MLHOps: Machine Learning for Healthcare Operations
It addresses the problem of reliable and ethical ML deployment for healthcare developers and clinicians, but is incremental as it synthesizes existing work into guidelines.
This paper tackles the challenge of deploying and maintaining machine learning models in healthcare by providing a survey and guidelines for MLHOps, covering processes from setup to long-term monitoring and ethical considerations.
Machine Learning Health Operations (MLHOps) is the combination of processes for reliable, efficient, usable, and ethical deployment and maintenance of machine learning models in healthcare settings. This paper provides both a survey of work in this area and guidelines for developers and clinicians to deploy and maintain their own models in clinical practice. We cover the foundational concepts of general machine learning operations, describe the initial setup of MLHOps pipelines (including data sources, preparation, engineering, and tools). We then describe long-term monitoring and updating (including data distribution shifts and model updating) and ethical considerations (including bias, fairness, interpretability, and privacy). This work therefore provides guidance across the full pipeline of MLHOps from conception to initial and ongoing deployment.