SLM4Health 2025
Improving Healthcare with Small Language Models

Table of Contents

• Preface
  Summary: There were 6 papers submitted for peer-review to this workshop; 6 papers were accepted for this volume; 6 are regular papers.

Session 1: From Detection to Diagnosis: Small Language Models in Medical and Assistive AI

• A Health-Focused Risk Taxonomy for AI: Assessing Unsafe Content Detection with Small Language Models (SLMs) 1-9
  Lantana Hewitt, Abdel-Karim Al Tamimi, Robert Copeland, Richard Moore, Shaman Jhanji
• Enhancing Aphasia Speech Interpretation using Small Language Models (SLMs) 10-18
  Abdel-Karim Al-Tamimi, Kate Radford, Jacqueline Benfield, Jacob Andrews, Catherine Sweby
• Exploring the Role of Knowledge Graph-Based RAG in Japanese Medical Question Answering with Small-Scale LLMs 19-28
  Yingjian Chen, Feiyang Li, Xingyu Song, Tianxiao Li, Zixin Xu, Xiujie Chen, Issey Sukeda, Irene Li

Session 2: SLM for Clinical NLP – Power and Limitations

• Exploring the Missing Medical Context in Generated Radiology Reports 25-32
  Karan Bania, Harshvardhan Mestha, Tanmay Verlekar
• Embedding-Based Acronym Disambiguation Supported by Large Language Models in German Clinical Narratives 33-40
  Amila Kugic, Stefan Schulz, Markus Kreuzthaler
• Large language models versus static word embeddings to predict Acute Kidney Injury in the Intensive Care Unit: Does context matter? 41-50
  Rick van Slobbe, Drahomira Herrmannova, Elia S Lima-Walton, Ameen Abu-Hanna, Iacopo Vagliano