Format

Send to

Choose Destination
Artif Intell Med. 2019 Jun;97:79-88. doi: 10.1016/j.artmed.2018.11.004. Epub 2018 Nov 23.

Comparative effectiveness of convolutional neural network (CNN) and recurrent neural network (RNN) architectures for radiology text report classification.

Author information

1
Department of Biomedical Data Science, Stanford University School of Medicine, Stanford, CA, USA. Electronic address: imonb@stanford.edu.
2
Artificial Intelligence Laboratory, Philips Research North America, Cambridge, MA, USA.
3
Department of Radiology, Stanford University School of Medicine, Stanford, CA, USA.
4
Department of Bioinformatics, University of Utah Medical Center, UT, USA.
5
Department of Neuroradiology, Duke University School of Medicine, NC, USA.
6
Department of Radiology, Children Hospital Colorado, CO, USA.
7
Department of Biomedical Data Science, Stanford University School of Medicine, Stanford, CA, USA; Department of Radiology, Stanford University School of Medicine, Stanford, CA, USA.

Abstract

This paper explores cutting-edge deep learning methods for information extraction from medical imaging free text reports at a multi-institutional scale and compares them to the state-of-the-art domain-specific rule-based system - PEFinder and traditional machine learning methods - SVM and Adaboost. We proposed two distinct deep learning models - (i) CNN Word - Glove, and (ii) Domain phrase attention-based hierarchical recurrent neural network (DPA-HNN), for synthesizing information on pulmonary emboli (PE) from over 7370 clinical thoracic computed tomography (CT) free-text radiology reports collected from four major healthcare centers. Our proposed DPA-HNN model encodes domain-dependent phrases into an attention mechanism and represents a radiology report through a hierarchical RNN structure composed of word-level, sentence-level and document-level representations. Experimental results suggest that the performance of the deep learning models that are trained on a single institutional dataset, are better than rule-based PEFinder on our multi-institutional test sets. The best F1 score for the presence of PE in an adult patient population was 0.99 (DPA-HNN) and for a pediatrics population was 0.99 (HNN) which shows that the deep learning models being trained on adult data, demonstrated generalizability to pediatrics population with comparable accuracy. Our work suggests feasibility of broader usage of neural network models in automated classification of multi-institutional imaging text reports for a variety of applications including evaluation of imaging utilization, imaging yield, clinical decision support tools, and as part of automated classification of large corpus for medical imaging deep learning work.

KEYWORDS:

Convolutional neural network (CNN); Pulmonary embolism; Radiology report analysis; Recurrent neural network (RNN); Text report classification

PMID:
30477892
PMCID:
PMC6533167
[Available on 2020-06-01]
DOI:
10.1016/j.artmed.2018.11.004

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center