Abstract | Background: Though there has been considerable effort to implement machine learning methods for healthcare, clinical implementation has lagged. Incorporating explainable machine learning methods through the development of a decision support tool using a design thinking approach is expected to lead to greater uptake of such tools.
Objective: The objective of this work was to explore how constant engagement of clinician end-users can address the lack of adoption of ML tools in clinical contexts due to their lack of transparency, and address challenges related to presenting explainability in a decision support interface.
Methods: A Design Thinking framework was proposed that can be applied when developing an ML-based clinical decision support tool that incorporates explainability. The framework was then used in the development of a clinician-facing interface for the quantification of COVID-19 severity from chest x-ray images. We tested the resulting prototype with clinicians to verify domain appropriate representation, potential actionability, and consistency of the tool.
Results: During testing with clinicians, there was an overall positive reception concerning the design, usefulness, and applicability of the tool. Results show that using the proposed approach can improve clinician’s receptivity of clinical decision support tools.
Conclusions: The Design Thinking approach was augmented by using frameworks to help inform the development focus and direction, and to continuously evaluate the design on elements that would improve clinician trust. The approach outlines a direction for machine learning experts, user experience designers, and clinician end users on how to collaborate in the creation of trustworthy and usable XML based clinical decision support tools. Clinical Trial: N/A |
---|