A Column Generation Approach Based on Region Growth

A Treatment Planning System for Intensity-Modulated Small Animal Radiotherapy Realized Using an Existing Movable Rectangular Collimator

This paper presents our research findings on the direct detection and position determination (DDPD) problem, which is to estimate the positions and the total number of multiple sources with intermittent emission, using the received signals intercepted by a moving antenna array at multiple observation slots. We combine the direct position determination (DPD) concept and the Minimum Variance Distortionless Response (MVDR) criterion to solve the DDPD problem without the prior knowledge of the effective number of sources (the number of sources at every observation slot). We find out that the parallel combination of the MVDR spectrum in the cost function results in its own sensitivity to the missing emission, which makes the location estimator bias. Therefore, we use the K-means clustering algorithm to identify the sources with intermittent emission and their emitting slots adaptively, based on the MVDR spectrum values over all observation slots. Then, we use the spectrum values of the identified emitting slots to reconstruct the MVDR spectrum. Finally, we find the peaks of the reconstructed MVDR spectrum, which are corresponding to the source positions. The results of the simulations demonstrate that the proposed method gets asymptotic performance with the signal to noise ratio (SNR), both in the aspects of the root mean square error (RMSE) and the bias error, and high resolution as a generalized MVDR based method.

Nowadays, large-scale failures in elastic optical inter-data center networks (EO-DCNs) are critical issues for network survivability. Once the original connection or the primary DC fails, an alternative data center (DC) can serve the cloud services with replicated content. In this paper, we study disaster-resilient cloud services provisioning with distance-adaptive modulation formats, considering content placement, routing with modulation format adaption, protection of path and content, and spectrum allocation simultaneously. Both dedicated end-to-content backup path protection (DEBPP) and shared end-to-content backup path protection (SEBPP) are studied. First, integer linear program (ILP) models are formulated with the objective of minimum spectrum usage. For the sake of scalability, a column generation (CG) approach is developed to decompose the ILP model into a master problem and a pricing problem. The CG approach is shown to achieve comparable results to ILP models. Numerical simulations are conducted to investigate the spectrum usage of DEBPP and SEBPP with different traffic parameters and different network topologies.

A column generation (CG) approach for the solution of timetabling problems is presented. This methodology could be used for various instances of the timetabling problem, although in this paper the solution of the high-school situation in Greece is presented. The results obtained show clearly that the CG approach that has been extremely successful in recent years in the solution of airline crew scheduling problems could also be very efficient and robust for the solution of timetabling problems. Several large timetabling problems corresponding to real problems have been successfully solved, with the solutions obtained feasible and of very high quality in accordance with the problem definition. In addition, none of the solutions contained any idle hour for any of the teachers, which was one of the main goals of this optimization effort.

We consider the problem of intensity-modulated radiation therapy (IMRT) treatment planning using direct aperture optimization. While this problem has been relatively well studied in recent years, most approaches employ a heuristic approach to the generation of apertures. In contrast, we use an exact approach that explicitly formulates the fluence map optimization (FMO) problem as a convex optimization problem in terms of all multileaf collimator (MLC) deliverable apertures and their associated intensities. However, the number of deliverable apertures, and therefore the number of decision variables and constraints in the new problem formulation, is typically enormous. To overcome this, we use an iterative approach that employs a subproblem whose optimal solution either provides a suitable aperture to add to a given pool of allowable apertures or concludes that the current solution is optimal. We are able to handle standard consecutiveness, interdigitation and connectedness constraints that may be imposed by the particular MLC system used, as well as jaws-only delivery. Our approach has the additional advantage that it can explicitly account for transmission of dose through the part of an aperture that is blocked by the MLC system, yielding a more precise assessment of the treatment plan than what is possible using a traditional beamlet-based FMO problem. Finally, we develop and test two stopping rules that can be used to identify treatment plans of high clinical quality that are deliverable very efficiently. Tests on clinical head-and-neck cancer cases showed the efficacy of our approach, yielding treatment plans comparable in quality to plans obtained by the traditional method with a reduction of more than 75% in the number of apertures and a reduction of more than 50% in beam-on time, with only a modest increase in computational effort. The results also show that delivery efficiency is very insensitive to the addition of traditional MLC constraints; however, jaws-only treatment requires about a doubling in beam-on time and number of apertures used. Finally, we showed the importance of accounting for transmission effects when assessing or, preferably, optimizing treatment plan quality.

