# Random 2.5D U-net for Fully 3D Segmentation

arXiv: https://arxiv.org/abs/1910.10398 · HTML: https://mlsanity.com/paper/1910.10398/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Oct 2019.

## Built on by (most cited first, 2 of 2)

- [Recurrent U-net for automatic pelvic floor muscle segmentation on 3D ultrasound](https://mlsanity.com/paper/2107.13833/) (2021)
- [Validation of Various Normalization Methods for Brain Tumor Segmentation: Can Federated Learning Overcome This Heterogeneity?](https://mlsanity.com/paper/2510.07126/) (2025)

## Abstract (arXiv)

Convolutional neural networks are state-of-the-art for various segmentation tasks. While for 2D images these networks are also computationally efficient, 3D convolutions have huge storage requirements and therefore, end-to-end training is limited by GPU memory and data size. To overcome this issue, we introduce a network structure for volumetric data without 3D convolution layers. The main idea is to include projections from different directions to transform the volumetric data to a sequence of images, where each image contains information of the full data. We then apply 2D convolutions to these projection images and lift them again to volumetric data using a trainable reconstruction algorithm. The proposed architecture can be applied end-to-end to very large data volumes without cropping or sliding-window techniques. For a tested sparse binary segmentation task, it outperforms already known standard approaches and is more resistant to generation of artefacts.