Proceedings

Two-Fold Patch Perturbation for Efficient Self-Supervised Learning in 3D Medical Imaging

Tirthajit Baruah, Kabir Jamadar, Punit Rathore

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. AI and Health. Pages 6654-6662.

https://doi.org/10.24963/ijcai.2026/740

Keywords

  • Medical knowledge representation: Medical knowledge representation
  • Self-supervised learning: Self-supervised learning
  • Medical imaging: Medical imaging
  • Medical diagnosis: Medical diagnosis

Abstract

Self-supervised pre-training has become a key paradigm for reducing annotation costs in 3D medical imaging, yet many recent approaches rely on complex objectives or incur substantial computational overhead. We propose a simple and efficient self-supervised pre-training framework for 3D medical images based on a two-fold patch-wise perturbation strategy. The method applies Bernoulli patch masking and discrete rotations, and trains a shared encoder with a three-head objective for reconstruction, perturbation localization, and rotation prediction. This design encourages spatially aware and transferable representations while remaining computationally lightweight. Experiments across diverse segmentation and classification benchmarks, including modality-shift scenarios, demonstrate consistent improvements over general self-supervised baselines and competitive or superior performance compared to recent medical self-supervised methods, while requiring substantially less memory, computation, and training time than the state-of-the-art pre-training pipelines.

Tirthajit Baruah, Kabir Jamadar, and Punit Rathore. 2026. Two-Fold Patch Perturbation for Efficient Self-Supervised Learning in 3D Medical Imaging. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 6654–6662. https://doi.org/10.24963/ijcai.2026/740
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards