TY - JOUR
T1 - Visibility Attribute Extraction and Anomaly Detection for Chinese Diagnostic Report Based on Cascade Networks
AU - Zhang, Jitong
AU - Jiang, Huiyan
AU - Huang, Liangliang
AU - Yao, Yu Dong
AU - Li, Siqi
N1 - Publisher Copyright:
© 2013 IEEE.
PY - 2019
Y1 - 2019
N2 - 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.
AB - 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.
KW - CNN
KW - CRF
KW - GRU
KW - Image diagnosis report
KW - PET/CT image
KW - anomaly detection
KW - visibility attribute
UR - http://www.scopus.com/inward/record.url?scp=85097342695&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85097342695&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2019.2932842
DO - 10.1109/ACCESS.2019.2932842
M3 - Article
AN - SCOPUS:85097342695
VL - 7
SP - 116402
EP - 116412
JO - IEEE Access
JF - IEEE Access
M1 - 8786226
ER -