mlsanityData to 14 Oct 2026

To forget is to preserve: Machine Unlearning for 3D medical image segmentation

arXiv 2606.16180 ↗First posted 15 Jun 2026cs.CV cs.LGcited by 0

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 26 of its 37 references that are in the map. First posted on arXiv 15 Jun 2026.

Builds on

26 of 37 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024Erase to Enhance: Data-Efficient Machine Unlearning in MRI Reconstruction24 May
  2. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation19 Oct
  3. 2023Model Sparsity Can Simplify Machine Unlearning11 Apr
  4. 2022A Survey of Machine Unlearning6 Sep
  5. 2021Fast Yet Effective Machine Unlearning17 Nov
  6. 2021Machine Unlearning of Features and Labels26 Aug

Other references · 20

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  5. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  6. 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,292
  7. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  8. 20163D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotationcited by 1,476
  9. 2017A Survey on Deep Learning in Medical Image Analysiscited by 1,186
  10. 2019Machine Unlearningcited by 1,061
  11. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 821
  12. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 618
8 more
  1. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 465
  2. 2020Amnesiac Machine Learningcited by 319
  3. 2019Med3D: Transfer Learning for 3D Medical Image Analysiscited by 288
  4. 2016Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learningcited by 209
  5. 2020Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observationscited by 180
  6. 2022The Right to be Forgotten in Federated Learning: An Efficient Realization with Rapid Retrainingcited by 118
  7. 2022Recommendation Unlearningcited by 80
  8. 2024Machine Unlearning for Image-to-Image Generative Modelscited by 47

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Erase to Retain: Low Rank Adaptation Guided Selective Unlearning in Medical Segmentation Networks
  2. 2025Targeted Unlearning Using Perturbed Sign Gradient Methods With Applications On Medical Images
  3. 2024Machine Unlearning for Medical Imaging
  4. 2026Lethe: How Hard Is It to Forget? A Benchmark for Federated Unlearning in Medical Imaging
  5. 2025Forget-MI: Machine Unlearning for Forgetting Multimodal Information in Healthcare Settings
  6. 2026Models Designed to Forget: Machine Unlearning via Key Deletion
  7. 2024Label-Agnostic Forgetting: A Supervision-Free Unlearning in Deep Models
  8. 2025When to Forget? Complexity Trade-offs in Machine Unlearning