文献速递乳腺OCT

基于卷积神经网络的光学相干断层扫描图像分类方法实现乳房肿块切除术后切缘全自动评估

Fully Automated Postlumpectomy Breast Margin Assessment Utilizing Convolutional Neural Network Based Optical Coherence Tomography Image Classification Method.
PubMed 文献 · Acad Radiol · 2019-07-17

本研究开发基于卷积神经网络(CNN)的深度学习方法,对超高分辨率光学相干断层扫描(UHR-OCT)图像中乳腺组织癌性与非癌性区域进行分类,用于术中切缘评估。CNN对癌与非癌的二分类达到94%准确率、96%敏感度和92%特异度。作者认为该全自动方法可行,可克服人工判读的观察者间变异和速度限制,可能实现实时术中切缘评估。

要点

  • CNN对OCT图像癌与非癌二分类达94%准确率、96%敏感度、92%特异度。
  • 四类组织分类中浸润性导管癌F1最高(0.89±0.09),导管原位癌最低(0.65±0.15)。
  • 该系统采用轴向分辨率2.7 μm、横向分辨率5.5 μm的UHR-OCT,作者认为可能实现实时术中切缘评估。

核心结论

  • 使用基于CNN的算法准确区分OCT图像中的癌性区域是可行的,该全自动方法可克服人工判读的观察者间变异和判读速度等局限,并可能实现实时术中切缘评估。

研究设计与样本

  • 本研究为深度学习分类方法开发研究,采用定制超高分辨率OCT系统(轴向分辨率2.7 μm、横向分辨率5.5 μm),算法基于A-scan分类方案,CNN为11层架构、由连续的3×3卷积核组成,分类四类组织:脂肪、间质、导管原位癌和浸润性导管癌。

关键影像表现/征象

  • OCT图像分类涵盖脂肪、间质、导管原位癌和浸润性导管癌四类组织。
  • 癌与非癌二分类的CNN达到94%准确率、96%敏感度和92%特异度。
  • 五折交叉验证平均F1评分以浸润性导管癌最高(0.89±0.09),脂肪次之(0.79±0.17),间质为0.74±0.18,导管原位癌最低(0.65±0.15)。

诊断标准与关键数值

  • 癌与非癌二分类:准确率94%、敏感度96%、特异度92%。
  • 五折交叉验证平均F1评分:浸润性导管癌0.89±0.09、脂肪0.79±0.17、间质0.74±0.18、导管原位癌0.65±0.15。

临床应用与阅片要点

  • 作者认为该全自动CNN方法可克服人工判读的观察者间变异和判读速度限制,并可能实现实时术中切缘评估。

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

BACKGROUND: The purpose of this study was to develop a deep learning classification approach to distinguish cancerous from noncancerous regions within optical coherence tomography (OCT) images of breast tissue for potential use in an intraoperative setting for margin assessment.

METHODS: A custom ultrahigh-resolution OCT (UHR-OCT) system with an axial resolution of 2.7 μm and a lateral resolution of 5.5 μm was used in this study. The algorithm used an A-scan-based classification scheme and the convolutional neural network (CNN) was implemented using an 11-layer architecture consisting of serial 3 × 3 convolution kernels. Four tissue types were classified, including adipose, stroma, ductal carcinoma in situ, and invasive ductal carcinoma.

RESULTS: The binary classification of cancer versus noncancer with the proposed CNN achieved 94% accuracy, 96% sensitivity, and 92% specificity. The mean five-fold validation F1 score was highest for invasive ductal carcinoma (mean standard deviation, 0.89 ± 0.09) and adipose (0.79 ± 0.17), followed by stroma (0.74 ± 0.18), and ductal carcinoma in situ (0.65 ± 0.15).

CONCLUSION: It is feasible to use CNN based algorithm to accurately distinguish cancerous regions in OCT images. This fully automated method can overcome limitations of manual interpretation including interobserver variability and speed of interpretation and may enable real-time intraoperative margin assessment.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Mojahed D, Ha RS, Chang P, et al. Fully Automated Postlumpectomy Breast Margin Assessment Utilizing Convolutional Neural Network Based Optical Coherence Tomography Image Classification Method. Acad Radiol. 2019. PMID: 31324579. doi:10.1016/j.acra.2019.06.018

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。