# Statistically Valid Post-Deployment Monitoring Should Be Standard for AI-Based Digital Health

arXiv: https://arxiv.org/abs/2506.05701 · HTML: https://mlsanity.com/paper/2506.05701/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 117 references that are in the map.
- First posted on arXiv 6 Jun 2025.

## Builds on (most cited first, 23 of 23)

- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [Continual Lifelong Learning with Neural Networks: A Review](https://mlsanity.com/paper/1802.07569/) (2018)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Deep Bayesian Active Learning with Image Data](https://mlsanity.com/paper/1703.02910/) (2017)
- [Computational Optimal Transport](https://mlsanity.com/paper/1803.00567/) (2018)
- [Inherent Trade-Offs in the Fair Determination of Risk Scores](https://mlsanity.com/paper/1609.05807/) (2016)
- [Learning from Noisy Labels with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2007.08199/) (2020)
- [Learning under Concept Drift: A Review](https://mlsanity.com/paper/2004.05785/) (2020)
- [Confounding variables can degrade generalization performance of radiological deep learning models](https://mlsanity.com/paper/1807.00431/) (2018)
- [Detecting and Correcting for Label Shift with Black Box Predictors](https://mlsanity.com/paper/1802.03916/) (2018)
- [Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift](https://mlsanity.com/paper/1810.11953/) (2018)
- [Change is Hard: A Closer Look at Subpopulation Shift](https://mlsanity.com/paper/2302.12254/) (2023)
- [A Unified View of Label Shift Estimation](https://mlsanity.com/paper/2003.07554/) (2020)
- [Feature Robustness in Non-stationary Health Records: Caveats to Deployable Model Performance in Common Clinical Machine Learning Tasks](https://mlsanity.com/paper/1908.00690/) (2019)
- [Weak Supervision for Fake News Detection via Reinforcement Learning](https://mlsanity.com/paper/1912.12520/) (2019)
- [SAFFRON: an adaptive algorithm for online control of the false discovery rate](https://mlsanity.com/paper/1802.09098/) (2018)
- [CheXstray: Real-time Multi-Modal Data Concordance for Drift Detection in Medical Imaging AI](https://mlsanity.com/paper/2202.02833/) (2022)
- [Concept Drift Detection and Adaptation with Hierarchical Hypothesis Testing](https://mlsanity.com/paper/1707.07821/) (2017)
- [A label-efficient two-sample test](https://mlsanity.com/paper/2111.08861/) (2021)
- [Active Sequential Two-Sample Testing](https://mlsanity.com/paper/2301.12616/) (2023)
- [MLHOps: Machine Learning for Healthcare Operations](https://mlsanity.com/paper/2305.02474/) (2023)
- [Validity problems in clinical machine learning by indirect data labeling using consensus definitions](https://mlsanity.com/paper/2311.03037/) (2023)
- [Advanced Tutorial: Label-Efficient Two-Sample Tests](https://mlsanity.com/paper/2501.03568/) (2025)

## Built on by (most cited first, 8 of 8)

- [Pre-Trained AI Model Assisted Online Decision-Making under Missing Covariates: A Theoretical Perspective](https://mlsanity.com/paper/2507.07852/) (2025)
- [Monitoring Deployed AI Systems in Health Care](https://mlsanity.com/paper/2512.09048/) (2025)
- [Adversarial Drift-Aware Predictive Transfer: Toward Durable Clinical AI](https://mlsanity.com/paper/2601.11860/) (2026)
- [Active Multiple-Prediction-Powered Inference](https://mlsanity.com/paper/2605.08429/) (2026)
- [NeurIPS Should Require Reproducibility Standards for Frontier AI Safety Claims](https://mlsanity.com/paper/2605.08192/) (2026)
- [The Open-Box Fallacy: Why AI Deployment Needs a Calibrated Verification Regime](https://mlsanity.com/paper/2605.10601/) (2026)
- [An Investigation of the NeurIPS and ICML 2025 Position Tracks](https://mlsanity.com/paper/2608.16894/) (2026)
- [Comparison of techniques for fine-tuning open-weight models for entity extraction from radiology reports](https://mlsanity.com/paper/2609.40236/) (2026)

## Abstract (arXiv)

This position paper argues that post-deployment monitoring in clinical AI is underdeveloped and proposes statistically valid and label-efficient testing frameworks as a principled foundation for ensuring reliability and safety in real-world deployment. A recent review found that only 9% of FDA-registered AI-based healthcare tools include a post-deployment surveillance plan. Existing monitoring approaches are often manual, sporadic, and reactive, making them ill-suited for the dynamic environments in which clinical models operate. We contend that post-deployment monitoring should be grounded in label-efficient and statistically valid testing frameworks, offering a principled alternative to current practices. We use the term "statistically valid" to refer to methods that provide explicit guarantees on error rates (e.g., Type I/II error), enable formal inference under pre-defined assumptions, and support reproducibility--features that align with regulatory requirements. Specifically, we propose that the detection of changes in the data and model performance degradation should be framed as distinct statistical hypothesis testing problems. Grounding monitoring in statistical rigor ensures a reproducible and scientifically sound basis for maintaining the reliability of clinical AI systems. Importantly, it also opens new research directions for the technical community--spanning theory, methods, and tools for statistically principled detection, attribution, and mitigation of post-deployment model failures in real-world settings.