Multisite evaluation of radiomic feature reproducibility and discriminability for identifying peripheral zone prostate tumors on MRI
Multisite evaluation of radiomic feature reproducibility and discriminability for identifying peripheral zone prostate tumors on MRI
复制标题
DOI：
10.1117/1.jmi.6.2.024502
复制
发表时间：
2019-04-01
影响因子：
2.4
通讯作者：
Viswanath, Satish E.
中科院分区：
文献类型：
--
作者：
Chirra, Prathyush;Leo, Patrick;Viswanath, Satish E.
放射组学领域的最新进展已经使得能够开发用于各种疾病的许多基于预后和预测成像的工具。然而，这些工具在临床上的广泛采用取决于它们在多个站点和扫描仪上的通用性。这在从T1或T2加权磁共振图像（MRI）导出的放射组学特征的上下文中可能特别相关，其中已知信号强度值缺乏组织特异性意义并且基于机构之间的不同采集协议而变化。我们提出了第一个实证研究基准五个不同的放射组学特征的家庭在多站点设置的再现性和可辨别性，特别是识别前列腺肿瘤在MRI上的外周区。我们的队列包括来自四个不同站点的147名患者的T2加权MRI数据集，所有数据集均首先进行预处理，以校正采集相关伪影，例如偏场、不同体素分辨率和强度漂移（非标准化）。在跨部位设置中评估了来自五个不同家族（灰色、Haralick、梯度、Laws和Gabor）的约406个三维体素放射组学特征，以确定（a）它们在相对均匀的非肿瘤组织区域内的可重复性如何以及（B）它们如何区分肿瘤区域和非肿瘤区域。我们的研究结果表明，大多数流行的Haralick特征在超过99%的跨站点比较中是可重复的，并且实现了出色的跨站点可分辨性（分类精度约为0.8）。相比之下，大多数Laws特征在站点之间具有高度可变性（在
Recent advances in the field of radiomics have enabled the development of a number of prognostic and predictive imaging-based tools for a variety of diseases. However, wider clinical adoption of these tools is contingent on their generalizability across multiple sites and scanners. This may be particularly relevant in the context of radiomic features derived from T1- or T2-weighted magnetic resonance images (MRIs), where signal intensity values are known to lack tissue-specific meaning and vary based on differing acquisition protocols between institutions. We present the first empirical study of benchmarking five different radiomic feature families in terms of both reproducibility and discriminability in a multisite setting, specifically, for identifying prostate tumors in the peripheral zone on MRI. Our cohort comprised 147 patient T2-weighted MRI datasets from four different sites, all of which are first preprocessed to correct for acquisition-related artifacts such as bias field, differing voxel resolutions, and intensity drift (nonstandardness). About 406 three-dimensional voxelwise radiomic features from five different families (gray, Haralick, gradient, Laws, and Gabor) were evaluated in a cross-site setting to determine (a) how reproducible they are within a relatively homogeneous nontumor tissue region and (b) how well they could discriminate tumor regions from nontumor regions. Our results demonstrate that a majority of the popular Haralick features are reproducible in over 99% of all cross-site comparisons, as well as achieve excellent cross-site discriminability (classification accuracy of approximate to 0.8). By contrast, a majority of Laws features are highly variable across sites (reproducible in