To forget is to preserve: Machine Unlearning for 3D medical image segmentation
2026-06-15 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMachine Learning
AI summaryⓘ
The authors studied how to make a machine learning model "forget" specific personal data to comply with privacy laws like GDPR. They tested different methods on a brain scan dataset using a 3D ResNet-50 model. Their experiments showed that one method called Noisy Label worked best by effectively removing the data it was told to forget while keeping most of the important information intact. Other methods either forgot too much or damaged the model's overall performance. This work helps set clear benchmarks for safely and effectively unlearning data from models.
General Data Protection Regulation (GDPR)machine learning unlearning3D ResNet-50Med3D frameworkMRBrainS18 datasetDice similarity coefficientmean absolute error (MAE)Noisy Label strategymodel retention accuracydata privacy compliance
Authors
Nitesh Kumar Singh, Akhilesh Singh, Arjun Arora
Abstract
With new data privacy laws such as the General Data Protection Regulation (GDPR) [1] that allow individuals to ask that any of their personal information be erased from trained machine learning models, there has been a push to investigate the unlearning of data from models as a way to comply with these laws. In this regard, based on four mechanics, we consider several approximate unlearning strategies applied to the MRBrainS18 dataset [2]. We use a 3D ResNet-50 [3] as a backbone architecture for segmentation that has been pre-trained with the Med3D framework [4]. Considering the pre-trained model as a baseline, we evaluate respective retention accuracy on 2 types of subjects, i.e., retain and forget. We assess these approaches through their Dice similarity coefficient and mean absolute error (MAE) values using two separate training horizons 20 and 50 epochs. The results show that the Noisy Label strategy had the best overall trade-off with a decrease of 93% in the forget set while maintaining 84% accuracy for the retained set after 50 epochs. All other strategies showed extreme levels of forgetting at higher epoch numbers while also demonstrating catastrophic degradation of their retain set performance. The results of this study provide a strict baseline of performance metrics for unlearning on a subject-specific level and provide practitioners with clear criteria for selecting the proper strategies.