Related Experiment Video
Updated: Nov 18, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.1K
A Deep Learning Approach for Brain Tumor Classification and Segmentation Using a Multiscale Convolutional Neural
Francisco Javier Díaz-Pernas1, Mario Martínez-Zarzuela1, Míriam Antón-Rodríguez1
1Department of Signal Theory, Communications and Telematics Engineering, Telecommunications Engineering School, University of Valladolid, 47011 Valladolid, Spain.
Healthcare (Basel, Switzerland)
|February 5, 2021
Summary
This study introduces an automatic brain tumor segmentation and classification model using a Deep Convolutional Neural Network. The novel multiscale approach achieved high accuracy in identifying meningioma, glioma, and pituitary tumors from MRI scans.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neuroscience
Background:
- Accurate brain tumor segmentation and classification are crucial for effective treatment planning.
- Existing methods often require extensive preprocessing and struggle with diverse tumor types and views.
Purpose of the Study:
- To develop a fully automatic Deep Convolutional Neural Network (Deep Convolutional Neural Network) model for brain tumor segmentation and classification.
- To enhance tumor analysis by incorporating a multiscale processing approach inspired by the Human Visual System.
Main Methods:
- A Deep Convolutional Neural Network (Deep Convolutional Neural Network) model processing input MRI images across three spatial scales.
- The model analyzes sagittal, coronal, and axial views without requiring skull or vertebral column removal preprocessing.
- Evaluation on a dataset of 3064 MRI slices from 233 patients.
Main Results:
- Achieved a remarkable tumor classification accuracy of 0.973.
- Outperformed previous classical machine learning and deep learning methods on the same dataset.
- Demonstrated effective segmentation and classification of meningioma, glioma, and pituitary tumors.
Conclusions:
- The proposed fully automatic multiscale Deep Convolutional Neural Network (Deep Convolutional Neural Network) model offers a significant advancement in brain tumor analysis.
- The method's ability to process images without preprocessing and its high accuracy make it a promising tool for clinical applications.
