Safe AI for health and beyond -- Monitoring to transform a health service
This tackles the critical issue of model maintenance for safe AI deployment in healthcare, though it is incremental as it builds on existing monitoring concepts.
The paper addresses the problem of predictive models becoming obsolete in healthcare due to changes in data and practices, proposing infrastructure for monitoring and updating models to ensure safe long-term use, with examples from breast cancer prognosis and neurodegenerative disease stratification.
Machine learning techniques are effective for building predictive models because they identify patterns in large datasets. Development of a model for complex real-life problems often stop at the point of publication, proof of concept or when made accessible through some mode of deployment. However, a model in the medical domain risks becoming obsolete as patient demographics, systems and clinical practices change. The maintenance and monitoring of predictive model performance post-publication is crucial to enable their safe and effective long-term use. We will assess the infrastructure required to monitor the outputs of a machine learning algorithm, and present two scenarios with examples of monitoring and updates of models, firstly on a breast cancer prognosis model trained on public longitudinal data, and secondly on a neurodegenerative stratification algorithm that is currently being developed and tested in clinic.