Does Machine Unlearning Preserve Clinical Safety? A Risk Analysis for Medical Image Classification

TL;DR

This paper examines how machine unlearning impacts clinical safety in medical image classification, proposing a risk-aware method that balances data removal with patient safety considerations.

Contribution

It introduces SalUn-CRA, a clinical risk-aware unlearning method that reduces harmful associations and maintains safety in medical AI models.

Findings

SalUn-CRA achieves lower clinical risk than standard unlearning methods.

Standard unlearning can increase false negatives, risking patient safety.

Risk-aware unlearning maintains model utility while enhancing safety.

Abstract

The application of Deep Learning in medical diagnosis must balance patient safety with compliance with data protection regulations. Machine Unlearning enables the selective removal of training data from deployed models. However, most methods are validated primarily through efficiency and privacy-oriented metrics, with limited attention to clinically asymmetric error costs. In this work, we investigate how unlearning affects clinical risk in binary medical image classification. We show that standard unlearning strategies (Fine-Tuning, Random Labeling, and SalUn) may reduce test utility while increasing false-negative rates, thereby amplifying clinical risk. To mitigate this, we propose SalUn-CRA (Clinical Risk-Aware), a variant of SalUn that replaces random relabeling with entropy-based forgetting for malignant samples in the forget set, preventing the model from learning harmful benign associations. We evaluate on DermaMNIST and PathMNIST medical image datasets under 20% and 50% data removal. Using Global Risk metrics with asymmetric costs, SalUn-CRA achieves lower or comparable clinical risk to full retraining while preserving unlearning effectiveness. These results suggest that clinical risk should be an integral component of unlearning validation in medical systems.