Semi-Supervised Information Extraction for Cancer Pathology Reports

Qiu, John X.; Gao, Shang; Alawad, Mohammed; Schaefferkoetter, Noah; Alamudun, Folami; Yoon, Hong-Jun; Wu, Xiao-Cheng; Tourassi, Georgia

doi:10.1109/BHI.2019.8834470

Title: Semi-Supervised Information Extraction for Cancer Pathology Reports

Abstract

Pathology reports are a main source of data for cancer surveillance programs. Manual coding of pathology reports is labor-intensive but necessary for obtaining labeled data to train automated information extraction systems. In this study, we investigated semi-supervised deep learning, improving the performance of a multitask information extraction system for automated annotation of pathology reports. We used a set of over 374,000 pathology reports from the Louisiana Tumor Registry and a novel convolutional attention-based auto-encoder. We performed a set of experiments comparing supervised training augmented with unlabeled data at 1%, 5%, 10%, and 50% of the original data size. We also compared the impact of extending text processing to include unlabeled tokens. We find that semi-supervised training consistently improved individual performance with increased micro-averaged F-scores between 0.012 and 0.064 and increased macro-averaged F-scores of up to 0.158. This demonstrates that semantic information learned via unsupervised learning can be used to improve supervised clinical task performance.

Authors:

Qiu, John X. ^[1]; Gao, Shang ^[1];

^[1]; Schaefferkoetter, Noah ^[1];

^[1];

^[1]; Wu, Xiao-Cheng ^[2];

^[1]

ORNL
LSUHSC-Louisiana Tumor Registry

Publication Date:: 2019-05-01

Research Org.:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Org.:: USDOE

OSTI Identifier:: 1564225

DOE Contract Number:: AC05-00OR22725

Resource Type:: Conference

Resource Relation:: Conference: IEEE EMBS International Conference on Biomedical & Health Informatics (IEEE-EMBS BHI 2019) - Chicago, Illinois, United States of America - 5/19/2019 8:00:00 AM-5/22/2019 8:00:00 AM

Country of Publication:: United States

Language:: English

Citation Formats


                    Qiu, John X., Gao, Shang, Alawad, Mohammed, Schaefferkoetter, Noah, Alamudun, Folami, Yoon, Hong-Jun, Wu, Xiao-Cheng, and Tourassi, Georgia. Semi-Supervised Information Extraction for Cancer Pathology Reports.  United States: N. p., 2019. 
        Web.  doi:10.1109/BHI.2019.8834470.

Copy to clipboard


                    Qiu, John X., Gao, Shang, Alawad, Mohammed, Schaefferkoetter, Noah, Alamudun, Folami, Yoon, Hong-Jun, Wu, Xiao-Cheng, & Tourassi, Georgia. Semi-Supervised Information Extraction for Cancer Pathology Reports.  United States.  https://doi.org/10.1109/BHI.2019.8834470

Copy to clipboard


                    Qiu, John X., Gao, Shang, Alawad, Mohammed, Schaefferkoetter, Noah, Alamudun, Folami, Yoon, Hong-Jun, Wu, Xiao-Cheng, and Tourassi, Georgia. 2019.  
        "Semi-Supervised Information Extraction for Cancer Pathology Reports".  United States.  https://doi.org/10.1109/BHI.2019.8834470.  https://www.osti.gov/servlets/purl/1564225.

Copy to clipboard


                    
@article{osti_1564225,

  title        = {Semi-Supervised Information Extraction for Cancer Pathology Reports},

  author       = {Qiu, John X. and Gao, Shang and Alawad, Mohammed and Schaefferkoetter, Noah and Alamudun, Folami and Yoon, Hong-Jun and Wu, Xiao-Cheng and Tourassi, Georgia},

  abstractNote = {Pathology reports are a main source of data for cancer surveillance programs. Manual coding of pathology reports is labor-intensive but necessary for obtaining labeled data to train automated information extraction systems. In this study, we investigated semi-supervised deep learning, improving the performance of a multitask information extraction system for automated annotation of pathology reports. We used a set of over 374,000 pathology reports from the Louisiana Tumor Registry and a novel convolutional attention-based auto-encoder. We performed a set of experiments comparing supervised training augmented with unlabeled data at 1%, 5%, 10%, and 50% of the original data size. We also compared the impact of extending text processing to include unlabeled tokens. We find that semi-supervised training consistently improved individual performance with increased micro-averaged F-scores between 0.012 and 0.064 and increased macro-averaged F-scores of up to 0.158. This demonstrates that semantic information learned via unsupervised learning can be used to improve supervised clinical task performance.},

  doi          = {10.1109/BHI.2019.8834470},

  url          = {https://www.osti.gov/biblio/1564225},
  journal      = {},
number       = ,

  volume       = ,

  place        = {United States},

  year         = {2019},

  month        = {5}

}

Copy to clipboard

Conference:

View Conference (0.94 MB)

https://doi.org/10.1109/BHI.2019.8834470

Other availability

Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:

Export Metadata

Save to My Library

Works referenced in this record:

Hierarchical attention networks for information extraction from cancer pathology reports
journal, November 2017

Gao, Shang; Young, Michael T.; Qiu, John X.
Journal of the American Medical Informatics Association, Vol. 25, Issue 3
https://doi.org/10.1093/jamia/ocx131

Scalable deep text comprehension for Cancer surveillance on high-performance computing
journal, December 2018

