A novel dose fall-off index and preliminary application in brain and lung stereotactic radiotherapy.

BACKGROUND: Stereotactic radiotherapy (SRT) has been widely used for the treatment of brain metastases and early stage non-small-cell lung cancer (NSCLC). Excellent SRT plans are characterized by steep dose fall-off, making it critical to accurately and comprehensively predict and evaluate dose fall-off. PURPOSE: A novel dose fall-off index was proposed to ensure high-quality SRT planning. METHODS: The novel gradient index (NGI) had two different modes: NGIx V for three-dimensions and NGIx r for one-dimension. NGIx V and NGIx r were defined as the ratios of the decreased percentage dose (x%) to the corresponding isodose volume and equivalent sphere radii, respectively. A total of 243 SRT plans at our institution between April 2020 and March 2022 were enrolled, including 126 brain and 117 lung SRT plans. Measurement-based verifications were performed using SRS MapCHECK. Ten plan complexity indexes were calculated. Dosimetric parameters related to radiation injuries were also extracted, including the normal brain volume exposed to 12 Gy (V12 ) and 18 Gy (V18 ) during single-fraction SRT (SF-SRT) and multi-fraction SRT (MF-SRT), respectively, and the normal lung volume exposed to 12 Gy (V12 ). The performance of NGI and other common dose fall-off indexes, gradient index (GI), R50% and D2cm were evaluated using Spearman correlation analysis to explore their correlations with the PTV size, gamma passing rate (GPR), plan complexity indexes, and dosimetric parameters. RESULTS: There were statistically significant correlations between NGI and PTV size (r = -0.98, P < 0.01 for NGI50 V and r = -0.93, P < 0.01 for NGI50 r), which were the strongest correlations compared with GI (r = 0.11, P = 0.13), R50% (r = -0.08, P = 0.19) and D2cm (r = 0.84, P < 0.01). The fitted formulas of NGI50 V = 23.86V-1.00 and NGI50 r = 113.5r-1.05 were established. The GPRs of enrolled SRT plans were 98.6 ± 1.7%, 94.2 ± 4.7% and 97.1 ± 3.1% using the criteria of 3%/2 mm, 3%/1 mm, and 2%/2 mm, respectively. NGI50 V achieved the strongest correlations with various plan complexity indexes (|r| ranged from 0.67 to 0.91, P < 0.01). NGI50 V also showed the highest r values with V12 (r = -0.93, P < 0.01) and V18 (r = -0.96, P < 0.01) of the normal brain during SF-SRT and MF-SRT, respectively, and V12 (r = -0.86, P < 0.01) of the normal lung during lung SRT. CONCLUSIONS: Compared with GI, R50% and D2cm , the proposed dose fall-off index, NGI, had the strongest correlations with the PTV size, plan complexity and V12 /V18 of the normal tissues. These correlations established on NGI are more helpful and reliable for SRT planning, quality control, and reducing the risk of radiation injuries.

Correlation between the γ passing rates of IMRT plans and the volumes of air cavities and bony structures in head and neck cancer.

BACKGROUND: Both patient-specific dose recalculation and γ passing rate analysis are important for the quality assurance (QA) of intensity modulated radiotherapy (IMRT) plans. The aim of this study was to analyse the correlation between the γ passing rates and the volumes of air cavities (Vair) and bony structures (Vbone) in target volume of head and neck cancer. METHODS: Twenty nasopharyngeal carcinoma and twenty nasal natural killer T-cell lymphoma patients were enrolled in this study. Nine-field sliding window IMRT plans were produced and the dose distributions were calculated by anisotropic analytical algorithm (AAA), Acuros XB algorithm (AXB) and SciMoCa based on the Monte Carlo (MC) technique. The dose distributions and γ passing rates of the targets, organs at risk, air cavities and bony structures were compared among the different algorithms. RESULTS: The γ values obtained with AAA and AXB were 95.6 ± 1.9% and 96.2 ± 1.7%, respectively, with 3%/2 mm criteria (p > 0.05). There were significant differences (p < 0.05) in the γ values between AAA and AXB in the air cavities (86.6 ± 9.4% vs. 98.0 ± 1.7%) and bony structures (82.7 ± 13.5% vs. 99.0 ± 1.7%). Using AAA, the γ values were proportional to the natural logarithm of Vair (R2 = 0.674) and inversely proportional to the natural logarithm of Vbone (R2 = 0.816). When the Vair in the targets was smaller than approximately 80 cc or the Vbone in the targets was larger than approximately 6 cc, the γ values of AAA were below 95%. Using AXB, no significant relationship was found between the γ values and Vair or Vbone. CONCLUSION: In clinical head and neck IMRT QA, greater attention should be paid to the effect of Vair and Vbone in the targets on the γ passing rates when using different dose calculation algorithms.

