Visibility Attribute Extraction and Anomaly Detection for Chinese Diagnostic Report Based on Cascade Networks
Visibility Attribute Extraction and Anomaly Detection for Chinese Diagnostic Report Based on Cascade Networks
复制标题
基于级联网络的中文诊断报告可见性属性提取与异常检测
DOI：
10.1109/access.2019.2932842
复制
发表时间：
2019
期刊：
影响因子：
3.9
通讯作者：
Siqi Li
中科院分区：
文献类型：
--
作者：
Jitong Zhang;Huiyan Jiang;Liangliang Huang;Yu-dong Yao;Siqi Li
在正电子发射断层扫描/计算机断层扫描（PET/CT）图像诊断报告中，图像表现部分的语义分析是医学图像自动诊断的重要组成部分，是提取诊断报告中的关键词和异常句子的重要步骤。为此，本文将可见性属性提取网络（vee - net）和双向门控循环单元（BiGRU）结合成级联网络，解决属性提取和异常检测的任务。首先，定义可见性属性（VA），根据图像发现中的语言特征将词汇总结为12种模式；其次，开发了残差卷积神经网络（residual CNN）、BiGRU和条件随机场（conditional random field， CRF）组成的可见性属性提取网络（vee - net），用于从词嵌入中自动提取可见性属性。最后，将词嵌入和相应的VA输入到BiGRU和softmax中进行句子级异常检测。在中文PET/CT诊断报告数据集上对该方法进行了评价，属性提取的f1得分为94.35%，句子级异常检测的f1得分为96.40%，病例级异常检测的f1得分为96.77%。此外，使用国家生物技术信息中心（NCBI）公开的英文疾病语料库数据集进行外部验证，疾病检测的f1得分为95.81%。实验结果表明，与其他相关方法相比，所提出的级联网络具有优势。
In the positron emission tomography/computed tomography (PET/CT) image diagnosis report, the semantic analysis of image findings section is an important part of the automatic diagnosis of medical image, which is an essential step for extracting keywords and abnormal sentences in the diagnostic report. To this end, this paper combines visibility attribute extraction network (VAE-Net) and bi-directional gated recurrent unit (BiGRU) into cascade networks to solve the tasks of attribute extraction and anomaly detection. First, a visibility attribute (VA) is defined to summary the vocabulary into 12 patterns based on the language characteristics in image findings. Second, a visibility attribute extraction network (VAE-Net) is developed to automatically extract VA from word embeddings, which is composed of residual convolutional neural network (residual CNN), BiGRU, and conditional random field (CRF). Finally, word embeddings and the corresponding VA are input into BiGRU and softmax to perform sentence-level anomaly detections. We evaluate the proposed method on a proprietary Chinese PET/CT diagnostic report dataset with an F1-score of 94.35% in the attribute extraction, an F1-score of 96.40% in sentence-level anomaly detection, and an F1-score of 96.77% in case-level anomaly detection. Besides, a publicity English national center for biotechnology information (NCBI) disease corpus dataset is used for externed validation with an F1-score of 95.81% in disease detection. The experimental results demonstrate the advantage of the proposed cascade networks as compared to other related methods.
登录
查看更多内容
DOI：
--
发表时间：
2016-12
期刊：
--
影响因子：
--
作者：
Nut Limsopatham;Nigel Collier
通讯作者：
Nut Limsopatham;Nigel Collier
DOI：
10.1007/978-3-540-68636-1_10
发表时间：
2008-01
期刊：
--
影响因子：
--
作者：
Yuejie Zhang;Zhiting Xu;Tao Zhang
通讯作者：
Yuejie Zhang;Zhiting Xu;Tao Zhang
DOI：
10.1136/jamia.1994.95236146
发表时间：
1994-03-01
影响因子：
6.4
作者：
FRIEDMAN, C;ALDERSON, PO;JOHNSON, SB
通讯作者：
JOHNSON, SB
影响因子：
10.6
作者：
Li, Xiaomeng;Chen, Hao;Heng, Pheng-Ann
通讯作者：
Heng, Pheng-Ann
DOI：
--
发表时间：
2004-12
期刊：
--
影响因子：
--
作者：
Sunita Sarawagi;William W. Cohen
通讯作者：
Sunita Sarawagi;William W. Cohen