Physician Experience Design (PXD): More Usable Machine Learning Prediction for Clinical Decision Making

Lu Wang; Mark Chignell; Yilun Zhang; Andrew Pinto; Fahad Razak; Kathleen Sheehan; Amol Verma

Physician Experience Design (PXD): More Usable Machine Learning Prediction for Clinical Decision Making

AMIA Jt Summits Transl Sci Proc. 2022 May 23:2022:476-485. eCollection 2022.

Authors

Lu Wang¹, Mark Chignell¹, Yilun Zhang¹, Andrew Pinto^{2

3}, Fahad Razak^{4

5

6}, Kathleen Sheehan^{7

8}, Amol Verma^{4

5

6}

Affiliations

¹ Dept. of Mechanical & Industrial Engineering, University of Toronto (UofT), Toronto, ON M5S 2E4, Canada (CA).
² Dalla Lana School of Public Health, UofT, Toronto, ON M5S 3G8, CA.
³ Dept. of Family and Community Medicine, St. Michael's Hospital, Unity Health, Toronto, ON M5B 1W8, CA.
⁴ Dept. of Psychiatry, UofT, Toronto, ON M5B 1W8, CA.
⁵ Dept. of Medicine, UofT, Toronto, ON M5S 1A8, CA.
⁶ Inst. of Health Policy, Management and Evaluation, UofT, Toronto, ON M5T 3M6, CA.
⁷ GEMINI, Unity Health, Toronto, ON M5B 1W8, CA.
⁸ Dalla Lana School of Public Health, UofT, Toronto, ON M5T 2S8, CA.

PMID: 35854747
PMCID: PMC9285165

Abstract

Delirium is an acute neurocognitive disorder, which is difficult to identify and predict. Using GEMINI, Canada's largest hospital data and analytics study, we had a labeled sample of around 4,000 cases with approximately 25% of cases being labeled as having delirium. Based on this labeled data, we developed machine learning (ML) models and interacted with physicians to interpret the ML models and their predictions. We developed a preliminary Explainable Artificial Intelligence (XAI) framework for physician experience design (PXD) to improve the uptake of ML models by improving the transparency of model results, thereby increasing physician trust in models as well as the uptake of model results for clinical decision making. We developed our PXD approach first with Conceptual Investigation to collect and extract physicians' feedback on ML models and their evaluation requirements. We carried out a case study, working closely with the physicians in a participatory design process to develop a dashboard that presents ML delirium identification results interactively based on physician selections and inputs. In this approach a physician-preferred ML model for clinical decision making is selected through PXD evaluation.