• Home
  • Search
  • Two-headed UNetEfficientNets for parallel execution of segmentation and classification of brain tumors: incorporating postprocessing techniques with connected component labelling
  • Cite Icon25
  • https://doi.org/10.1007/s00432-024-05718-1Copy DOI Icon

Two-headed UNetEfficientNets for parallel execution of segmentation and classification of brain tumors: incorporating postprocessing techniques with connected component labelling

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

PurposeThe purpose of this study is to develop accurate and automated detection and segmentation methods for brain tumors, given their significant fatality rates, with aggressive malignant tumors like Glioblastoma Multiforme (GBM) having a five-year survival rate as low as 5 to 10%. This underscores the urgent need to improve diagnosis and treatment outcomes through innovative approaches in medical imaging and deep learning techniques.MethodsIn this work, we propose a novel approach utilizing the two-headed UNetEfficientNets model for simultaneous segmentation and classification of brain tumors from Magnetic Resonance Imaging (MRI) images. The model combines the strengths of EfficientNets and a modified two-headed Unet model. We utilized a publicly available dataset consisting of 3064 brain MR images classified into three tumor classes: Meningioma, Glioma, and Pituitary. To enhance the training process, we performed 12 types of data augmentation on the training dataset. We evaluated the methodology using six deep learning models, ranging from UNetEfficientNet-B0 to UNetEfficientNet-B5, optimizing the segmentation and classification heads using binary cross entropy (BCE) loss with Dice and BCE with focal loss, respectively. Post-processing techniques such as connected component labeling (CCL) and ensemble models were applied to improve segmentation outcomes.ResultsThe proposed UNetEfficientNet-B4 model achieved outstanding results, with an accuracy of 99.4% after postprocessing. Additionally, it obtained high scores for DICE (94.03%), precision (98.67%), and recall (99.00%) after post-processing. The ensemble technique further improved segmentation performance, with a global DICE score of 95.70% and Jaccard index of 91.20%.ConclusionOur study demonstrates the high efficiency and accuracy of the proposed UNetEfficientNet-B4 model in the automatic and parallel detection and segmentation of brain tumors from MRI images. This approach holds promise for improving diagnosis and treatment planning for patients with brain tumors, potentially leading to better outcomes and prognosis.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00432-024-05718-1.

Loading PDF

Similar Papers
  • Research Article

A Novel Approach for Detection, Segmentation, and Classification of Brain Tumors in MRI Images Using Neural Network and Special C Means Fuzzy Clustering Techniques

  • Aug 25, 2024
  • Advances in Nonlinear Variational Inequalities
  • Sangeeta Kakarwal
  • Research Article
  • Citations2

K-NET+SEGAN-BASED SEGMENTATION WITH GANNET AQUILA OPTIMIZATION ALGORITHM-ENABLED DEEP MAXOUT NETWORK FOR BRAIN TUMOR CLASSIFICATION USING MRI

  • Jun 01, 2023
  • Journal of Mechanics in Medicine and Biology
  • Sakthi Ulaganathan +2
  • Research Article
  • Citations36

Deep multi-task learning structure for segmentation and classification of supratentorial brain tumors in MR images

  • Nov 25, 2023
  • Interdisciplinary Neurosurgery
  • Shirin Kordnoori +3
  • PDF
  • Research Article
  • Citations188

A Survey of Brain Tumor Segmentation and Classification Algorithms.

  • Sep 06, 2021
  • Journal of Imaging
  • Erena Siyoum Biratu +3
  • Research Article
  • Citations2

A Novel Deep Learning Approach for Brain Tumor Segmentation and Classification

  • Apr 04, 2024
  • Journal of Electrical Systems
  • Sangeeta Kakarwal, Rahul Mapari
  • Research Article
  • Citations1

Improvements to the Brain Tumor Segmentation and Classification System Using Convolutional Neural Networks

  • Jul 24, 2023
  • Tuijin Jishu/Journal of Propulsion Technology
  • Gaurav Kumar Verma, Shailendra Kumar
  • Research Article
  • Citations2

Automated Brain Tumor Detection and Classification in MRI Images: A Hybrid Image Processing Techniques

  • Jan 01, 2024
  • Journal of Cybersecurity and Information Management
  • N N
  • PDF
  • Research Article

Brain Tumor Prediction and Segmentation with Morphological Region-based Active Contour Model and Refinement using Boltzmann Monte Carlo Method in MRI Images

  • May 17, 2024
  • Indian Journal of Science and Technology
  • K R Srivaishnavi +2
  • Book Chapter
  • Citations14

Automated Brain Tumor Segmentation and Classification Through MRI Images

  • Jan 01, 2022
  • Sahar Gull +4
  • Book Chapter
  • Citations65

Artificial Intelligence in Brain Tumor Detection through MRI Scans

  • Jul 07, 2021
  • Sahar Gull +1
  • Research Article
  • Citations27

A secure hybrid deep learning framework for brain tumor detection and classification

  • Mar 24, 2025
  • Journal of Big Data
  • Sandeep Kumar Mathivanan +5
  • Addendum
  • Citations72

RETRACTED ARTICLE: Effective segmentation and classification of brain tumor using rough K means algorithm and multi kernel SVM in MR images

  • Aug 20, 2020
  • Journal of Ambient Intelligence and Humanized Computing
  • S Krishnakumar +1
  • Conference Article
  • Citations3

Fully automated segmentation of brain tumor from multiparametric MRI using 3D context u-net with deep supervision

  • Feb 15, 2021
  • Mingquan Lin +7
  • Conference Article
  • Citations9

Automated MRI Brain Tumour Segmentation and Classification Based on Deep Learning Techniques

  • Apr 21, 2022
  • K Srilatha +4
  • Research Article
  • Citations247

Robust clinical applicable CNN and U-Net based algorithm for MRI classification and segmentation for brain tumor

  • Oct 28, 2023
  • Expert Systems With Applications
  • Atika Akter +7
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.