In 2012, almost 700,000 new head and neck cancer (HNSCC) cases were estimated to occur worldwide with two thirds of those cases occurring in developing countries. Within the geographic areas in Latin America, characterized by high incidence rates of head and neck cancer, the region comprising Argentina, southern Brazil and Uruguay has the highest levels of incidence. Despite current therapy approaches, the survival rate is around 50%, and the main reason for treatment failure is the frequent development of regional recurrences, affecting around 30% of patients. In this context the use of ctDNA (specific biomarker for tumors) using techniques of mass sequencing (high sensitivity and specificity) is presented as a promising tool for noninvasive diagnosis and monitoring of HNSCC cases. Our primary goal was to provide a comprehensive evaluation of the presence of ctDNA in both plasma and oral rinses from a series of HNSCC cases without previous knowledge of tumor mutational status. 43 cases diagnosed as head and neck squamous cell carcinoma were selected from a multi center case-control study conducted in Argentina between 1998 and 2002. Subsite and stage distribution was as follows: 14 oral cavity, 6 oropharynx, 1 hypopharynx, 15 larynx and 7 overlapping lesions. 6 cases were stage III and 17 stage IV. Additionally, 50 controls matched by center, age, sex, smoking and alcohol status were included to assess a threshold of the background noise after sequencing. Targeted sequencing was performed independently in duplicate on plasma, oral rinses, germline DNA and the tumor material using the Ion Proton™ System. Sequencing included the entire coding region of TP53 as the most frequently mutated gene in HNSCC. Both Plasma and oral rinses samples were sequenced at an average depth of 10000X. For somatic variant calling we used a recently developed new statistical model and a bioinformatics pipeline based on Poisson distribution for estimation of sequencing errors distribution to accurately identify variants above the sequencing background. Statistical tests were carried out in R statistical programming environment. 74% of tumor samples sequenced harboured at least one TP53 somatic mutation. 10 TP53 variants were identified in the corresponding plasma samples and 1 of these variants matched the tumor mutation, possibly reflecting high tumor heterogeneity. In contrast, 90% of the variants detected in the tumor were also found in in the oral rinses with allelic frequencies ranging from 0.002 to 0.05 showing higher specificity of oral rinses as a surrogate sample for mutation detection. The detection and analysis of ctDNA in noninvasive samples might be included as a potential biomarker to improve early diagnosis and patient monitoring in HNSCC cases. Citation Format: Sandra Perdomo, Patrice Avogbe, Marta Vilensky, James Mckay, Paul Brennan. Detection of circulating tumor DNA in oral rinses and plasma from head and neck cancer cases in Argentina. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 483.

A column generation approach to capacitated p-median problems

A column generation approach for the unconstrained binary quadratic programming problem

The problem of scheduling and tree routing in WiMAX/802.16 based mesh networks were not defined in the standard and are thus subject to extensive research. In this paper, we consider the problem of joint routing and scheduling in 802.16-based wireless mesh network, with the objective of determining a minimum length schedule that satisfies a given (uplink/downlink) end-to-end traffic demand. Minimizing the schedule length amounts to maximizing the spectrum spatial reuse by concurrently transmitting on as many links as possible, which we refer to as a transmission configuration (a group of links that can simultaneously transmit without violating the signal-to-interferenceplus- noise ratio (SINR) requirement). Our model is referred to as maximum spatial reuse (MSR). Since there is an overwhelming number of possible transmission configurations to be assigned to time slots, we adopt the column generation technique to construct our MSR model. We present two formulations for modeling MSR, namely the link-based column generation (CGLink) formulation and the path-based column generation (CGPath) formulation. These two formulations differ mainly in the number of routing decision variables. Our experimental results indicate that the path-based formulation needs much less computational (CPU) time than the link-based formulation in order to determine the (same) optimized solution with the same spatial reuse gain.

