# Dense Multi-path U-Net for Ischemic Stroke Lesion Segmentation in Multiple Image Modalities

arXiv: https://arxiv.org/abs/1810.07003 · HTML: https://mlsanity.com/paper/1810.07003/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Oct 2018.

## Built on by (most cited first, 13 of 13)

- [Medical Image Segmentation Review: The success of U-Net](https://mlsanity.com/paper/2211.14830/) (2022)
- [Dense Nested Attention Network for Infrared Small Target Detection](https://mlsanity.com/paper/2106.00487/) (2021)
- [DFENet: A Novel Dimension Fusion Edge Guided Network for Brain MRI Segmentation](https://mlsanity.com/paper/2105.07962/) (2021)
- [Multi-scale Super-resolution Magnetic Resonance Spectroscopic Imaging with Adjustable Sharpness](https://mlsanity.com/paper/2206.08984/) (2022)
- [APIS: A paired CT-MRI dataset for ischemic stroke segmentation challenge](https://mlsanity.com/paper/2309.15243/) (2023)
- [Flow-based Visual Quality Enhancer for Super-resolution Magnetic Resonance Spectroscopic Imaging](https://mlsanity.com/paper/2207.10181/) (2022)
- [Neural Network-derived perfusion maps: a Model-free approach to computed tomography perfusion in patients with acute ischemic stroke](https://mlsanity.com/paper/2101.05992/) (2021)
- [DXM-TransFuse U-net: Dual Cross-Modal Transformer Fusion U-net for Automated Nerve Identification](https://mlsanity.com/paper/2202.13304/) (2022)
- [Stroke Lesion Segmentation with Visual Cortex Anatomy Alike Neural Nets](https://mlsanity.com/paper/2105.06544/) (2021)
- [Identification of Physical Processes and Unknown Parameters of 3D Groundwater Contaminant Problems via Theory-guided U-net](https://mlsanity.com/paper/2205.00134/) (2022)
- [Deep Attention Unet: A Network Model with Global Feature Perception Ability](https://mlsanity.com/paper/2304.10829/) (2023)
- [CDSE-UNet: Enhancing COVID-19 CT Image Segmentation with Canny Edge Detection and Dual-Path SENet Feature Fusion](https://mlsanity.com/paper/2403.01513/) (2024)
- [DM-QPMNET: Dual-modality fusion network for cell segmentation in quantitative phase microscopy](https://mlsanity.com/paper/2511.00218/) (2025)

## Abstract (arXiv)

Delineating infarcted tissue in ischemic stroke lesions is crucial to determine the extend of damage and optimal treatment for this life-threatening condition. However, this problem remains challenging due to high variability of ischemic strokes' location and shape. Recently, fully-convolutional neural networks (CNN), in particular those based on U-Net, have led to improved performances for this task. In this work, we propose a novel architecture that improves standard U-Net based methods in three important ways. First, instead of combining the available image modalities at the input, each of them is processed in a different path to better exploit their unique information. Moreover, the network is densely-connected (i.e., each layer is connected to all following layers), both within each path and across different paths, similar to HyperDenseNet. This gives our model the freedom to learn the scale at which modalities should be processed and combined. Finally, inspired by the Inception architecture, we improve standard U-Net modules by extending inception modules with two convolutional blocks with dilated convolutions of different scale. This helps handling the variability in lesion sizes. We split the 93 stroke datasets into training and validation sets containing 83 and 9 examples respectively. Our network was trained on a NVidia TITAN XP GPU with 16 GBs RAM, using ADAM as optimizer and a learning rate of 1×10-5 during 200 epochs. Training took around 5 hours and segmentation of a whole volume took between 0.2 and 2 seconds, as average. The performance on the test set obtained by our method is compared to several baselines, to demonstrate the effectiveness of our architecture, and to a state-of-art architecture that employs factorized dilated convolutions, i.e., ERFNet.