JMIR Medical Informatics is pleased to announce the electronic collection (e-collection) Theme Issue 2024: Health Natural Language Processing and Applications with Large Language Models. This is a “closed group theme issue”—the e-collection is not open to everybody to submit. Only authors of selected papers are invited to submit to this e-collection.
The rapid development of natural language processing (NLP) has ushered in a new era of innovation, particularly with the advent of large language models (LLMs). Generative AI and LLMs are rapidly transforming various domains, including health care. LLMs like Google Med-PaLM have shown promising generative capabilities in responding to various health care questions. They can process and analyze vast amounts of medical text, engage in patient Q&A, generate various medical reports, and improve the understanding of unstructured electronic health records. This technology will be rapidly developed and applied in the health care field in the foreseeable future.
However, LLMs also bring new challenges and problems in medical informatics. For instance, a phenomenon known as the "hallucination and confabulation issue" often occurs in LLMs and is particularly dangerous in the health care field, as it can lead to the dissemination of inaccurate or misleading information, potentially causing serious health risks and adverse outcomes. Additionally, general-domain LLMs may perform less effectively in medical fields due to domain-specific characteristics. Therefore, combining domain knowledge to enhance the performance of medical LLMs is a crucial area of focus. Researchers and practitioners must address these challenges to fully harness the potential of this transformative technology.
To that end, the China Conference on Health Information Processing (CHIP) proposes the e-collection, Theme Issue 2024: Health Natural Language Processing and Applications with Large Language Models to promote interdisciplinary research. CHIP is the annual symposium of the Chinese Information Processing Society’s Technical Committee of Medical, Health and Biological Information Processing. CHIP is a leading international conference specialized in the field of health information processing in China. The conference provides a platform for researchers and practitioners from academia, industry, and government worldwide to share their ideas, research findings, and experiences to advance the field. Previous CHIP conferences were successfully held annually from 2015 to 2023, with experts from the United States, Germany, France, United Kingdom, Japan, Australia, China, and other countries presenting their latest research and achievements.
CHIP will select the best-scored papers and recommend them for the e-collection. This e-collection will contain articles on the technologies, methodologies, and applications of NLP in the health/medical sector and will also include viewpoints and reviews. The e-collection will showcase various new developments in these areas. Authors of selected papers are encouraged to submit high-quality original research articles, mainly describing original research and presenting results that advance state-of-the-art.
Invited papers submitted to this e-collection for consideration may explore, but are not limited to, the following topics (sample titles):
- Health information extraction and retrieval using LLMs
- Question-answering technologies for health application
- Benchmarking datasets and evaluation metrics of LLMs in medicine and health
- LLMs for medical report generation and interpretability
- Clinical decision support and informatics with LLMs
- Techniques for the enrichment of medical knowledge graphs incorporated
- within LLMs
- Medical and health-specific LLM methods
- Multimodal LLMs in health care
- Health care knowledge representation and reasoning
- Medical ontologies
- LLM techniques for the personalization of medicine
- Safety and bias in medical LLMs
- Applications of LLMs in medical and health care
- Trends and challenges in health and medical LLMs
How to Submit
Only invited authors identified by the guest editors, on behalf of the conference organizing committee, may submit their manuscripts to this JMIR Medical Informatics e-collection. Authors of selected papers are asked to submit their work by selecting “Theme Issue 2024: Health Natural Language Processing and Applications with Large Language Models” in the “Section” drop-down list. See the article How do I submit to a theme issue? in our Knowledge Base for more information. Invited papers should adhere to JMIR Medical Informatics’ formatting guidelines for relevant articles types outlined in our Instructions for Authors. Manuscripts submitted to this e-collection that have not been invited or have been deemed unsuitable may be transferred to another JMIR Publications journal if the authors agree in accordance with JMIR Publications’ portable peer-review (also called cascading peer-review) process. A manuscript submitted to this e-collection that is transferred to another JMIR journal and accepted would still appear as part of the Theme Issue 2024: Health Natural Language Processing and Applications with Large Language Models e-collection.
Submission Guidelines
All submissions will undergo a rigorous peer-review process, and accepted articles will be published as part of the e-collection Theme Issue 2024: Health Natural Language Processing and Applications with Large Language Models.
A 20% discount on the article processing fee (APF) is available for articles that are published as a part of this e-collection.
Submission Deadline: March 31, 2025
Submissions not reviewed or accepted for publication in this JMIR Medical Informatics e-collection may be offered a cascading peer review or a transfer to other JMIR Publications journals, according to standard publisher policies. For example, early-stage formative work that informs the design of future interventions or research may better fit the scope for JMIR Formative Research. Authors are encouraged to submit study protocols or grant proposals to JMIR Research Protocols before data acquisition to preregister the study (Registered Reports—subsequent acceptance in one of the JMIR Publications journals is then guaranteed).
All articles submitted to this e-collection will be shared and published rapidly through the following mechanisms:
- All peer-reviewed articles in this e-collection will be immediately and permanently made open access. This is the standard for all titles within the JMIR Publications portfolio.
- Articles can be made immediately available in JMIR Preprints (with a DOI) after submission if authors select the preprint option to enable this service.
Guest Editors
Qiao Jin, MD,
National Library of Medicine, National Institutes of Health, USA
Tianyong Hao, PhD
School of Computer Science, South China Normal University, China
Ling Luo,
School of Computer Science and Technology, Dalian University of Technology, China
Contact information for questions and presubmission enquiries:
Ling Luo
lingluo@dlut.edu.cn