We consider two approaches for solving the classical minimum vertex coloring problem—that is, the problem of coloring the vertices of a graph so that adjacent vertices have different colors and minimizing the number of used colors—namely, constraint programming and column generation. Constraint programming is able to solve very efficiently many of the benchmarks but suffers from a lack of effective bounding methods. On the contrary, column generation provides tight lower bounds by solving the fractional vertex coloring problem exploited in a branch-and-price algorithm, as already proposed in the literature. The column generation approach is here enhanced by using constraint programming to solve the pricing subproblem and to compute heuristic solutions. Moreover, new techniques are introduced to improve the performance of the column generation approach in solving both the linear relaxation and the integer problem. We report extensive computational results applied to the benchmark instances: we are able to prove optimality of 11 new instances and to improve the best-known lower bounds on 17 other instances. Moreover, we extend the solution approaches to a generalization of the problem known as the minimum vertex graph multicoloring problem, where a given number of colors has to be assigned to each vertex.

The time required to deliver intensity-modulated radiation therapy (IMRT) treatments can be significantly longer than conventional treatments, especially for the segmented multileaf collimator (sMLC) delivery system with a large record and verification (R&V) overhead. In this work, we evaluate the impact of the number of intensity-modulated beams (IMBs) and the number of intensity levels (ILs) on the quality and delivery efficiency of IMRT plans, generated by the Corvus planning system for sMLC delivery on a Siemens LINAC with the Lantis R&V system. Detailed studies were performed for three image data sets of previously treated oropharyngeal patients. Treatment plans for patient 1 were developed using 5, 7, 9, or 15 evenly spaced axial IMBs as well as one with 7 axial IMBs whose directions were user-selected, each using ILs of 3, 5, 10, or 20. For patients 2 and 3, plans with 15 IMBs and 20 ILs were not attempted. A total of 42 plans were developed using three oropharyngeal cancer CT image data sets. Plan quality was evaluated by assessing compliance with the Radiation Therapy Oncology Group (RTOG) H-0022 protocol criteria and the physician's clinical judgment. Plan efficiency was accessed by the number of segments of each plan. We found that for our treatment-planning and delivery system, an IMRT plan that uses a moderate number of IMBs and ILs, such as 7 or 9 IMBs with 3 or 5 ILs, would appear to be the optimal approach when both quality of the plan and delivery efficiency are considered. Based on this study, we have routinely used 9 IMBs with 3 ILs or 7 IMBs with 5 ILs for head and neck patients. A retrospective comparison indicates that delivery efficiency is improved on the order of 30% compared to plans generated with 9 IMBs with 5 ILs. PACS number: 87.53.Tf

Objective To investigate whether the change of beam set-up methods will influence the dosimetric quality of intensity modulated radiation therapy (IMRT) for non-small cell lung cancer (NSCLC).Methods Twenty-one stage Ⅰ-Ⅲ NSCLC patients were selected for this study.The technique of step and shoot was used and three different beam set-up methods were chosen for IMRT planning,including IMRT-7 with nine equal-spaced beams angled 0°,51°,102°,153°,204°,255°and 306°; IMRT-5 with five equal-spaced beams angled 0°,72°,144°,216°and 288°; and IMRT-5m which was created from IMRT-7 but excluded 2 fields (51°and 102° were omitted if there was lesion in the right lung,while 255°and 306° were excluded if there was lesion in the left lung).The dose constrains ofnormal lungs for IMRT were set according to V5-V60 of normal lungs obtained from the same patient's actually treated 3D-CRT dose volume histogram.The prescription dose for IMRT started from 65 Gy,and then escalated or decreased step by step by 2 Gy once a time until the best plan was obtained.Results For normal lung dose,IMRT-5m had lower V5-V25 than the other two groups; but there was no significant difference in V30-V40.IMRT-5 was the worst for V45-V60; and mean lung dose was lowest in IMRT-5m.Dose parameters of esophagus and spinal cord,target conformity index,and total monitor units were all similar among difference plans.IMRT-5m had lowest heart V40 compared to the other two groups.For target heterogeneity index,IMRT-5 was higher than IMRT-7,but there were no significant differences among IMRT-5m,IMRT-5 and IMRT-7.Compared to 3D-CRT,the prescription dose could be increased by (5.1 ±4.6) Gy for IMRT-7,(3.1 ±5.3) Gy for IMRT-5,and (5.5 ±4.8)Gy for IMRT-5m.Conclusion Fewer beams and modified beam angles could result in similar,even better plan quality. Key words: Non-small cell lung cancer; Intensity modulated radiation therapy; Dosimetry; Beam angle; Beam number