Multi-phase simultaneous segmentation of tumor in lung 4D-CT data with context information.

Lung 4D computed tomography (4D-CT) plays an important role in high-precision radiotherapy because it characterizes respiratory motion, which is crucial for accurate target definition. However, the manual segmentation of a lung tumor is a heavy workload for doctors because of the large number of lung 4D-CT data slices. Meanwhile, tumor segmentation is still a notoriously challenging problem in computer-aided diagnosis. In this paper, we propose a new method based on an improved graph cut algorithm with context information constraint to find a convenient and robust approach of lung 4D-CT tumor segmentation. We combine all phases of the lung 4D-CT into a global graph, and construct a global energy function accordingly. The sub-graph is first constructed for each phase. A context cost term is enforced to achieve segmentation results in every phase by adding a context constraint between neighboring phases. A global energy function is finally constructed by combining all cost terms. The optimization is achieved by solving a max-flow/min-cut problem, which leads to simultaneous and robust segmentation of the tumor in all the lung 4D-CT phases. The effectiveness of our approach is validated through experiments on 10 different lung 4D-CT cases. The comparison with the graph cut without context constraint, the level set method and the graph cut with star shape prior demonstrates that the proposed method obtains more accurate and robust segmentation results.

[Super-resolution reconstruction for 4-dimensional computed tomography of the lung using graph cuts].

Four-dimensional computer tomography (4D-CT) has a great value in lung cancer radiotherapy for its capability in providing lung information with respiratory motion. We employed a global graph cuts super-resolution (SR) reconstruction method to reconstruct high-resolution lung 4D-CT images. First, the high-resolution images reconstruction energy function was built based on a Maximum a posteriori Markov Random Field (MAP-MRF) formulation. The energy function was then transformed to a graph formulation, which was solved using graph cut algorithm. All the evaluation results showed that this approach outperformed the line interpolation and projection onto convex sets (POCS) approach with an improved structural clarity.

[Automatic Segmentation of Four Dimensional Computed Tomography of Lung Tumor Based on Star Shape Prior and Graph Cuts].

Lung four dimensional computed tomography(4D-CT)is of great value in tumor target localization and precise cancer radiotherapy.However,it is hard to segment tumors in 4D-CT data manually,since the data may contain a great number of slices with tumor.Meanwhile,auto-segmentation does not certainly guarantee the accuracy due to the complexity of images.Therefore,a new automatic segmentation technique based on Graph Cuts with star shape prior was proposed to increase automation and guarantee the accuracy of segmentation in our laboratory.Firstly,an object seed was selected in the image of initial phase and an initial target block was formed centering the selected seed.Then,the full search block-matching algorithm was adopted to obtain the most similar target block in the next phase and compute the motion field between them,and so on.Afterwards,the center seeds of each phase were obtained according to the motion fields,which would be set to the center point of star shape prior.Finally,tumors could be automatically segmented with Graph Cuts algorithm and star shape prior.Both qualitative and quantitative evaluation results showed that our approach could not only guarantee the accuracy of segmentation but also increase automation,compared with the traditional Graph Cuts algorithm.