Natural Language Processing in Electronic Health Records in relation to healthcare decision-making: A systematic review

Here we present deep-learning techniques for healthcare, centering our discussion on deep learning in computer vision, natural language processing, reinforcement learning, and generalized methods. We describe how these computational techniques can impact a few key areas of medicine and explore how to build end-to-end systems. Our discussion of computer vision focuses largely on medical imaging, and we describe the application of natural language processing to domains such as electronic health record data. Similarly, reinforcement learning is discussed in the context of robotic-assisted surgery, and generalized deep-learning methods for genomics are reviewed.

Systematic reviews aide the analysis and dissemination of evidence, using rigorous and transparent methods to generate empirically attained answers to focused research questions. Identifying all evidence relevant to the research questions is an essential component, and challenge, of systematic reviews. Gray literature, or evidence not published in commercial publications, can make important contributions to a systematic review. Gray literature can include academic papers, including theses and dissertations, research and committee reports, government reports, conference papers, and ongoing research, among others. It may provide data not found within commercially published literature, providing an important forum for disseminating studies with null or negative results that might not otherwise be disseminated. Gray literature may thusly reduce publication bias, increase reviews' comprehensiveness and timeliness, and foster a balanced picture of available evidence. Gray literature's diverse formats and audiences can present a significant challenge in a systematic search for evidence. However, the benefits of including gray literature may far outweigh the cost in time and resource needed to search for it, and it is important for it to be included in a systematic review or review of evidence. A carefully thought out gray literature search strategy may be an invaluable component of a systematic review. This narrative review provides guidance about the benefits of including gray literature in a systematic review, and sources for searching through gray literature. An illustrative example of a search for evidence within gray literature sources is presented to highlight the potential contributions of such a search to a systematic review. Benefits and challenges of gray literature search methods are discussed, and recommendations made.