QQ客服 官方微博 反馈留言

	Semi-supervised representation learning for infant’s biliary atresia screening using Deep CNN-based Varia-tional Autoencoder
	Li,Ling; Song,Kuiliang; Wang,Bin; Zhang,DaBao; Wan,ZhiYong; Qin,Wenjian
	2018
会议日期	2018
会议地点	昆明
英文摘要	The challenge of convolutional networks (CNNs) for medical imaging analysis is to train the network model with limited well labeled dataset. Since a variational autoencoder (VAE) is able to learn the probability distribution on data for de-scribing an observation in terms of its latent attributes by unsupervised manner, it has emerged as one of the most popular unsupervised learning technology in computer vision applications. In this paper, we presented a novel semi-supervised representation learning approach for screening infant’s biliary atresia using convolutional variational autoencoder (CVAE). Firstly, we leveraged a smartphone’s camera for infant’s stool images collection. Secondly, a pre-trained deep convolutional variation autoencoder was used to train the feature extractor for the infant’s stool image-features extraction, and finally we fine-tuned the last classification layers for identifying acholic from normal stools. We compared our screening approach with “tradition stool color card” method; the results demon-strated CVAE model has a higher accuracy rate 92.16%. Universal screening bil-iary atresia by semi-supervised representation learning may be a valuable tech-nology to help parents to recognition acholic stools in the preoperative period, which may ultimately lead to improved native liver survival probabilities.
内容类型	会议论文
源URL	[http://ir.siat.ac.cn:8080/handle/172644/14496]
专题	深圳先进技术研究院_医工所
推荐引用方式 GB/T 7714	Li,Ling,Song,Kuiliang,Wang,Bin,et al. Semi-supervised representation learning for infant’s biliary atresia screening using Deep CNN-based Varia-tional Autoencoder[C]. 见:. 昆明. 2018.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。