Qiu, John X.; Yoon, Hong-Jun; Srivastava, Kshitij
BMC Bioinformatics, Vol. 19, Issue S18
https://doi.org/10.1186/s12859-018-2511-9

Explainable Prediction of Medical Codes from Clinical Text
conference, January 2018

Mullenbach, James; Wiegreffe, Sarah; Duke, Jon
Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)
https://doi.org/10.18653/v1/N18-1100

Deep learning
journal, May 2015

LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
Nature, Vol. 521, Issue 7553
https://doi.org/10.1038/nature14539

Natural Language Processing in Medicine: An Overview
journal, September 1996

Spyns, P.
Methods of Information in Medicine, Vol. 35, Issue 04/05
https://doi.org/10.1055/s-0038-1634681

Similar records in OSTI.GOV collections:

Semi-supervised learning approach for crack detection and identification in advanced gas-cooled reactor graphite bricks - 111

Conference Berry, Craig; West, Graeme; McArthur, Stephen; ...

One of the life-limiting components of an Advanced Gas cooled Reactor (AGR) is its graphite core. The bricks present in the core undergo radiolytic oxidation throughout their lifetime which causes graphite weight loss and irradiation which can result in some of the bricks developing cracks. Understanding the nature and extent of brick cracking within the core is key to ensuring continued and extended operation of the AGR fleet. A semi-supervised machine learning classification algorithm is proposed as a method for improving the detection of cracked graphite bricks, by combining the labels derived from infrequent, detailed inspections of the core, withmore »« less
Extraction of Tumor Site from Cancer Pathology Reports using Deep Filters

Conference Dubey, Abhishek; Hinkle, Jacob; Christian, Blair; ...

Purpose: Pathology reports are the primary source of information concerning the millions of cancer cases across the United States. % Cancer registries manually process the pathology reports to extract the pertinent information including primary tumor site, behavior, histology, laterality, and grade. % Processing a large volume of the pathology reports in a timely manner is a continuing challenge for cancer registries. % The purpose of this study is to develop an information extraction pipeline to reliably and efficiently extract reportable information.\\ Method: % We have developed a novel inverse-regression (IR) based information extraction pipeline. % The inverse-regression based supervised filtermore »« less
https://doi.org/10.1145/3307339.3342173

Full Text Available
Extraction of Tumor Site from Cancer Pathology Reports using Deep Filters

Conference Dubey, Abhishek; Hinkle, Jacob; Christian, Blair; ...

Purpose: Pathology reports are the primary source of information concerning the millions of cancer cases across the United States. Cancer registries manually process the pathology reports to extract the pertinent information including primary tumor site, behavior, histology, laterality, and grade. Processing a large volume of the pathology reports in a timely manner is a continuing challenge for cancer registries. The purpose of this study is to develop an information extraction pipeline to reliably and efficiently extract reportable information. Method: We have developed a novel inverse-regression (IR) based information extraction pipeline. The inverse-regression based supervised filter has been successfully applied tomore »« less
https://doi.org/10.1145/3307339.3342173

Full Text Available
Automatic extraction of cancer registry reportable information from free-text pathology reports using multitask convolutional neural networks

Journal Article Alawad, Mohammed; Gao, Shang; Qiu, John; ... - Journal of the American Medical Informatics Association

Objective: In this work we implement 2 different multitask learning (MTL) techniques, hard parameter sharing and cross-stitch, to train a word-level convolutional neural network (CNN) specifically designed for automatic extraction of cancer data from unstructured text in pathology reports. We show the importance of learning related information extraction (IE) tasks leveraging shared representations across the tasks to achieve state-of-the-art performance in classification accuracy and computational efficiency. Materials and Methods: Multitask CNN (MTCNN) attempts to tackle document information extraction by learning to extract multiple key cancer characteristics simultaneously. We trained our MTCNN to perform 5 information extraction tasks: (1) primary cancermore »« less
Cited by 22
https://doi.org/10.1093/jamia/ocz153
Coarse-to-Fine Multi-Task Training of Convolutional Neural Networks for Automated Information Extraction from Cancer Pathology Reports

Conference Alawad, Mohammed; Yoon, Hong-Jun; Tourassi, Georgia

Information extraction and coding of free-text pathology reports is an important activity for cancer registries to support national cancer surveillance. Cancer registrars must process high volumes of pathology reports on an annual basis. In this study, we investigated an automated approach using a coarse-to-fine training of convolutional neural networks (CNNs) for extracting the primary site, histological grade and laterality from unstructured cancer pathology text reports. Our proposed training scheme consists of two stages. In the first stage, the multi-task learning (MTL) with hard parameter sharing approach is used to train a multi-task MT-CNN model for all the tasks. Then, themore »« less
https://doi.org/10.1109/BHI.2018.8333408

Full Text Available

Similar Records

Title: Semi-Supervised Information Extraction for Cancer Pathology Reports

Abstract

Citation Formats

Hierarchical attention networks for information extraction from cancer pathology reports journal, November 2017

Scalable deep text comprehension for Cancer surveillance on high-performance computing journal, December 2018

Explainable Prediction of Medical Codes from Clinical Text conference, January 2018

Deep learning journal, May 2015

Natural Language Processing in Medicine: An Overview journal, September 1996

Hierarchical attention networks for information extraction from cancer pathology reports
journal, November 2017

Scalable deep text comprehension for Cancer surveillance on high-performance computing
journal, December 2018

Explainable Prediction of Medical Codes from Clinical Text
conference, January 2018

Deep learning
journal, May 2015

Natural Language Processing in Medicine: An Overview
journal, September 1996