3D Multimodal Brain Tumor Segmentation and Grading Scheme based on Machine, Deep, and Transfer Learning Approaches

Abstract

Glioma is one of the most common tumors of the brain. The detection and grading of glioma at an early stage is very critical for increasing the survival rate of the patients. Computer-aided detection (CADe) and computer-aided diagnosis (CADx) systems are essential and important tools that provide more accurate and systematic results to speed up the decision-making process of clinicians. In this paper, we introduce a method consisting of the variations of the machine, deep, and transfer learning approaches for the effective brain tumor (i.e., glioma) segmentation and grading on the multimodal brain tumor segmentation (BRATS) 2020 dataset. We apply popular and efficient 3D U-Net architecture for the brain tumor segmentation phase. We also utilize 23 different combinations of deep feature sets and machine learning/fine-tuned deep learning CNN models based on Xception, IncResNetv2, and EfficientNet by using 4 different feature sets and 6 learning models for the tumor grading phase. The experimental results demonstrate that the proposed method achieves 99.5% accuracy rate for slice-based tumor grading on BraTS 2020 dataset. Moreover, our method is found to have competitive performance with similar recent works.

Authors and Affiliations

Erdal Tasci, Aybars Ugur, Kevin Camphausen, Ying Zhuge, Rachel Zhao, Andra Krauze

Keywords

Related Articles

Cross-Project Fault Prediction using Artificial Intelligence

Software defect prediction project focuses on finding errors or flaws in software and aiming to improve accuracy which gives evolution batch with detectable results while adding to modern outcomes and advancement liabili...

Advancements in Neuroradiology via Artificial Intelligence and Machine Learning

Neuroradiology is significantly showing the broad impact in field of Artificial intelligence research and also in Machine learning. Neuro-radiology includes methods such as neuro-imaging which simply diagnose and charact...

Utilizing CRISPR as a Novel Tool for the Induction of Cell Reprogramming

Researchers can now target specific DNA sequences and easily modify them thanks to recent developments in CRISPR technology, enabling genome manipulation with unmatched precision. Furthermore, cell reprogramming is one o...

Review of Bioinformatics Tools and Techniques to Accelerate Ovarian Cancer Research

Since the history of humans there was no definitive cure for cancer. The rapid development in the field of bioinformatics has resulted in acceleration of advancement of cancer research. As computing and IT technology imp...

The COVID-19 Omicron Wave in the Framework of a New Mathematical Modeling in Few European Countries and the Right Time for Lifting Restrictions

Objectives: The COVID-19 Omicron wave in Romania, Bulgaria and Germany is considered in retrospective till begin of March 2022. The aim is to describe both country specific features and common trends related the same und...

Download PDF file
  • EP ID EP724395
  • DOI https://doi.org/10.61797/ijbic.v1i2.153
  • Views 82
  • Downloads 0

How To Cite

Erdal Tasci, Aybars Ugur, Kevin Camphausen, Ying Zhuge, Rachel Zhao, Andra Krauze (2022). 3D Multimodal Brain Tumor Segmentation and Grading Scheme based on Machine, Deep, and Transfer Learning Approaches. International Journal of Bioinformatics and Intelligent Computing, 1(2), -. https://europub.co.uk/articles/-A-724395