Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

An Ensemble Approach for Brain Tumor Segmentation and Synthesis

Juampablo E. Heras Rivera, Agamdeep S. Chopra, Tianyi Ren, Hitender Oswal, Yutong Pan, Zineb Sordo, Sophie Walters, William Henry, Hooman Mohammadi, Riley Olson, Fargol Rezayaraghi, Tyson Lam, Akshay Jaikanth, Pavan Kancharla, Jacob Ruzevick, Daniela Ushizima, Mehmet Kurt

TL;DR

A deep learning framework is proposed that ensembles these state-of-the-art architectures to achieve accurate segmentation and produce finely synthesized images in brain tumor segmentation.

Abstract

The integration of machine learning in magnetic resonance imaging (MRI), specifically in neuroimaging, is proving to be incredibly effective, leading to better diagnostic accuracy, accelerated image analysis, and data-driven insights, which can potentially transform patient care. Deep learning models utilize multiple layers of processing to capture intricate details of complex data, which can then be used on a variety of tasks, including brain tumor classification, segmentation, image synthesis, and registration. Previous research demonstrates high accuracy in tumor segmentation using various model architectures, including nn-UNet and Swin-UNet. U-Mamba, which uses state space modeling, also achieves high accuracy in medical image segmentation. To leverage these models, we propose a deep learning framework that ensembles these state-of-the-art architectures to achieve accurate segmentation and produce finely synthesized images.

An Ensemble Approach for Brain Tumor Segmentation and Synthesis

TL;DR

A deep learning framework is proposed that ensembles these state-of-the-art architectures to achieve accurate segmentation and produce finely synthesized images in brain tumor segmentation.

Abstract

The integration of machine learning in magnetic resonance imaging (MRI), specifically in neuroimaging, is proving to be incredibly effective, leading to better diagnostic accuracy, accelerated image analysis, and data-driven insights, which can potentially transform patient care. Deep learning models utilize multiple layers of processing to capture intricate details of complex data, which can then be used on a variety of tasks, including brain tumor classification, segmentation, image synthesis, and registration. Previous research demonstrates high accuracy in tumor segmentation using various model architectures, including nn-UNet and Swin-UNet. U-Mamba, which uses state space modeling, also achieves high accuracy in medical image segmentation. To leverage these models, we propose a deep learning framework that ensembles these state-of-the-art architectures to achieve accurate segmentation and produce finely synthesized images.

Paper Structure

This paper contains 28 sections, 1 figure, 4 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Data
  4. BraTS Adult Glioma Post Treatment (AGPT):
  5. BraTS-Africa Challenge dataset:
  6. BraTS Generalizability Across Tumors (BraTS-GoAT):
  7. BraTS Inpainting:
  8. Evaluation Metrics
  9. Model Architectures
  10. Optimized U-Net
  11. Training details
  12. RhizoNet
  13. nn-UNet
  14. Swin-UNetR
  15. U-Mamba
  16. ...and 13 more sections

Figures (1)

  • Figure 1: Architecture of our proposed Mask Aware 3D T-Former network for inpainting tasks. There are 2 flavours of the Transformer block, a standard gated linear attention block, and a gated linear attention block with an additional gated Fourier block for enhanced global contextual learning.