Implementation of the structural SIMilarity (SSIM) index as a quantitative evaluation tool for dose distribution error detection
Implementation of the structural SIMilarity (SSIM) index as a quantitative evaluation tool for dose distribution error detection
复制标题
DOI：
10.1002/mp.14010
复制
发表时间：
2020-01-28
期刊：
影响因子：
3.8
通讯作者：
Cai, Bin
中科院分区：
文献类型：
--
作者：
Peng, Jiayuan;Shi, Chengyu;Cai, Bin
目的将结构相似性指数（SSIM）应用于放射治疗剂量验证领域，评价其对不同剂量分布误差的反映能力。方法SSIM指数由亮度、对比度和结构三个子指数组成。给定两幅图像，亮度分析比较局部平均结果，对比度分析比较局部标准差，而结构指数表示局部Pearson相关性。设计了三种测试误差模式（绝对剂量误差、剂量梯度误差和剂量结构误差），以表征SSIM及其子指标的响应，并建立指标与不同剂量误差类型之间的相关性。在建立相关性后，通过计算每个指标对四个放射治疗计划（一个MLC栅栏测试计划，一个脑立体定向放射治疗计划和两个头颈部计划）进行测试，并与伽马分析结果进行比较，以确定它们的异同。结果在3种测试误差模式中，当绝对剂量一致性从100%下降到5%时，亮度指数从1下降到0.1;当剂量梯度一致性从100%下降到10%时，对比度指数从1下降到0.36;因此，亮度、对比度和结构指数可以分别检测绝对剂量误差、梯度差异和剂量结构误差。对于四个临床病例，当伽马分析仅提供有限信息时，子指标可以揭示错误的类型。结论建立了SSIM指数各分量与剂量分布误差类型之间的相关性。SSIM指数提供了额外的错误信息相比，伽马分析提供。
Purpose To apply an imaging metric of the structural SIMilarity (SSIM) index to the radiotherapy dose verification field and evaluate its capability to reveal the different types of errors between two dose distributions. Method The SSIM index consists of three sub-indices: luminance, contrast, and structure. Given two images, luminance analysis compares the local mean result, contrast analysis compares the local standard deviation, and the structure index represents the local Pearson correlation. Three test error patterns (absolute dose error, dose gradient error, and dose structure error) were designed to characterize the response of SSIM and its sub-indices and establish the correlation between the indices and different dose error types. After establishing the correlation, four radiotherapy plans (one MLC picket-fence test plan, one brain stereotactic radiotherapy plan, and two head-and-neck plans) were tested by computing each index and compared with the gamma analysis results to determine their similarities and differences. Results Among the three test error patterns, the luminance index decreased from 1 to 0.1 when the absolute dose agreement fell from 100% to 5%, the contrast index decreased from 1 to 0.36 when the dose gradient agreement fell from 100% to 10%, and the structure index decreased from 1 to 0.23 when the periodical dose pattern shifted (leading to a lower correlation). Thus, the luminance, contrast and structure index can detect the absolute dose error, gradient discrepancy, and dose structure error, respectively. For the four clinical cases, the sub-indices can reveal the type of error when gamma analysis only provided limited information. Conclusions The correlation between the subcomponents of the SSIM index and the error types of the dose distribution were established. The SSIM index provides additional error information compared to that provided by gamma analysis.