TuNet: End-to-end Hierarchical Brain Tumor Segmentation using Cascaded Networks

TL;DR

This paper introduces TuNet, an end-to-end hierarchical cascaded network for brain tumor segmentation that leverages tumor sub-region structure, ResNet-like blocks, and ensemble techniques to improve accuracy.

Contribution

It presents a novel cascaded network architecture that explicitly models tumor hierarchy and integrates advanced modules for improved segmentation performance.

Findings

Achieved dice scores of 88.06, 80.84, and 80.29 for tumor regions.

Reduced Hausdorff Distance (95th percentile) to 6.10, 5.17, and 2.21.

Demonstrated effectiveness of hierarchical modeling and ensemble methods.

Abstract

Glioma is one of the most common types of brain tumors; it arises in the glial cells in the human brain and in the spinal cord. In addition to having a high mortality rate, glioma treatment is also very expensive. Hence, automatic and accurate segmentation and measurement from the early stages are critical in order to prolong the survival rates of the patients and to reduce the costs of the treatment. In the present work, we propose a novel end-to-end cascaded network for semantic segmentation that utilizes the hierarchical structure of the tumor sub-regions with ResNet-like blocks and Squeeze-and-Excitation modules after each convolution and concatenation block. By utilizing cross-validation, an average ensemble technique, and a simple post-processing technique, we obtained dice scores of 88.06, 80.84, and 80.29, and Hausdorff Distances (95th percentile) of 6.10, 5.17, and 2.21 for the whole tumor, tumor core, and enhancing tumor, respectively, on the online test set.