Special Issue of Artificial Intelligence in Medicine on Large language Models for Medicine

1. AIMS AND SCOPE

Large language models (LLMs), trained on extensive datasets encompassing scientific content from numerous disciplines, have demonstrated immense versatility in tasks such as content generation, summarization, translation, prediction, and recognition. With the release of ChatGPT in 2022, there has been a soaring interest in LLMs, and they have already been shown to be surprisingly useful AI tools for a variety of domains and tasks.

LLMs have a great potential to make a positive impact in the medical domain. They can be used in a clinical context both as interactive decision support tools for medical experts, assist in triage processes, enhance symptom collection, and streamline administrative functions like the consolidation of patient notes and discharge documents. In the education of doctors, LLMs can be utilized in preparing learning materials, supporting interactive learning and aid in the assessment of medical knowledge. Moreover, for medical researchers, LLMs offer the advantage of easy access to relevant literature, efficient knowledge summaries, and support during manuscript drafting.

At the same time, the medical domain poses more difficult challenges to LLMs than many other domains. Medicine is a mission critical domain in which inaccuracies can have disastrous consequences. Hence, the demands for precision, comprehensiveness, up-to-date information, and clarity are considerably heightened in this field. Added layers of complexity arise from the stringent privacy norms and the extensive legal frameworks that heavily regulate the medical sector

To embrace the challenges and opportunities in designing, validating and deploying LLMs in a medical context, this special issue aims to encourage submissions of scientific findings from both academia and healthcare industry that present fundamental theory, techniques, practical experiences of LLMs in medicine as well as roadmaps for directions of research and development.

This special issue seeks scientific papers that contain novelty from an AI perspective. Papers that simply apply existing algorithms and systems to a medical domain are outside the scope of the issue and will be bench rejected.

2. TOPICS COVERED

The topics of this special issue include, but are not limited to:

• Novel text resources for training clinical LLMs
• Translating doctor-to-patient language
• Training and refining LLMs for medical education
• Training and refining LLMs for medical research
• Local language models
• Combining LLMs and medical knowledge graphs
• Accuracy and recency of LLMs for medical applications
• New evaluation paradigms for medical LLMs
• Explanations of LLMs for medical applications
• Addressing multi-linguality requirements in clinical LLMs
• Real-time Web access for medical LLMs
• Addressing bias in medical LLMs
• Ethical and legal aspects of medical LLMs

We encourage authors to publish training data and models on platforms such as at Hugging Face or Github.

3. IMPORTANT DATES

Papers can be submitted anytime prior to the submission deadline and will enter the reviewing process on a rolling basis. The submission deadline indicates the latest day a paper can be submitted to the special issue, and the Final Manuscript Due the latest date the final manuscript can be submitted.

Submission Deadline – 02/04/2024
Notification of First Decision – 3 months after submission
Final Manuscript Due – 01/12/2024

4. SUBMISSION

The submission website is: https://www2.cloud.editorialmanager.com/aiim/

To ensure that all manuscripts are correctly identified for inclusion into this special issue, it is important that authors select VSI: LLMs for Medicine when they reach the “Article Type” step in the submission process.

5. GUEST EDITORS

Grigoris Antoniou is Research Professor of Artificial Intelligence at the University of Huddersfield, UK, and Leibniz University Hannover, Germany. His research activity spans over knowledge representation, semantic web technologies, hybrid AI learning and AI for medicine. He has published over 300 scientific papers that have attracted over 14,000 citations. Among others, he was guest editor of a special issue of Artificial Intelligence in Medicine on Medical Analytics for Healthcare Intelligence in 2021. He is Fellow of IEEE, European AI Society and Asia-Pacific AI Association.
Email: g.antoniou@hud.ac.uk

Keno Bressem is a board certified radiologist at Charité Universitätsmedizin Berlin, Germany and Digital Clinician Scientist at the Berlin Institute of Health, Germany. His research activity focusses on applied AI in medicine including computer vision and natural language processing. He is coordinator of the EU-funded project COMFORT (https://comfort-ai.eu/) a multinational initiative that focusses on the development of AI solutions of urologic cancers. He is member of the Radiologic societies of North America, Europe and Germany.
Email: keno-kyrill.bressem@charite.de

Frank van Harmelen is professor of Knowledge Representation and Reasoning at the Vrije Universiteit Amsterdam. As one of the early researchers in semantic web technologies (currently known as linked data and knowledge graphs), he co-defined the Web ontology language OWL, which has become a worldwide standard and is in wide academic and commercial use. He is principal investigator of the Hybrid Intelligence Centre (https://hybrid-intelligence-centre.org), a 20m€, 10 year research project into AI systems that collaborate with people instead of replacing them. He is a fellow of the European AI Society and of the Asia-Pacific AI Association. He was elected a member of the Academia Europae, and of the Royal Netherlands Society of Sciences and Humanities (KNAW). Among other journals, he serves on the editorial board of the AI in Medicine journal.
Email: frank.van.harmelen@cs.vu.nl

Alexander Löser is Professor for Data Science and Text-Mining at the Berliner Hochschule für Technik and Director of the Data Science Research Center. His research interests lie at the intersection of natural language processing and machine learning, in particular for clinical applications. He is expert for LLMs in the ‘Plattform Lernende Systeme,‘ a platform of leading AI experts sponsored by BMBF, the German Federal Ministry of Education and Research. Alexander has a well-established track record of innovation and technology transfer. He worked as an independent consultant with eBay, IBM, Zalando, MunichRe, SpringerNature, Fresenius, Krohne, Babbel, among others. Over the time he helped these organizations to create six data platforms with more than 45 data products.
Email: aloeser@bht-berlin.de

Wolfgang Nejdl is Professor of Computer Science at Leibniz University. He heads the L3S Research Center, www.L3S.de, as well as the Data Science Institute / Knowledge Based Systems, and does research in the areas of Information Retrieval, Artificial Intelligence, Social and Semantic Web, Digital Libraries and Technology Enhanced Learning. He was PI of the ERC Advanced Grant ALEXANDRIA, from 2014 - 2019, working on foundations for temporal retrieval, exploration and analytics in Web archives. Current projects include NoBIAS, SoBigData++, and the International Leibniz Future Lab on Artificial Intelligence, with a special focus on personalized medicine. Wolfgang Nejdl (https://www.kbs.uni-hannover.de/~nejdl), has published more than 430 scientific articles, with an h-index (based on Google Scholar) of 78.
Email: nejdl@L3S.de