To deliver effective training of clinical communication, this project utilizes artificial intelligence techniques to provide clinical staffs customized and lifelike practices with timely feedback under various scenario. Thank You all frontline heroes for fighting COVID-19!
CLINIC-ISBAR is a clinical handover dataset collected in Queen Elizabeth Hospital (QEH). It consists of 100 handover reports (1895 sentences) from two real clinical emergency cases and each sentence is labeled based on the ISBAR (Identify-Situation-Background-Assessment-Recommendation) framework.
The trainee practices the communication skill by talking with a specifically designed Chatbot, which can play the role as senior doctors. The basic structure of the conversation system is a hierarchical Long Short-Term Memory (LSTM) model. The LSTM model processes the sequential data in three levels including sentence level, components level, and the conversation level.
The conversation between the trainee and the Chatbot is multi-turn and the trainee will be evaluated based on the ISBAR framework. The domain knowledge, such as the specific requirement of standard communication framework in clinical practice, will be integrated in the form of knowledge graph and benefit both the conversation generation and the performance assessment.
A friendly interface will be designed for the convenience of non-IT users. Trianers can input their tasks, good examples, bad examples, and evaluation criteria. And the system will automatic generate the conversations for the practice of the clinical staffs and evaluate the performance of the trainees. By this way, the usage of the platform can be greatly expanded to different kinds of applications.
Research team from POLYU and clical experts from QEH.
Project Coordinate Associate Professor Department of computing The Hong Kong Polytechnic University
Associate Professor Department of computing The Hong Kong Polytechnic University
Consultant, Intensive Care Unit Associate Director, Multi-displinary Simulation & Skill Centre Queen Elizabeth Hospital
Director, Multi-displinary Simulation & Skill Centre Consultant, Department of Surgery Queen Elizabeth Hospital
Consultant, Department of Anaesthesiology & OT Services Associate Director, Multi-displinary Simulation & Skill Centre Queen Elizabeth Hospital
Hospital Manager Queen Elizabeth Hospital
Assistant Officer, Multi-displinary Simulation & Skill Centre Queen Elizabeth Hospital
Research Assistant Department of computing The Hong Kong Polytechnic University