Objective To explore the effect of horizontal beam on IMRT plan for lung cancer. Methods Totally 18 patients were enrolled, including 10 cases of left lung cancer and 8 cases of right lung cancer. Two intensity modulated radiation therapy (IMRT) plans were generated: one was the normal IMRT plan and another was the IMRT plan including a horizontal beam (H-IMRT). MUs, sub-field and the dose distribution of target and organs at risk (OARs) were compared. Results There were no statistically significant differences between two plans in Dmax, Dminand Dmean of PTV, the V30and Dmean of the ipsilateral lung, V20 and V30of the contralateral lung, and V30 of the total lung. When the horizontal beam was added, the PTV heterogeneity index(HI) and conformity index (CI) were significantly increased (t=-9.33, 10.88, P<0.05), the other dose index in lung was inferior (t=-1.55--7.58, P<0.05), whle heart sparing were clearly better (t=1.84-3.99, P<0.05), the number of MUs and segments were less(t=12.57, 3.19, P<0.05). Conclusions IMRT in the treatment of lung cancer can further improve the PTV coverage when the horizontal beam is added. However, the increasing lung volume of low dose irradiation is not conducive to lung protection. Key words: Horizontal beam; Intensity modulated radiation therapy; Lung cancer

&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;The purpose of this study is to analyze and compare the dosimetric parameters of three dimensional conformal radiotherapy (3DCRT) and intensity modulated radiotherapy (IMRT) in selected non-small cell lung cancer (NSCLC) cases. &lt;strong&gt;&lt;/strong&gt;&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Ten patients with inoperable NSCLC were selected for this study. The 3DCRT and IMRT plans were generated for all patients following Radiation Therapy Oncology Group (RTOG) guidelines. Generated plans were then compared on the basis of planning target volume (PTV) coverage, dose delivered to organs at risk, homogeneity index (HI), and conformity index (CI) for the prescribed dose (PD) of 50 Gy in 25 fractions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The mean D&lt;sub&gt;95 &lt;/sub&gt;and D&lt;sub&gt;99 &lt;/sub&gt;(dose to the 95% and 99% volume) for the PTV were found better in the 3DCRT plans compared to the ones in the IMRT plans. On an average, the volume receiving 20 Gy (V&lt;sub&gt;20&lt;/sub&gt;) of contralateral lung was 2.91% and 3.03% in the 3DCRT and IMRT plans, respectively. The D&lt;sub&gt;mean&lt;/sub&gt; of contralateral lung was 3.17 Gy (3DCRT) versus 4.2 Gy (IMRT), whereas the D&lt;sub&gt;mean&lt;/sub&gt; of ipsilateral lung was 12.69 Gy (3DCRT) and 13.82 Gy (IMRT). The V&lt;sub&gt;20&lt;/sub&gt; of ipsilateral lung was found to be slightly lower in the 3DCRT (25.67%) when compared to the IMRT (30.50%). The dose to the heart was comparable in the 3DCRT and IMRT plans (mean dose: 4.42 Gy versus 4.48 Gy; D&lt;sub&gt;33&lt;/sub&gt;: 3.77 Gy versus and 4.02 Gy). For the spinal cord, the D&lt;sub&gt;max &lt;/sub&gt;was found to be lower in the 3DCRT plans (18.40 Gy) when compared to the IMRT plans (25.49 Gy). The HI was 1.08 versus 1.41 in the 3DCRT and IMRT plans, respectively. The CI was identical (1.67) in both sets of plans.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;Based on the results of this study, the PTV coverage was found to be slightly better in the 3DCRT plans when compared to the one in the IMRT plans. On average, the dose to the organs at risk were found to be comparable.&lt;/p&gt;

Inverse and forward optimization of one- and two-dimensional intensity-modulated radiation therapy-based treatment of concave-shaped planning target volumes: the case of prostate cancer.