Clinical Decision Support and Natural Language Processing in Medicine: Systematic Literature Review.

Area of Science:

• Health Informatics
• Artificial Intelligence in Medicine
• Clinical Decision Support

Background:

• Accurate health information is crucial for patient diagnosis and treatment.
• Health professionals increasingly use online resources for clinical data.
• A streamlined approach to accessing health information is needed.

Purpose of the Study:

• To systematically review artificial intelligence (AI) and natural language processing (NLP) in healthcare.
• To assess their use in electronic health records and automated information searches.

Main Methods:

• Systematic search of PubMed, Embase, ScienceDirect, Scopus, and Web of Science (Jan 2000-Apr 2023).
• Inclusion of original research and systematic reviews on AI/NLP for clinical decision support.
• Quality assessment using the Critical Appraisal Skills Programme tool.

Main Results:

• 26 studies were included from 707 articles.
• NLP was used for data collection in 81% of studies.
• Electronic health records were a data source in 69% of studies.
• 19% of studies used combined NLP strategies for clinical data.
• 62% presented stand-alone data review algorithms.
• 35% utilized clinical decision support systems for immediate clinical use.

Conclusions:

• Natural language processing (NLP) engines significantly improve the accuracy of clinical decision systems.
• Biphasic tools integrating AI and human judgment enhance clinical diagnosis and treatment workflows.