Loading [MathJax]/extensions/MathMenu.js
Ensemble Machine Learning Approach for Brain Tumor Classification Analysis | IEEE Conference Publication | IEEE Xplore

Ensemble Machine Learning Approach for Brain Tumor Classification Analysis


Abstract:

Diagnosis of Cancer has become very common these days, and brain tumor is diagnosed second commonly after breast cancer. Classification of brain tumors is an important ou...Show More

Abstract:

Diagnosis of Cancer has become very common these days, and brain tumor is diagnosed second commonly after breast cancer. Classification of brain tumors is an important outcome in identifying the presence of tumors and deciding on treatment. Due to its low survival rate, the importance of proper classification is required. To solve this problem, CAD model is used to find the easiest way, and machine learning algorithms are used with incredible accuracy to get optimistic results. Recent advances in building better models have received attention and various deep learning algorithms have been used. In this work, several classification algorithms were implemented to achieve the best results. MRI images of the brain are considered more important than CT images. This dataset contains about 300 MRI images along with the dataset from Cancer Archives and is divided into abnormal and normal classes to obtain binary classification results. In the initial stage, the preprocessing method is used, and in the later final stage, the classification algorithm is used. To improve accuracy, an ensemble of different classifiers is used with a set of respective input data. Thus, the proposed model proved to provide more precise accuracy than the previous classifier models used in the existing method.
Date of Conference: 16-18 February 2022
Date Added to IEEE Xplore: 10 May 2022
ISBN Information:
Conference Location: Trichy, India

I. Introduction

Magnetic Resonance Imaging (MRI) is a non-invasive, medical imaging that delivers excellent images of the human body and organs in both 2 dimensional (D) and 3D formats [1]. It is widely used and considered to be one of the most accurate techniques for cancer detection and classification, due to its high-resolution images on the brain tissue. It has had a significant influence on the field of automated medical image analysis because of its capacity to deliver a wealth of information on brain anatomy and abnormalities. Tumors may have various shapes and there may not be enough visible landmarks in the image to contribute to an accurate decision. Tumors come in a variety of forms, and there may not be enough apparent markers in the picture to provide an appropriate diagnosis. As a result, it's believed that human diagnosis is inherently unreliable. Furthermore, a misdiagnosis of the kind of brain tumor can be a major problem since it prevents patients from responding well to medical intervention and reduces their chances of survival. On the other hand, an accurate diagnosis will enable the patient to promptly begin the appropriate therapy and live a longer life.

Contact IEEE to Subscribe

References

References is not available for this document.