Artificial intelligence methods in computer-aided diagnostic tools and decision support analytics for clinical informatics

Abstract The massive amount of health data acquired from electronic health records (EHRs), gives us unprecedented opportunities to build clinical infrastructures for clinical informatics (CI), research-based studies, and precision medicine. However, a significant part of the patients’ treatment data is stored in an unstructured format, such as free-text clinical notes. Machine learning (ML) methods and natural language processing (NLP) techniques are popular computational strategies used to extract information from these data silos. Artificial intelligence (AI) methods can be applied to deduce clinically significant inferences and assist the treatment workflow. In this chapter, we present the underpinnings, mathematical background, and relevant applications for ML methods and natural language processing techniques that are used in computer-aided diagnostics and clinical decision support analytics. We present several case studies within the healthcare domain. Finally, as a practical application, we present the design and details of an informatics-driven clinical platform powered with ML algorithms and NLP techniques for the domain of radiation oncology.