当前位置: X-MOL 学术Artif. Intell. Med. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Classification of glomerular hypercellularity using convolutional features and support vector machine.
Artificial Intelligence in Medicine ( IF 7.5 ) Pub Date : 2020-01-25 , DOI: 10.1016/j.artmed.2020.101808
Paulo Chagas 1 , Luiz Souza 1 , Ikaro Araújo 2 , Nayze Aldeman 3 , Angelo Duarte 4 , Michele Angelo 4 , Washington L C Dos-Santos 5 , Luciano Oliveira 1
Affiliation  

Glomeruli are histological structures of the kidney cortex formed by interwoven blood capillaries, and are responsible for blood filtration. Glomerular lesions impair kidney filtration capability, leading to protein loss and metabolic waste retention. An example of lesion is the glomerular hypercellularity, which is characterized by an increase in the number of cell nuclei in different areas of the glomeruli. Glomerular hypercellularity is a frequent lesion present in different kidney diseases. Automatic detection of glomerular hypercellularity would accelerate the screening of scanned histological slides for the lesion, enhancing clinical diagnosis. Having this in mind, we propose a new approach for classification of hypercellularity in human kidney images. Our proposed method introduces a novel architecture of a convolutional neural network (CNN) along with a support vector machine, achieving near perfect average results on FIOCRUZ data set in a binary classification (lesion or normal). Additionally, classification of hypercellularity sub-lesions was also evaluated, considering mesangial, endocapilar and both lesions, reaching an average accuracy of 82%. Either in binary task or in the multi-classification one, our proposed method outperformed Xception, ResNet50 and InceptionV3 networks, as well as a traditional handcrafted-based method. To the best of our knowledge, this is the first study on deep learning over a data set of glomerular hypercellularity images of human kidney.



中文翻译:

使用卷积特征和支持向量机对肾小球高细胞性进行分类。

肾小球是由交织的毛细血管形成的肾皮质的组织学结构,负责血液过滤。肾小球病变损害肾脏滤过能力,导致蛋白质损失和代谢废物滞留。病变的一个例子是肾小球高细胞性,其特征是肾小球不同区域的细胞核数目增加。肾小球细胞肥大是在不同肾脏疾病中常见的病变。肾小球高细胞性的自动检测将加速对病变的扫描组织切片的筛选,从而增强临床诊断。考虑到这一点,我们提出了一种新的分类人类肾脏图像中细胞过多的方法。我们提出的方法引入了卷积神经网络(CNN)的新颖架构以及支持向量机,从而以二进制分类(病变或正常)对FIOCRUZ数据集实现了近乎完美的平均结果。此外,还考虑了肾小球系膜,毛细血管内膜和两个皮损,对高细胞性亚皮损的分类进行了评估,平均准确率达82%。无论是在二进制任务中还是在多分类任务中,我们提出的方法都优于Xception,ResNet50和InceptionV3网络,以及传统的基于手工的方法。据我们所知,这是对人类肾脏的肾小球高细胞性图像数据集进行深度学习的第一项研究。在二进制分类(病变或正常)的FIOCRUZ数据集上获得接近完美的平均结果。此外,还考虑了肾小球系膜,毛细血管内膜和两个皮损,对高细胞性亚皮损的分类进行了评估,平均准确率达82%。无论是在二进制任务中还是在多分类任务中,我们提出的方法都优于Xception,ResNet50和InceptionV3网络,以及传统的基于手工的方法。据我们所知,这是对人类肾脏的肾小球高细胞性图像数据集进行深度学习的第一项研究。在二进制分类(病变或正常)的FIOCRUZ数据集上获得接近完美的平均结果。此外,还考虑了肾小球系膜,毛细血管内膜和两个皮损,对高细胞性亚皮损的分类进行了评估,平均准确率达82%。无论是在二进制任务中还是在多分类任务中,我们提出的方法都优于Xception,ResNet50和InceptionV3网络,以及传统的基于手工的方法。据我们所知,这是对人类肾脏的肾小球高细胞性图像数据集进行深度学习的第一项研究。我们提出的方法优于Xception,ResNet50和InceptionV3网络,以及传统的基于手工方法。据我们所知,这是对人类肾脏的肾小球高细胞性图像数据集进行深度学习的第一项研究。我们提出的方法优于Xception,ResNet50和InceptionV3网络,以及传统的基于手工方法。据我们所知,这是对人类肾脏的肾小球高细胞性图像数据集进行深度学习的第一项研究。

更新日期:2020-01-25
down
wechat
bug