Our Name Entity Recognition tool for Emergency Medical Service Report (NEREMSR) is an online natural language processing system to identify the potential entities in the paramedics report for auditing purpose.
What exactly can this tool do?
NEREMSR can detect 17 entities mentioned in an EMS paramedics report, including common medications like Nitroglycerin (GTN) and Aspirin, procedures like ECG and stroke assessment, and clinical findings like bleeding and signs of obvious death. When the paramedics report is audited, the auditor can use this tool to tell whether the necessary procedures were missed.
NEREMSR was trained on a large unlabelled corpus from Singapore Civil Defence Force Emergency Medical Services Department with multiple natural language processing techniques. For the details of how the system is built, please refer to our paper below. You can try our system in the Demo page.
Wang Han* and Wesley Yeung*, Angeline Tung, Joey Tay Ai Meng, Feng Mengling, Shalini Arulanadam. An Emergency Medical Services Clinical Audit System driven by Weakly-Supervised Named Entity Recognition from Deep Learning. arXiv, 2020