Computational Segmentation and Classification of Diabetic Glomerulosclerosis
Computational Segmentation and Classification of Diabetic Glomerulosclerosis
复制标题
DOI：
10.1681/asn.2018121259
复制
发表时间：
2019-10-01
影响因子：
13.6
通讯作者：
Sarder, Pinaki
中科院分区：
文献类型：
--
作者：
Ginley, Brandon;Lutnick, Brendon;Sarder, Pinaki
背景病理学家使用肾小球病变的视觉分类来评估糖尿病肾病（DN）患者的样本。结果可能因病理学家而异。数字算法可以减少这种变异性，并提供更一致的图像结构interpretation.Methods我们开发了一个数字管道，从DN患者的肾活检进行分类。我们将传统的图像分析与现代机器学习相结合，以有效地捕获重要结构，最大限度地减少人工工作和监督，并将生物先验信息强制应用到我们的模型中。为了计算量化肾小球结构，尽管其复杂性，我们将其简化为三个组成部分，包括细胞核，毛细血管腔和鲍曼空间;和高碘酸-希夫阳性结构。我们使用卷积神经网络从整个载玻片图像中检测肾小球边界和细胞核，并使用为此目的专门开发的无监督技术检测其余肾小球结构。我们定义了一组数字特征，量化DN的结构进展，和一个经常性的网络架构，将这些特征处理成一个classification.Results我们的数字分类同意与高级病理学家，其分类被用作地面真理与中度科恩的kappa kappa = 0.55和95%的置信区间[0.50，0.60]。另外两名肾脏病理学家同意数字分类，kappa（1）= 0.68，95%区间[0.50，0.86]和kappa（2）= 0.48，95%区间[0.32，0.64]。我们的研究结果表明，计算方法与人类视觉分类方法相当，并且可以在临床决策工作流程中提供更高的精度。我们检测肾小球边界与0.93 +/- 0.04平衡准确性，肾小球核与0.94的敏感性和0.93特异性，肾小球结构成分与0.95的敏感性和0.99 specificity.Conclusions计算得出的，组织学图像特征具有显着的诊断信息，可以增强临床诊断。
Background Pathologists use visual classification of glomerular lesions to assess samples from patients with diabetic nephropathy (DN). The results may vary among pathologists. Digital algorithms may reduce this variability and provide more consistent image structure interpretation.Methods We developed a digital pipeline to classify renal biopsies from patients with DN. We combined traditional image analysis with modern machine learning to efficiently capture important structures, minimize manual effort and supervision, and enforce biologic prior information onto our model. To computationally quantify glomerular structure despite its complexity, we simplified it to three components consisting of nuclei, capillary lumina and Bowman spaces; and Periodic Acid-Schiff positive structures. We detected glomerular boundaries and nuclei from whole slide images using convolutional neural networks, and the remaining glomerular structures using an unsupervised technique developed expressly for this purpose. We defined a set of digital features which quantify the structural progression of DN, and a recurrent network architecture which processes these features into a classification.Results Our digital classification agreed with a senior pathologist whose classifications were used as ground truth with moderate Cohen's kappa kappa = 0.55 and 95% confidence interval [0.50, 0.60]. Two other renal pathologists agreed with the digital classification with kappa(1) = 0.68, 95% interval [0.50, 0.86] and kappa(2) = 0.48, 95% interval [0.32, 0.64]. Our results suggest computational approaches are comparable to human visual classification methods, and can offer improved precision in clinical decision workflows. We detected glomerular boundaries from whole slide images with 0.93 +/- 0.04 balanced accuracy, glomerular nuclei with 0.94 sensitivity and 0.93 specificity, and glomerular structural components with 0.95 sensitivity and 0.99 specificity.Conclusions Computationally derived, histologic image features hold significant diagnostic information that may augment clinical diagnostics.