HI-Net: Hyperdense Inception 3D UNet for Brain Tumor Segmentation

TL;DR

This paper introduces HI-Net, a novel 3D UNet architecture with hyperdense inception modules that effectively captures multi-scale features for improved brain tumor segmentation in MRI images.

Contribution

The paper proposes a hyperdense inception 3D UNet (HI-Net) that enhances multi-scale feature extraction using hyper dense connections and factorized convolutional layers.

Findings

Achieved dice scores of 0.79457 for ET, 0.87494 for WT, and 0.83712 for TC on BRATS 2020 dataset.

Demonstrated improved segmentation performance over existing methods.

Validated effectiveness of hyperdense connections in capturing contextual information.

Abstract

The brain tumor segmentation task aims to classify tissue into the whole tumor (WT), tumor core (TC), and enhancing tumor (ET) classes using multimodel MRI images. Quantitative analysis of brain tumors is critical for clinical decision making. While manual segmentation is tedious, time-consuming, and subjective, this task is at the same time very challenging to automatic segmentation methods. Thanks to the powerful learning ability, convolutional neural networks (CNNs), mainly fully convolutional networks, have shown promising brain tumor segmentation. This paper further boosts the performance of brain tumor segmentation by proposing hyperdense inception 3D UNet (HI-Net), which captures multi-scale information by stacking factorization of 3D weighted convolutional layers in the residual inception block. We use hyper dense connections among factorized convolutional layers to extract more contexual information, with the help of features reusability. We use a dice loss function to cope with class imbalances. We validate the proposed architecture on the multi-modal brain tumor segmentation challenges (BRATS) 2020 testing dataset. Preliminary results on the BRATS 2020 testing set show that achieved by our proposed approach, the dice (DSC) scores of ET, WT, and TC are 0.79457, 0.87494, and 0.83712, respectively.