From Data to Evidence: Transformative AI for Real-World Data

Dear Colleagues, 

Real-world data (RWD), particularly electronic health records (EHRs), are increasingly being used to convert into real-world evidence (RWE) that guides clinical practice, care planning and decision-making.  Advanced artificial intelligence (AI) techniques, such as natural language processing (NLP, especially the recent surge of large language models [LLMs]), unlock value from unstructured EHR text at scale, enabling concept extraction, improved phenotyping and clinical summarization. Causal AI provides credible effect estimation and decision support from observational RWD by addressing confounding, transportability and counterfactual reasoning. Multimodal AI (e.g., health digital twins that fuse EHRs with biomedical imaging, physiological signals and knowledge graphs), supports individualized simulation, prognosis and treatment planning.  Together, these capabilities can accelerate clinical research, improve diagnostic tool development and advance the generation of robust RWE. However, important gaps remain for real-world deployment, including LLM hallucinations, privacy protection for EHRs and bias and fairness in such AI models. 

This Special Issue of the Journal of Informatics aims to improve understanding of how advanced AI tools can leverage EHRs to improve clinical research in real-world settings. We welcome methodology papers, applications, reviews and reproducible resources (datasets, benchmarks and code). High-quality submissions accepted for publication may be considered for discounts at the Editorial Office’s discretion.

Pillars and Topics of Interest

(1) RWD and Causal AI

• Causal inference with RWD for clinical effectiveness and safety research
• Identification and mitigation of confounding, selection bias and transportability issues
• Causal structure learning from heterogeneous clinical data

(2) RWD and Multimodal AI

• Multimodal fusion of EHRs with biomedical imaging, physiological signals and knowledge graphs
• Health digital twins for prognosis, simulation and treatment planning
• Robustness, calibration and generalizability in multimodal models

(3) LLMs/NLP on EHRs

• Novel models and applications in predictive analytics, clinical NLP and LLMs for unstructured EHRs
• NLP for de-identification, synthetic text generation and privacy-preserving workflows
• Bias, fairness, robustness and hallucination mitigation for LLMs 

Prof. Dr. Jiang Bian
Dr. Yu Huang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Informatics is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• electronic health record (EHR)
• real-world evidence (RWE)
• large language models (LLMs)
• digital health
• multimodal fusion
• causal AI