Digital Scribe

Currently, general practitioners (GPs) spend more than 50% of their time documenting clinical encounters rather than patient care. In this project we aim to design an Artificial Intelligence (AI) assistant that would listen to doctor-patient conversations during consultations and automatically generate summary notes. We performed three user research and evaluation methods: a contextual investigation of doctor-patient technology relationships at consultation rooms, a series of co-design workshops, and a “Wizard of Oz” (WoZ) study to test different collaboration models between GPs and AI.

Study 1: A contextual investigation of doctor-patient technology: We observed over 30 primary care consultations, and the doctor-patient conversations were transcribed and analysed. The outcomes suggest that consultation activities are fragmented and flow in a non-linear way.

Study 2: We conducted three co-design workshops with 16 GPs. The outcomes included design considerations for future AI-enabled consultations.

Study 3: We performed Five WoZ experiments to evaluate 5 different types of doctor-AI interaction models. The outcomes included the suitability of different interaction models for different consultation activities. 

  • Dr A. baki Kocaballi
  • Dr Kiran Ijaz
  • Liliana Laranjo
  • Juan C Quiroz
  • Dana Rezazadegan
  • huong Ly Tong
  • Simon Willcock
  • Shlomo Berkovsky
  • Enrico Coiera
  • Rezazadegan, D., Berkovsky, S., Quiroz, J. C., Kocaballi, A. B., Wang, Y., Laranjo, L., & Coiera, E. (2021). Symbolic and Statistical Learning Approaches to Speech Summarization: A Scoping Review. Computer Speech & Language, 101305.
  • Kocaballi, A. B., Ijaz, K., Laranjo, L., Quiroz, J. C., Rezazadegan, D., Tong, H. L., … & Coiera, E. (2020). Envisioning an artificial intelligence documentation assistant for future primary care consultations: A co-design study with general practitioners. Journal of the American Medical Informatics Association, 27(11), 1695-1704.
  • Quiroz, J. C., Laranjo, L., Kocaballi, A. B., Briatore, A., Berkovsky, S., Rezazadegan, D., & Coiera, E. (2020). Identifying relevant information in medical conversations to summarize a clinician-patient encounter. Health Informatics Journal, 26(4), 2906-2914.
  • Kocaballi, A. B., Coiera, E., Tong, H. L., White, S. J., Quiroz, J. C., Rezazadegan, F., … & Laranjo, L. (2019). A network model of activities in primary care consultations. Journal of the American Medical Informatics Association, 26(10), 1074-1082.
  • Coiera, E., Kocaballi, B., Halamka, J., & Laranjo, L. (2018). The digital scribe. NPJ digital medicine, 1(1), 1-5.