Searching the PDF Haystack: Automated Knowledge Discovery in Scanned EHR Documents

Abstract

Background Clinicians express concern that they may be unaware of important information contained in voluminous scanned and other outside documents contained in electronic health records (EHRs). An example is “unrecognized EHR risk factor information,” defined as risk factors for heritable cancer that exist within a patient's EHR but are not known by current treating providers. In a related study using manual EHR chart review, we found that half of the women whose EHR contained risk factor information meet criteria for further genetic risk evaluation for heritable forms of breast and ovarian cancer. They were not referred for genetic counseling. Objectives The purpose of this study was to compare the use of automated methods (optical character recognition with natural language processing) versus human review in their ability to identify risk factors for heritable breast and ovarian cancer within EHR scanned documents. Methods We evaluated the accuracy of the chart review by comparing our criterion standard (physician chart review) versus an automated method involving Amazon's Textract service (Amazon.com, Seattle, Washington, United States), a clinical language annotation modeling and processing toolkit (CLAMP) (Center for Computational Biomedicine at The University of Texas Health Science, Houston, Texas, United States), and a custom-written Java application. Results We found that automated methods identified most cancer risk factor information that would otherwise require clinician manual review and therefore is at risk of being missed. Conclusion The use of automated methods for identification of heritable risk factors within EHRs may provide an accurate yet rapid review of patients' past medical histories. These methods could be further strengthened via improved analysis of handwritten notes, tables, and colloquial phrases.

PCaGuard: A Software Platform to Support Optimal Management of Prostate Cancer

“…38 Since electronic health records (EHRs) contain more and more data, researchers develop methods to extract characteristics either from data 39 40 or scanned and other outside documents contained in EHRs. 41 In the market, data management systems available clinical, such as REDCap, OpenClinica, and eClinicalOS, but these systems are either commercial or not suitable for this project because the whole workflow is time consuming, error prone, and requires the integration of different components.…”

Improving Cancer Care Communication: Identifying Sociodemographic Differences in Patient Portal Secure Messages Not Authored by the Patient

“…The use of natural language processing (NLP) models to extract data from unstructured free text in the electronic health record (EHR) has demonstrated efficiency and potential to reduce missed clinically significant information compared with manual review. 10,11 The complete methodology is described in a manuscript detailing the development of our NLP model to classify the authorship of PSMs. 12 In brief, we trained an NLP model on a dataset of 1,850 individual messages with manual annotation of patient or presumed proxy as our gold standard in concordance with other NLP models categorizing PSMs.…”

A Smartphone App to Increase Immunizations in the Pediatric Solid Organ Transplant Population: Development and Initial Usability Study

“…The participants suggested that in the future it would be ideal if one could take a photograph of their child's prior vaccine record with their phone's camera, upload the photograph into the Immunize PediatricTransplant app, and then have the app automatically recognize vaccine names and dates. Optimal character recognition with natural language processing has recently been shown to have the potential to accurately identify clinically relevant information contained within the EMR [44][45][46][47].…”