About this Research Topic
Realization of electronic health records (EHR) as a digital version of patient’s paper charts has been an important milestone towards automatic information processing and data management of, and knowledge discovery from, patient health records. Recent advances in artificial intelligence, in particular natural language processing and machine learning, have enabled the secondary use of EHRs as complementary to randomized clinical trials in today’s predominant framework of evidence-based medicine, thanks to much bigger sizes, larger population coverages, and longer time periods of EHRs as compared to those of clinical trials. Although there has been much research effort in exploitation of EHRs as a valuable evidence source for patient data analysis, medical research, and clinical decision-making support, there are still improvements to do and open problems to solve.
This Research Topic focuses on mining data and discovering knowledge from EHRs for medical research. It may require development of new, or application of existing, methods in biostatistics, machine learning, and natural language processing.
We invite submissions of both novel methodological and application papers for data mining and knowledge discovery from EHRs. Attention is given to (1) hybrid approaches that combine state-of-the-art statistical, machine learning, and/or natural language processing techniques; and (2) methods that exploit both structured data and clinical notes in EHRs and external sources of clinical knowledge. Topics of interest include, but not limited to:
• Health data analytics
• Cohort identification
• Clinical prediction
• Disease phenotyping
• Symptom extraction
• Drug effectiveness and interaction
• Treatment regimen learning
• Clinical information extraction
• Clinical embeddings
We also welcome papers that deal with the limitations of EHRs to improve medical research on EHRs, e.g. missing data or data biases.
Keywords: clinical information extraction, clinical prediction, phenotyping, treatment regimen, data imputation, data bias, biostatistics, machine learning, natural language processing
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.