文献速递泌尿生殖腹部CTUS

使用卷积神经网络对肾脏肿瘤进行分类以检测嗜酸细胞瘤

Classification of renal tumour using convolutional neural networks to detect oncocytoma.
PubMed 文献 · Eur J Radiol · 2020-10-11

本研究收集2015年1月至2018年9月369例患者的20000张二维CT图像,采用改良ResNet50V2卷积神经网络在患者层面区分嗜酸细胞瘤与肾细胞癌;独立测试集DataTest_1的AUC为0.973、准确率93.3%、特异度93.5%,小数据集DataTest_2逐图评估AUC为0.946、准确率90.0%、特异度98.0%;按每例患者51%多数票判定时准确率升至100%、假阴性降为零。作者认为该深度学习方法能以高准确率将肾脏肿瘤分类为嗜酸细胞瘤,可能避免肾肿块患者的过度治疗。

要点

  • 研究在患者层面而非单张图像层面划分数据集,避免同一患者图像跨训练与测试集。
  • DataTest_1的AUC为0.973、准确率93.3%、特异度93.5%。
  • 按每例患者51%多数票判定时准确率升至100%,假阴性降为零。

核心结论

  • 卷积神经网络和深度学习技术能以高准确率将肾脏肿瘤肿块分类为嗜酸细胞瘤。
  • 作者认为该诊断方法可能避免肾肿块患者的过度治疗。

研究设计与样本

  • 研究收集2015年1月至2018年9月369例患者的数据,共20000张二维CT图像。
  • 良恶性真实标签由术后组织学结果或超声引导下经皮穿刺活检确定。
  • 数据随机分为训练集70%、验证集10%和独立测试集20%(DataTest_1),另用小型数据集DataTest_2对训练模型做额外验证。
  • 数据按患者层面而非单张图像层面划分数据集。
  • 采用改良版ResNet50V2模型。

诊断标准与关键数值

  • DataTest_1测试结果:AUC为0.973,准确率93.3%,特异度93.5%。
  • DataTest_2逐图评估结果:AUC为0.946,准确率90.0%,特异度98.0%。
  • 未出现多张假阴性图像来自同一患者的情况。
  • 按每例患者51%多数票评估时,准确率升至100%,假阴性发生率降至零。

临床应用与阅片要点

  • 模型以患者层面51%多数票整合逐图分类结果,可作为阅片时整例判读的参考方式。
  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

PURPOSE: To investigate the ability of convolutional neural networks (CNNs) to facilitate differentiation of oncocytoma from renal cell carcinoma (RCC) using non-invasive imaging technology.

METHODS: Data were collected from 369 patients between January 2015 and September 2018. True labelling of scans as benign or malignant was determined by subsequent histological findings post-surgery or ultrasound-guided percutaneous biopsy. The data included 20,000 2D CT images. Data were randomly divided into sets for training (70 %), validation (10 %) and independent testing (20 %, DataTest_1). A small dataset (DataTest_2) was used for additional validation of the training model. Data were divided into sets at the patient level, rather than by individual image. A modified version of the ResNet50V2 was used. Accuracy of detecting benign or malignant renal mass was evaluated by a 51 % majority vote of individual image classifications to determine the classification for each patient.

RESULTS: Test results from DataTest_1 indicate an area under the curve (AUC) of 0.973 with 93.3 % accuracy and 93.5 % specificity. Results from DataTest_2 indicate an AUC of 0.946 with 90.0 % accuracy and 98.0 % specificity when evaluation is performed image by image. There is no case in which multiple false negative images originate from the same patient. When evaluated with 51 % majority of scans for each patient, the accuracy rises to 100 % and the incidence of false negatives falls to zero.

CONCLUSION: CNNs and deep learning technology can classify renal tumour masses as oncocytoma with high accuracy. This diagnostic method could prevent overtreatment for patients with renal masses.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Pedersen M, Andersen MB, Christiansen H, et al. Classification of renal tumour using convolutional neural networks to detect oncocytoma. Eur J Radiol. 2020. PMID: 33120238. doi:10.1016/j.ejrad.2020.109343

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。