Saved in:
Bibliographic Details
Main Authors: Burgon, Alexis, Sahiner, Berkman, Petrick, Nicholas A, Pennello, Gene, Samala, Ravi K
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.04878
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915918019297280
author Burgon, Alexis
Sahiner, Berkman
Petrick, Nicholas A
Pennello, Gene
Samala, Ravi K
author_facet Burgon, Alexis
Sahiner, Berkman
Petrick, Nicholas A
Pennello, Gene
Samala, Ravi K
contents This work addresses challenges in evaluating adaptive artificial intelligence (AI) models for medical devices, where iterative updates to both models and evaluation datasets complicate performance assessment. We introduce a novel approach with three complementary measurements: learning (model improvement on current data), potential (dataset-driven performance shifts), and retention (knowledge preservation across modification steps), to disentangle performance changes caused by model adaptations versus dynamic environments. Case studies using simulated population shifts demonstrate the approach's utility: gradual transitions enable stable learning and retention, while rapid shifts reveal trade-offs between plasticity and stability. These measurements provide practical insights for regulatory science, enabling rigorous assessment of the safety and effectiveness of adaptive AI systems over sequential modifications.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04878
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning, Potential, and Retention: An Approach for Evaluating Adaptive AI-Enabled Medical Devices
Burgon, Alexis
Sahiner, Berkman
Petrick, Nicholas A
Pennello, Gene
Samala, Ravi K
Artificial Intelligence
Performance
This work addresses challenges in evaluating adaptive artificial intelligence (AI) models for medical devices, where iterative updates to both models and evaluation datasets complicate performance assessment. We introduce a novel approach with three complementary measurements: learning (model improvement on current data), potential (dataset-driven performance shifts), and retention (knowledge preservation across modification steps), to disentangle performance changes caused by model adaptations versus dynamic environments. Case studies using simulated population shifts demonstrate the approach's utility: gradual transitions enable stable learning and retention, while rapid shifts reveal trade-offs between plasticity and stability. These measurements provide practical insights for regulatory science, enabling rigorous assessment of the safety and effectiveness of adaptive AI systems over sequential modifications.
title Learning, Potential, and Retention: An Approach for Evaluating Adaptive AI-Enabled Medical Devices
topic Artificial Intelligence
Performance
url https://arxiv.org/abs/2604.04878