundation, Grant No. DCR75-04544 at Stanford University. This work was carried out during the first author's stay at the Management Science Division of the University of British Columbia and the Systems Optimization Laboratory of Stanford University. The authors are indebted to Mr. S. Liu and Mrs. M. Ratner for their assistance in performing the computations.     

Multi-input Optimal Control Problems for Combined Tumor Anti-angiogenic and Radiotherapy Treatments

A cell-population-based model for tumor growth under anti-angiogenic treatment, with the tumor volume and its variable carrying capacity as variables, is combined with the linear-quadratic model for damage done by radiation ionization. The resulting multi-input system is analyzed as an optimal control problem with the objective of minimizing the tumor volume subject to isoperimetric constraints that limit the overall amounts of anti-angiogenic agents, respectively, the damage done to healthy tissue by radiotherapy. For various model formulations, explicit expressions for singular controls are derived for both the dosage of the anti-angiogenic therapeutic agent and the radiation dose schedule. Their role in the structure of optimal protocols is discussed.     

Invariant integral and statistical equations of paraxial light beam transmission in free space

The statistical behavior of arbitrary paraxial light beams propagating in free space is investigated by using the Hermite-Gaussian expansion method and Fock’s representation. A series of equivalent Gaussian parameters for paraxial beam and the statistical equations for these parameters are presented. The optical transmission problem in quasi-far field region is studied. The so-called general Hermite-Gaussian beam is defined. Project partly supported by the National Hi-tech Inertial Confinement Fusion Committee, the Natural Science Foundation of Guangdong Province, the Postdoctoral Foundation of China and Guangdong Province.     

Natural frequencies of structures with uncertain but nonrandom parameters

In this paper, we present a method for computing upper and lower bounds of the natural frequencies of a structure with parameters which are unknown, except for the fact that they belong to given intervals. These parameters are uncertain, yet they are not treated as being random, since no information is available on their probabilistic characteristics. The set of possible states of the system is described by interval matrices. By solving the generalized interval eigenvalue problem, the bounds on the natural frequencies of the structure with interval parameters are evaluated. Numerical results show that the proposed method is extremely effective.The research reported in this paper has been supported by the PRC National Natural Science Foundation and by the USA National Science Foundation Grant MSM-9215698 (Program Director Dr. K. P. Chong).     

On the optimal design centering,tolerancing, and tuning problem

In this note, we point out an erroneous result appearing in the literature on the optimal design centering, tolerancing, and tuning problem. Using the same framework of outer approximations, we then suggest alternate approaches to the solution of this problem, including a special-purpose subalgorithm.This research was supported by the National Science Foundation, Grant No. ECS-82-04452. The author wishes to thank Dr. E. Polak for his helpful comments.     

Optimal filter design subject to output sidelobe constraints: Theoretical considerations

The design of filters for detection and estimation in radar and communications systems is considered, with inequality constraints on the maximum output sidelobe levels. A constrained optimization problem in Hilbert space is formulated, incorporating the sidelobe constraints via a partial ordering of continuous functions. Generalized versions (in Hilbert space) of the Kuhn-Tucker and duality theorems allow the reduction of this problem to an unconstrained one in the dual space of regular Borel measures.This research was supported by the National Science Foundation under Grant No. GK-2645, by the National Aeronautics and Space Administration under Grant No. NGL-22-009(124), and by the Australian Research Grants Committee. The authors wish to express their gratitude to Dr. Robert McAulay and Professor Ian Rhodes for various comments and suggestions.     

Duality in infinite dimensional linear programming

We consider the class of linear programs with infinitely many variables and constraints having the property that every constraint contains at most finitely many variables while every variable appears in at most finitely many constraints. Examples include production planning and equipment replacement over an infinite horizon. We form the natural dual linear programming problem and prove strong duality under a transversality condition that dual prices are asymptotically zero. That is, we show, under this transversality condition, that optimal solutions are attained in both primal and dual problems and their optimal values are equal. The transversality condition, and hence strong duality, is established for an infinite horizon production planning problem.This material is based on work supported by the National Science Foundation under Grant No. ECS-8700836.     

