Department of Physiology, Pharmacology and Medical Physics, Alborz University of Medical Sciences, Karaj, Iran.
Department of Biomedical Engineering and Medical Physics, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Med Phys. 2021 Jul;48(7):3691-3701. doi: 10.1002/mp.14896. Epub 2021 May 17.
We evaluate the feasibility of treatment response prediction using MRI-based pre-, post-, and delta-radiomic features for locally advanced rectal cancer (LARC) patients treated by neoadjuvant chemoradiation therapy (nCRT).
This retrospective study included 53 LARC patients divided into a training set (Center#1, n = 36) and external validation set (Center#2, n = 17). T2-weighted (T2W) MRI was acquired for all patients, 2 weeks before and 4 weeks after nCRT. Ninety-six radiomic features, including intensity, morphological and second- and high-order texture features were extracted from segmented 3D volumes from T2W MRI. All features were harmonized using ComBat algorithm. Max-Relevance-Min-Redundancy (MRMR) algorithm was used as feature selector and k-nearest neighbors (KNN), Naïve Bayes (NB), Random forests (RF), and eXtreme Gradient Boosting (XGB) algorithms were used as classifiers. The evaluation was performed using the area under the receiver operator characteristic (ROC) curve (AUC), sensitivity, specificity and accuracy.
In univariate analysis, the highest AUC in pre-, post-, and delta-radiomic features were 0.78, 0.70, and 0.71, for GLCM_IMC1, shape (surface area and volume) and GLSZM_GLNU features, respectively. In multivariate analysis, RF and KNN achieved the highest AUC (0.85 ± 0.04 and 0.81 ± 0.14, respectively) among pre- and post-treatment features. The highest AUC was achieved for the delta-radiomic-based RF model (0.96 ± 0.01) followed by NB (0.96 ± 0.04). Overall. Delta-radiomics model, outperformed both pre- and post-treatment features (P-value <0.05).
Multivariate analysis of delta-radiomic T2W MRI features using machine learning algorithms could potentially be used for response prediction in LARC patients undergoing nCRT. We also observed that multivariate analysis of delta-radiomic features using RF classifiers can be used as powerful biomarkers for response prediction in LARC.
我们评估使用基于 MRI 的治疗前、后和 delta 放射组学特征对接受新辅助放化疗（nCRT）的局部晚期直肠癌（LARC）患者进行治疗反应预测的可行性。
这项回顾性研究纳入了 53 例 LARC 患者，分为训练集（中心#1，n=36）和外部验证集（中心#2，n=17）。所有患者均接受 T2 加权（T2W）MRI 扫描，在 nCRT 前 2 周和后 4 周进行。从 T2W MRI 分割的 3D 容积中提取 96 个放射组学特征，包括强度、形态和二阶和高阶纹理特征。使用 ComBat 算法对所有特征进行协调。使用最大相关性最小冗余度（MRMR）算法作为特征选择器，k-最近邻（KNN）、朴素贝叶斯（NB）、随机森林（RF）和极端梯度提升（XGB）算法作为分类器。使用接收器操作特征（ROC）曲线下面积（AUC）、敏感性、特异性和准确性进行评估。
在单变量分析中，治疗前、后和 delta 放射组学特征中 AUC 值最高的分别是 GLCM_IMC1、形状（表面积和体积）和 GLSZM_GLNU 特征，分别为 0.78、0.70 和 0.71。在多变量分析中，RF 和 KNN 在治疗前和治疗后特征中均获得了最高的 AUC（分别为 0.85±0.04 和 0.81±0.14）。基于 delta 放射组学的 RF 模型的 AUC 最高（0.96±0.01），其次是 NB（0.96±0.04）。总的来说，delta 放射组学模型优于治疗前和治疗后特征（P 值<0.05）。
使用机器学习算法对 T2W MRI 基于 delta 放射组学的多变量分析可能可用于预测接受 nCRT 的 LARC 患者的治疗反应。我们还观察到，使用 RF 分类器对 delta 放射组学特征进行多变量分析可用作预测 LARC 患者治疗反应的有力生物标志物。
