Towards credible computational models: Application of a risk based framework for establishing credibility

Abstract

The use of computational modeling and simulation (CMS) as a tool for gaining insight into the technical performance and safety of medical devices has emerged continuously over the past years. However, to rely on information and decisions derived from model predictions, it is essential to establish model credibility for the specific context of use. Limited regulatory requirements and lack of consensus on the level of verification and validation activities required result in rare use of CMS as a source of evidence in the medical device approval process. The American Society of Mechanical Engineers (ASME) developed a risk-informed framework to establish appropriate credibility requirements of a computational model: the ASME V&V 40?2018 standard. This paper aims to outline the concepts of this standard and to demonstrate its application using an example from the orthotics field. The necessary steps to establish model credibility for a custom?made 3D printed wrist hand orthosis (WHO) are presented. It is shown that the credibility requirements of each verification and validation activity depend on model risk by applying two different contexts of use to the same computational model.

Acknowledgments
We gratefully acknowledge the support provided by the Mecuris GmbH (Munich, Germany) throughout this research, especially by Mr. Ylli Binakaj. Their input, commitment and collaboration significantly enhanced the quality of our research. Research funding: Funded by the German Federal Ministry of Education and Research (BMBF).

Author’s statement
Conflict of interest: Ann-Kathrin Carl, Maxim Kirillov, and David Hochmann state no conflict of interest. Eric Quadrat is an employee of Simq GmbH, Grafing near Munich, Germany.