Situational AI Engineering for Automated Medical Reporting: Architecting and Experimentation with Care2Report Research Program
The emergence of generative pre-trained transformers based on large sets of natural language training data has triggered an abundance of applications ranging from chatbots, essay writing, and poem generation, to text mining. However, at the moment, there is little scientific evidence on the way these AI tools can be integrated into large software applications. AI engineering is the domain of software engineering concerned with the architecting and development of these applications.
In this keynote, we will present and discuss the Care2Report research program of Utrecht University that aims to design generic architectures of software applications for the automated reporting of human activity. Innovative interaction and reasoning are now becoming available using off-the-shelf AI technologies: machine learning, language models, speech recognition, action recognition, ontologies, knowledge graph databases, computational linguistics, and several more. We apply this general vision in the healthcare domain due to the societal need in this domain: high administrative burden where administrative duties are reported to take 20 to even 40% of the working time.
We will highlight the system architecture and the experimentation of the research program. We show how networks of architectural pipelines are being deployed to configure the AI technologies into one overall application for the reporting of medical consultations. Prompt engineering plays a significant role in the semantic interpretation of the natural language interaction for the summarization task. The application in the healthcare domain requires proper recognition of anatomic elements, symptoms, observations, diagnosis, and treatment policies. This recognition is configured based on a so-called medical guideline ontology derived from the publicly available guidelines of healthcare professionals. We discuss how prompt scripts can be composed by means of situational reuse of consultation reporting in different medical domains.
We end with an outlook on the future of this exciting, yet challenging, research endeavor.
Prof.dr. Sjaak Brinkkemper is full professor of Software Production at the Department of Information and Computing Sciences of Utrecht University, the Netherlands. He leads a group of about thirty researchers specialized in product software development and software entrepreneurship. The main research themes of the group are methodology of software production, software ecosystems, and responsibility and sustainability in the software industry. In 2019 he established the Care2Report research program (www.care2report.nl) that aims at providing complete medical reports from patient-doctor conversations using speech and action recognition. Brinkkemper has published several books and papers in his research interests: medical informatics, healthcare software production, requirements engineering, software architecture, and method engineering.