Wentland Andrew L, Yamashita Rikiya, Kino Aya, Pandit Prachi, Shen Luyao, Brooke Jeffrey R, Rubin Daniel, Kamaya Aya
Department of Radiology, University of Wisconsin School of Medicine & Public Health, 600 Highland Avenue, Madison, WI, 53792, USA.
Department of Medical Physics, University of Wisconsin School of Medicine & Public Health, Madison, WI, USA.
Abdom Radiol (NY). 2023 Feb;48(2):642-648. doi: 10.1007/s00261-022-03735-7. Epub 2022 Nov 12.
To assess the performance of a machine learning model trained with contrast-enhanced CT-based radiomics features in distinguishing benign from malignant solid renal masses and to compare model performance with three abdominal radiologists.
Patients who underwent intra-operative ultrasound during a partial nephrectomy were identified within our institutional database, and those who had pre-operative contrast-enhanced CT examinations were selected. The renal masses were segmented from the CT images and radiomics features were derived from the segmentations. The pathology of each mass was identified; masses were labeled as either benign [oncocytoma or angiomyolipoma (AML)] or malignant [clear cell, papillary, or chromophobe renal cell carcinoma (RCC)] depending on the pathology. The data were parsed into a 70/30 train/test split and a random forest machine learning model was developed to distinguish benign from malignant lesions. Three radiologists assessed the cohort of masses and labeled cases as benign or malignant.
148 masses were identified from the cohort, including 50 benign lesions (23 AMLs, 27 oncocytomas) and 98 malignant lesions (23 clear cell RCC, 44 papillary RCC, and 31 chromophobe RCCs). The machine learning algorithm yielded an overall accuracy of 0.82 for distinguishing benign from malignant lesions, with an area under the receiver operating curve of 0.80. In comparison, the three radiologists had significantly lower accuracies (p = 0.02) ranging from 0.67 to 0.75.
A machine learning model trained with CT-based radiomics features can provide superior accuracy for distinguishing benign from malignant solid renal masses compared to abdominal radiologists.
评估基于对比增强CT的放射组学特征训练的机器学习模型在区分肾实性肿块良恶性方面的性能，并将模型性能与三位腹部放射科医生进行比较。
在我们的机构数据库中识别出在部分肾切除术中接受术中超声检查的患者，并选择那些术前进行了对比增强CT检查的患者。从CT图像中分割出肾肿块，并从分割结果中提取放射组学特征。确定每个肿块的病理；根据病理结果将肿块标记为良性[嗜酸细胞瘤或血管平滑肌脂肪瘤（AML）]或恶性[透明细胞、乳头状或嫌色肾细胞癌（RCC）]。将数据解析为70/30的训练/测试分割，并开发了一个随机森林机器学习模型来区分良性和恶性病变。三位放射科医生评估了肿块队列，并将病例标记为良性或恶性。
从队列中识别出148个肿块，包括50个良性病变（23个AML、27个嗜酸细胞瘤）和98个恶性病变（23个透明细胞RCC、44个乳头状RCC和31个嫌色RCC）。机器学习算法区分良性和恶性病变的总体准确率为0.82，受试者操作特征曲线下面积为0.80。相比之下，三位放射科医生的准确率显著较低（p = 0.02），范围为0.67至0.75。
与腹部放射科医生相比，基于CT的放射组学特征训练的机器学习模型在区分肾实性肿块良恶性方面可提供更高的准确率。
