Deep learning-based NLP Data Pipeline for EHR Scanned Document Information Extraction

TL;DR

This study evaluates a deep learning NLP pipeline for extracting sleep apnea indicators from scanned EHR documents, emphasizing the importance of image preprocessing and document layout in improving extraction accuracy.

Contribution

It systematically assesses the impact of image preprocessing, NLP models, and document layout features on extracting clinical indicators from scanned health records.

Findings

Clinical BERT achieved AUROC of 0.9743 for AHI

Proper image preprocessing improves extraction accuracy

Document layout information enhances model performance

Abstract

Scanned documents in electronic health records (EHR) have been a challenge for decades, and are expected to stay in the foreseeable future. Current approaches for processing often include image preprocessing, optical character recognition (OCR), and text mining. However, there is limited work that evaluates the choice of image preprocessing methods, the selection of NLP models, and the role of document layout. The impact of each element remains unknown. We evaluated this method on a use case of two key indicators for sleep apnea, Apnea hypopnea index (AHI) and oxygen saturation (SaO2) values, from scanned sleep study reports. Our data that included 955 manually annotated reports was secondarily utilized from a previous study in the University of Texas Medical Branch. We performed image preprocessing: gray-scaling followed by 1 iteration of dilating and erode, and 20% contrast increasing. The OCR was implemented with the Tesseract OCR engine. A total of seven Bag-of-Words models (Logistic Regression, Ridge Regression, Lasso Regression, Support Vector Machine, k-Nearest Neighbor, Na\"ive Bayes, and Random Forest) and three deep learning-based models (BiLSTM, BERT, and Clinical BERT) were evaluated. We also evaluated the combinations of image preprocessing methods (gray-scaling, dilate & erode, increased contrast by 20%, increased contrast by 60%), and two deep learning architectures (with and without structured input that provides document layout information). Our proposed method using Clinical BERT reached an AUROC of 0.9743 and document accuracy of 94.76% for AHI, and an AUROC of 0.9523, and document accuracy of 91.61% for SaO2. We demonstrated the proper use of image preprocessing and document layout could be beneficial to scanned document processing.