Background: The increasing use of social media to share lived and living experiences of substance use presents a unique opportunity to obtain information on side effects, use patterns, and opinions on novel psychoactive substances. However, due to the large volume of data, obtaining useful insights through natural language processing technologies such as large language models is challenging.
Objective: This paper aims to develop a retrieval-augmented generation (RAG) architecture for medical question answering pertaining to clinicians' queries on emerging issues associated with health-related topics, using user-generated medical information on social media.
Background: The selection of the best donor for each specific patient is crucial for the success of allogeneic hematopoietic stem cell transplantation (HSCT). However, there is debate on the choice of the best donor when multiple suitable donors exist.
Methods: By using own data from two transplant centers, we have developed a calculator able to provide the patients' 2-year overall survival (OS) associated with each of the potential donor options during the selection process, in order to support the transplant physician during the choice.