# Automatic Liver Lesion Segmentation Using A Deep Convolutional Neural Network Method

## TL;DR

This paper presents a deep convolutional neural network approach for automatic liver lesion segmentation, utilizing a 2.5D U-Net and ResNet architecture, achieving top performance in the LiTS challenge.

## Contribution

The study introduces a novel 2.5D DCNN model combining U-Net and ResNet features for improved liver lesion segmentation accuracy.

## Key findings

- Achieved an average Dice score of 0.67 on test data
- Ranked first in the LiTS challenge at ISBI 2017
- Demonstrated effectiveness of 2.5D CNN for medical image segmentation

## Abstract

Liver lesion segmentation is an important step for liver cancer diagnosis, treatment planning and treatment evaluation. LiTS (Liver Tumor Segmentation Challenge) provides a common testbed for comparing different automatic liver lesion segmentation methods. We participate in this challenge by developing a deep convolutional neural network (DCNN) method. The particular DCNN model works in 2.5D in that it takes a stack of adjacent slices as input and produces the segmentation map corresponding to the center slice. The model has 32 layers in total and makes use of both long range concatenation connections of U-Net [1] and short-range residual connections from ResNet [2]. The model was trained using the 130 LiTS training datasets and achieved an average Dice score of 0.67 when evaluated on the 70 test CT scans, which ranked first for the LiTS challenge at the time of the ISBI 2017 conference.