The Xavier AI Initiative’s Good Machine Learning Practices (GMLP) Team, led by Pat Baird (Philips) and Rohit Nayak (Electronic Registry Systems), has just released a new whitepaper, Building Explainability and Trust for AI in Healthcare. The paper was developed under the leadership of Xavier Health, in partnership with industry professionals, and was presented during the 2019 Xavier AI Summit.
The paper opens a discussion of current approaches for explainability and trust by highlighting commonly used systems-development constructs oriented towards healthcare.
By highlighting a series of considerations and questions, readers are enabled to make their own conclusions. The paper also includes with an in-depth review of how the degree of risk can drive the level of explainability necessary for applications of AI within a healthcare setting.
The intended audience of this paper is broad, and includes developers, implementers, researchers, quality assurance, regulatory affairs, validation personnel, business managers, regulators and end-users faced with challenge of assessing and building trust in AI healthcare applications.
To download your free copy of Building Explainability and Trust for AI in Healthcare, click here.
If you’re interested in joining the GMLP Team, click here.