A Real Time Application for Recognition of Brain Tumor Using Convolutional Neural Network

Although lots of work has been done for the detection of the brain tumor still there is the gap to improve the results in terms of DSC and accuracy. Existing work still has limitations and challenges for identifying substructures of the tumor region and classification of healthy/unhealthy images. A

Project Title

A Real Time Application for Recognition of Brain Tumor Using Convolutional Neural Network

Project Area of Specialization

Computer Science

Project Summary

Although lots of work has been done for the detection of the brain tumor still there is the gap to improve the results in terms of DSC and accuracy. Existing work still has limitations and challenges for identifying substructures of the tumor region and classification of healthy/unhealthy images. A few challenges in accurate detection of brain tumor are reviewed below:

1. Brain tumor grows very fast and its average size grows rapidly. Accordingly, tumor diagnosis at a preliminary    phase is a really challenging task.

  2. Segmentation is a very important step for the accurate detection of the brain tumor but performing this step is a challenging task due to the following factors:

  a) the noise appear in MRI due to the magnetic field variation in the coil. A few issues emerge in the segmentation process such as bias field due to intensity consistency and biased volume impact.

  b) Brain tumor segmentation is a highly imbalanced data problem as it is very difficult to segment tumor substructure  due to the imbalanced tumor labels.

Project Objectives

This Project aims to detect brain tumor earlier using MRI which may increase patients survival rate. Because in MRI, tumor is shown more clearly that provide helps in the process of further treatment.

Project Implementation Method

 We will implement strategies/algorithms that  will automatically hit upon the tumor in MRI. Automated prognosis of the tumor by way of the use of MRI consists of numerous steps which encompass: preprocessing (Enhancement), segmentation, feature extraction, and classification .

 Preprocessing phase will enhances the Region of Interest (ROI) that directly affects the segmentation outcomes.

 In the feature extraction phase, extra adequate functions are extracted by means of candidate regions. These extracted features may be similarly concatenated/fused for the quality result of brain tumor category.

 Finally, Classification phase which will classify it as malignant or benign.

Benefits of the Project

The Benefits of our project is noise reduction and normalization of the photo intensity that enables for higher segmentation of brain tumor. In the feature extraction phase, extra adequate functions are extracted by means of candidate regions. These extracted features may be similarly concatenated/fused for the quality result of mind tumor category.

Technical Details of Final Deliverable

 E Draw - Methodology Diagrams

MS Word - Registration Form, Project Proposal

MS PowerPoint - Project Defense Presentation

 Documentation: Microsoft Word, Google Docx

 Software: MATLAB R2021a

Hardware: Intel Core i5 processor or advanced version ;Minimum 8GB of RAM; Minimum 1GB of Hard disk Space

Final Deliverable of the Project

Software System

Core Industry

Health

Other Industries

Core Technology

NeuroTech

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com