Varifocal-Net: A Chromosome Classification Approach Using Deep Convolutional Networks
Varifocal-Net: A Chromosome Classification Approach Using Deep Convolutional Networks
复制标题
Varifocal-Net：使用深度卷积网络的染色体分类方法
DOI：
10.1109/tmi.2019.2905841
复制
发表时间：
2019-11-01
影响因子：
10.6
通讯作者：
Yang, Guang-Zhong
中科院分区：
文献类型：
--
作者：
Qin, Yulei;Wen, Juan;Yang, Guang-Zhong
染色体分类对异常诊断的核型分析至关重要。为了加快诊断速度，我们提出了一种新的方法，称为Varifocal-Net，用于同时分类染色体的类型和极性使用深度卷积网络。该方法由一个全球规模网络（G-Net）和一个局部规模网络（L-Net）组成。它分为三个阶段。第一阶段是学习全局和局部特征。我们通过G-Net提取全局特征并检测更精细的局部区域。我们提出了一种可变机制，通过L-Net放大局部部分并提取局部特征。残差学习和多任务学习策略促进了高水平的特征提取。判别局部部分的检测由G-Net的定位子网完成，其训练过程包括监督学习和弱监督学习。第二阶段是构建两个多层感知器分类器，利用两个尺度的特征来提高分类性能。第三阶段是引入一种调度策略，通过利用核型的领域知识，在每个病例中分配每个染色体的类型。对1909例染色体组型病例的评估结果表明，所提出的Varifocal-Net在类型和极性任务上的准确率最高，为99.2。它优于最先进的方法，证明了我们的可变机制，多尺度特征集成和调度策略的有效性。该方法已用于辅助实际核型诊断。
Chromosome classification is critical for karyotyping in abnormality diagnosis. To expedite the diagnosis, we present a novel method named Varifocal-Net for simultaneous classification of chromosome's type and polarity using deep convolutional networks. The approach consists of one global-scale network (G-Net) and one local-scale network (L-Net). It follows three stages. The first stage is to learn both global and local features. We extract global features and detect finer local regions via the G-Net. By proposing a varifocal mechanism, we zoom into local parts and extract local features via the L-Net. Residual learning and multi-task learning strategies are utilized to promote high-level feature extraction. The detection of discriminative local parts is fulfilled by a localization subnet of the G-Net, whose training process involves both supervised and weakly supervised learning. The second stage is to build two multi-layer perceptron classifiers that exploit features of both two scales to boost classification performance. The third stage is to introduce a dispatch strategy of assigning each chromosome to a type within each patient case, by utilizing the domain knowledge of karyotyping. The evaluation results from 1909 karyotyping cases showed that the proposed Varifocal-Net achieved the highest accuracy per patient case () of 99.2 for both type and polarity tasks. It outperformed state-of-the-art methods, demonstrating the effectiveness of our varifocal mechanism, multi-scale feature ensemble, and dispatch strategy. The proposed method has been applied to assist practical karyotype diagnosis.