# LEM-UNet: an edge-guided network for 3D multimodal images segmentation in focal cortical dysplasia

## TL;DR

This paper introduces LEM-UNet, a new network for segmenting brain lesions in focal cortical dysplasia using edge information to improve accuracy.

## Contribution

The novel integration of Laplacian pyramid-based edge attention and multi-strategy feature fusion improves lesion segmentation in FCD.

## Key findings

- LEM-UNet achieved Dice Coefficients of 0.452 and 0.597 on two datasets, outperforming the baseline.
- The model's edge-focused design improves segmentation of lesions with blurred boundaries.
- The Laplacian pyramid enhances the capture of subtle lesion features.

## Abstract

Focal cortical dysplasia (FCD) is one of the common causes of refractory epilepsy. The subtle and indistinct edge of FCD lesions pose considerable challenges for accurate lesion localization. Therefore, we propose an edge guided segmentation network based on Laplacian pyramid to improve the localization performance of FCD lesions.

This is a retrospective study evaluated on two independent datasets. The proposed Laplacian Edge Mix UNet (LEM-UNet) builds upon the MedNeXt baseline and incorporates the Laplacian Edge Attention (LEA) block and the Multi-strategy Feature Fusion (MFF) block. LEA block captures lesion details and edge information during the encoding phase by integrating Laplacian pyramid feature maps with an attention mechanism, while MFF block fuses edge features with high level features during the decoding phase.

The model's performance was assessed through 5-fold cross-validation across both Open and Private Datasets, demonstrating superior performance. The average Dice Coefficient achieved 0.452 and 0.597 on the Open and Private Datasets, respectively, representing improvements of 2.40% and 2.90% compared to the baseline model.

The results demonstrate the importance of focusing on lesion edge in the FCD segmentation task. The integration of the Laplacian pyramid enhances the mode's ability to capture lesions with blurred edge and subtle features. LEM-UNet exhibits significant advantages over current FCD segmentation algorithms. The source code and pre trained model weights are available at https://github.com/simplify403/LEM-UNet.