Interactive Image Selection and Training for Brain Tumor Segmentation Network

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cerqueira, Matheus A., Sprenger, Flávia, Teixeira, Bernardo C. A., Falcão, Alexandre X.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914824885108736
author Cerqueira, Matheus A.
Sprenger, Flávia
Teixeira, Bernardo C. A.
Falcão, Alexandre X.
author_facet Cerqueira, Matheus A.
Sprenger, Flávia
Teixeira, Bernardo C. A.
Falcão, Alexandre X.
contents Medical image segmentation is a relevant problem, with deep learning being an exponent. However, the necessity of a high volume of fully annotated images for training massive models can be a problem, especially for applications whose images present a great diversity, such as brain tumors, which can occur in different sizes and shapes. In contrast, a recent methodology, Feature Learning from Image Markers (FLIM), has involved an expert in the learning loop, producing small networks that require few images to train the convolutional layers. In this work, We employ an interactive method for image selection and training based on FLIM, exploring the user's knowledge. The results demonstrated that with our methodology, we could choose a small set of images to train the encoder of a U-shaped network, obtaining performance equal to manual selection and even surpassing the same U-shaped network trained with backpropagation and all training images.
format Preprint
id arxiv_https___arxiv_org_abs_2406_03225
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Interactive Image Selection and Training for Brain Tumor Segmentation Network
Cerqueira, Matheus A.
Sprenger, Flávia
Teixeira, Bernardo C. A.
Falcão, Alexandre X.
Computer Vision and Pattern Recognition
68T07, 68T45
Medical image segmentation is a relevant problem, with deep learning being an exponent. However, the necessity of a high volume of fully annotated images for training massive models can be a problem, especially for applications whose images present a great diversity, such as brain tumors, which can occur in different sizes and shapes. In contrast, a recent methodology, Feature Learning from Image Markers (FLIM), has involved an expert in the learning loop, producing small networks that require few images to train the convolutional layers. In this work, We employ an interactive method for image selection and training based on FLIM, exploring the user's knowledge. The results demonstrated that with our methodology, we could choose a small set of images to train the encoder of a U-shaped network, obtaining performance equal to manual selection and even surpassing the same U-shaped network trained with backpropagation and all training images.
title Interactive Image Selection and Training for Brain Tumor Segmentation Network
topic Computer Vision and Pattern Recognition
68T07, 68T45
url https://arxiv.org/abs/2406.03225