Extending SOUP to ML Models When DesigningCertified Medical Systems

TL;DR

This paper explores the challenges of integrating third-party machine learning models into certified medical systems, proposing practical strategies to manage regulatory and design complexities.

Contribution

It extends the SOUP concept to ML models and introduces methods for managing their integration in regulated medical device development.

Findings

Proposes a framework for managing ML models as SOUP in medical systems

Addresses regulatory considerations for third-party ML components

Provides practical guidelines for integrating ML models into certified medical devices

Abstract

Software of Unknown Provenance, SOUP, refers to a software component that is already developed and widely available from a 3rd party, and that has not been developed, to be integrated into a medical device. From regulatory perspective, SOUP software requires special considerations, as the developers' obligations related to design and implementation are not applied to it. In this paper, we consider the implications of extending the concept of SOUP to machine learning (ML) models. As the contribution, we propose practical means to manage the added complexity of 3rd party ML models in regulated development.