Algohary Ahmad, Viswanath Satish, Shiradkar Rakesh, Ghose Soumya, Pahwa Shivani, Moses Daniel, Jambor Ivan, Shnier Ronald, Böhm Maret, Haynes Anne-Maree, Brenner Phillip, Delprado Warick, Thompson James, Pulbrock Marley, Purysko Andrei S, Verma Sadhna, Ponsky Lee, Stricker Phillip, Madabhushi Anant
Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio, USA.
Department of Radiology, Case Western Reserve University, Cleveland, Ohio, USA.
J Magn Reson Imaging. 2018 Feb 22. doi: 10.1002/jmri.25983.
Radiomic analysis is defined as computationally extracting features from radiographic images for quantitatively characterizing disease patterns. There has been recent interest in examining the use of MRI for identifying prostate cancer (PCa) aggressiveness in patients on active surveillance (AS).
To evaluate the performance of MRI-based radiomic features in identifying the presence or absence of clinically significant PCa in AS patients.
Retrospective.
MRI/TRUS (transperineal grid ultrasound) fusion-guided biopsy was performed for 56 PCa patients on AS who had undergone prebiopsy.
FIELD STRENGTH/SEQUENCE: 3T, T -weighted (T w) and diffusion-weighted (DW) MRI.
A pathologist histopathologically defined the presence of clinically significant disease. A radiologist manually delineated lesions on T w-MRs. Then three radiologists assessed MRIs using PIRADS v2.0 guidelines. Tumors were categorized into four groups: MRI-negative-biopsy-negative (Group 1, N = 15), MRI-positive-biopsy-positive (Group 2, N = 16), MRI-negative-biopsy-positive (Group 3, N = 10), and MRI-positive-biopsy-negative (Group 4, N = 15). In all, 308 radiomic features (First-order statistics, Gabor, Laws Energy, and Haralick) were extracted from within the annotated lesions on T w images and apparent diffusion coefficient (ADC) maps. The top 10 features associated with clinically significant tumors were identified using minimum-redundancy-maximum-relevance and used to construct three machine-learning models that were independently evaluated for their ability to identify the presence and absence of clinically significant disease.
Wilcoxon rank-sum tests with P < 0.05 considered statistically significant.
Seven T w-based (First-order Statistics, Haralick, Laws, and Gabor) and three ADC-based radiomic features (Laws, Gradient and Sobel) exhibited statistically significant differences (P < 0.001) between malignant and normal regions in the training groups. The three constructed models yielded overall accuracy improvement of 33, 60, 80% and 30, 40, 60% for patients in testing groups, when compared to PIRADS v2.0 alone.
Radiomic features could help in identifying the presence and absence of clinically significant disease in AS patients when PIRADS v2.0 assessment on MRI contradicted pathology findings of MRI-TRUS prostate biopsies.
3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.
放射组学分析被定义为通过计算从放射影像中提取特征以定量表征疾病模式。近期，人们对利用磁共振成像（MRI）来识别接受主动监测（AS）患者的前列腺癌（PCa）侵袭性产生了兴趣。
评估基于MRI的放射组学特征在识别AS患者中临床显著性PCa存在与否方面的性能。
回顾性研究。
对56例接受AS且已进行活检前检查的PCa患者进行了MRI/经会阴网格超声（TRUS）融合引导活检。
场强/序列：3T，T加权（T w）和扩散加权（DW）MRI。
一名病理学家通过组织病理学确定临床显著性疾病的存在。一名放射科医生在T w-MR图像上手动勾勒病变。然后三名放射科医生使用PIRADS v2.0指南评估MRI。肿瘤被分为四组：MRI阴性-活检阴性（第1组，N = 15），MRI阳性-活检阳性（第2组，N = 16），MRI阴性-活检阳性（第3组，N = 10），以及MRI阳性-活检阴性（第4组，N = 15）。总共从T w图像和表观扩散系数（ADC）图上的标注病变内提取了308个放射组学特征（一阶统计量、Gabor、Laws能量和Haralick）。使用最小冗余最大相关性确定与临床显著性肿瘤相关的前10个特征，并用于构建三个机器学习模型，独立评估它们识别临床显著性疾病存在与否的能力。
Wilcoxon秩和检验，P < 0.05被认为具有统计学显著性。
在训练组中，基于T w的7个特征（一阶统计量、Haralick、Laws和Gabor）以及基于ADC的3个放射组学特征（Laws、梯度和Sobel）在恶性和正常区域之间表现出统计学显著性差异（P < 0.001）。与单独使用PIRADS v2.0相比，构建的三个模型在测试组患者中的总体准确率分别提高了33%、60%、80%以及30%、40%、60%。
当MRI上的PIRADS v2.0评估与MRI-TRUS前列腺活检的病理结果相矛盾时，放射组学特征有助于识别AS患者中临床显著性疾病的存在与否。
3 技术效能：2级 J.Magn.Reson.Imaging 2018年