Existence theory for generalized nonlinear complementarity problems

The nonlinear complementarity problem is generalized by replacing the usual nonnegative ordering ofR ⁿ by an ordering generated by a convex cone. Two new classes of operators are introduced, each of which is used to guarantee the existence of a solution to the generalized problem. The new classes can be seen to be broader than previously studied classes. In addition, conditions are presented under which the solution set of the generalized linear complementarity problem is shown to have at most a finite number of solutions.This research was partially supported by National Science Foundation, Grant No. GP-16293, and constitutes part of the junior author's doctoral thesis. The authors are indebted to Dr. Carlton E. Lemke for many helpful discussions.     

Uncertainty and Option Value in Land Allocation Problems

In intensity-modulated radiation therapy (IMRT) not only is the shape of the beam controlled, but combinations of open and closed multileaf collimators modulate the intensity as well. In this paper, we offer a mixed integer programming approach which allows optimization over beamlet fluence weights as well as beam and couch angles. Computational strategies, including a constraint and column generator, a specialized set-based branching scheme, a geometric heuristic procedure, and the use of disjunctive cuts, are described. Our algorithmic design thus far has been motivated by clinical cases. Numerical tests on real patient cases reveal that good treatment plans are returned within 30 minutes. The MIP plans consistently provide superior tumor coverage and conformity, as well as dose homogeneity within the tumor region while maintaining a low irradiation to important critical and normal tissues.     

Co-volume methods for degenerate parabolic problems

Summary A complementary volume (co-volume) technique is used to develop a physically appealing algorithm for the solution of degenerate parabolic problems, such as the Stefan problem. It is shown that, these algorithms give rise to a discrete semigroup theory that parallels the continuous problem. In particular, the discrete Stefan problem gives rise to nonlinear semigroups in both the discreteL ¹ andH ^–1 spaces.The first author was supported by a grant from the Hughes foundation, and the second author was supported by the National Science Foundation Grant No. DMS-9002768 while this work was undertaken. This work was supported by the Army Research Office and the National Science Foundation through the Center for Nonlinear Analysis.     

Strong valid inequalities for fluence map optimization problem under dose-volume restrictions

Fluence map optimization problems are commonly solved in intensity modulated radiation therapy (IMRT) planning. We show that, when subject to dose-volume restrictions, these problems are NP-hard and that the linear programming relaxation of their natural mixed integer programming formulation can be arbitrarily weak. We then derive strong valid inequalities for fluence map optimization problems under dose-volume restrictions using disjunctive programming theory and show that strengthening mixed integer programming formulations with these valid inequalities has significant computational benefits.     

Intensity modulated radiation therapy treatment plan optimization

In this paper, we provide an overview of the state-of-the-art of optimization models for static radiation therapy treatment planning, focusing in particular on intensity modulated radiation therapy (IMRT) by (i) establishing a novel connection between risk management and radiation therapy treatment planning, and (ii) unifying and contrasting two different modeling approaches. In addition, we discussion recent and ongoing technological developments which show that this area of research is a lively and promising one that can continue to help patients by improving the clinical practice of radiation therapy. This invited paper is discussed in the comments available at: , , , , . This work was supported by the National Science Foundation under grant no. DMI-0457394/CMMI-0852727.     

Risk pooling strategy in a multi-echelon supply chain with price-sensitive demand

This paper studies a nonstationary inventory and pricing problem. We consider a two-echelon supply chain with one supplier and two retailers, in which the supplier carries all inventory to supply the retailers. Both the reserved and pooled inventory systems are analyzed. Results with normally distributed demands are compared. Assuming the random demand at each retailer is price-sensitive, we further consider the cases when the retailers have and do not have service level requirements. We start with analyzing inventory and pricing strategies for the supplier in a one-period scenario. Then we extend our analysis to both the backlogging and lost-sale scenarios in an infinite planning horizon. The first author’s research is sponsored by Grant No. 70502009 and No. 70432001 of the Chinese National Natural Science Foundation and the second author’s research is sponsored by Grant #W911NF-04-D-0003 of the US Army Research Office and Grant #DMI-0553310 of the US National Science Foundation.