Department of Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany.
Department of Neuroradiology, University of Heidelberg Medical Center, Heidelberg, Germany.
J Magn Reson Imaging. 2017 Aug;46(2):604-616. doi: 10.1002/jmri.25606. Epub 2017 Feb 2.
To assess radiomics as a tool to determine how well lesions found suspicious on breast cancer screening X-ray mammography can be categorized into malignant and benign with unenhanced magnetic resonance (MR) mammography with diffusion-weighted imaging and T -weighted sequences.
From an asymptomatic screening cohort, 50 women with mammographically suspicious findings were examined with contrast-enhanced breast MRI (ceMRI) at 1.5T. Out of this protocol an unenhanced, abbreviated diffusion-weighted imaging protocol (ueMRI) including T -weighted, (T w), diffusion-weighted imaging (DWI), and DWI with background suppression (DWIBS) sequences and corresponding apparent diffusion coefficient (ADC) maps were extracted. From ueMRI-derived radiomic features, three Lasso-supervised machine-learning classifiers were constructed and compared with the clinical performance of a highly experienced radiologist: 1) univariate mean ADC model, 2) unconstrained radiomic model, 3) constrained radiomic model with mandatory inclusion of mean ADC.
The unconstrained and constrained radiomic classifiers consisted of 11 parameters each and achieved differentiation of malignant from benign lesions with a .632 + bootstrap receiver operating characteristics (ROC) area under the curve (AUC) of 84.2%/85.1%, compared to 77.4% for mean ADC and 95.9%/95.9% for the experienced radiologist using ceMRI/ueMRI.
In this pilot study we identified two ueMRI radiomics classifiers that performed well in the differentiation of malignant from benign lesions and achieved higher performance than the mean ADC parameter alone. Classification was lower than the almost perfect performance of a highly experienced breast radiologist. The potential of radiomics to provide a training-independent diagnostic decision tool is indicated. A performance reaching the human expert would be highly desirable and based on our results is considered possible when the concept is extended in larger cohorts with further development and validation of the technique.
1 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:604-616.
评估放射组学作为一种工具，通过对乳腺 X 线摄影筛查中可疑病变进行无增强磁共振（MR）乳腺成像弥散加权成像和 T -加权序列检查，以确定其良恶性的能力。
从无症状的筛查队列中，对 50 名乳腺 X 线摄影可疑的女性患者在 1.5T 下进行对比增强乳腺 MRI（ceMRI）检查。从该方案中提取出无增强、简化的弥散加权成像方案（ueMRI），包括 T -加权（T w）、弥散加权成像（DWI）、背景抑制弥散加权成像（DWIBS）序列和相应的表观扩散系数（ADC）图。从 ueMRI 衍生的放射组学特征中，构建了三个 Lasso 监督机器学习分类器，并与一位经验丰富的放射科医生的临床表现进行了比较：1）单变量平均 ADC 模型，2）无约束放射组学模型，3）强制性纳入平均 ADC 的约束放射组学模型。
无约束和约束放射组学分类器分别由 11 个参数组成，恶性与良性病变的区分达到了 0.632 + bootstrap 接收者操作特征（ROC）曲线下面积（AUC）84.2%/85.1%，而平均 ADC 的 AUC 为 77.4%，经验丰富的放射科医生使用 ceMRI/ueMRI 的 AUC 为 95.9%/95.9%。
在这项初步研究中，我们确定了两种 ueMRI 放射组学分类器，它们在恶性与良性病变的区分中表现良好，性能优于平均 ADC 参数。分类低于经验丰富的乳腺放射科医生近乎完美的表现。放射组学提供一种独立于训练的诊断决策工具的潜力得到了证实。如果将该概念扩展到更大的队列中，并进一步开发和验证该技术，达到人类专家的水平是非常可取的。
1 技术功效：2 阶段 J. MAGN. RESON. IMAGING 2017;46:604-616.
