A novel enhanced softmax loss function for brain tumour detection using deep learning

Sunil Maharjan, Abeer Alsadoon, P. W.C. Prasad, Thair Al-Dalain, Omar Hisham Alsadoon

Research output: Contribution to journalArticle

Abstract

Background and Aim: In deep learning, the sigmoid function is unsuccessfully used for the multiclass classification of the brain tumour due to its limit of binary classification. This study aims to increase the classification accuracy by reducing the risk of overfitting problem and supports multi-class classification. The proposed system consists of a convolutional neural network with modified softmax loss function and regularization. Results: Classification accuracy for the different types of tumours and the processing time were calculated based on the probability score of the labeled data and their execution time. Different accuracy values and processing time were obtained when testing the proposed system using different samples of MRI images. The result shows that the proposed solution is better compared to the other systems. Besides, the proposed solution has higher accuracy by almost 2 % and less processing time of 40∼50 ms compared to other current solutions. Conclusion: The proposed system focused on classification accuracy of the different types of tumours from the 3D MRI images. This paper solves the issues of binary classification, the processing time, and the issues of overfitting of the data.

Original languageEnglish
Article number108520
JournalJournal of Neuroscience Methods
Volume330
Early online date11 Nov 2019
DOIs
Publication statusPublished - 15 Jan 2020

Fingerprint

Brain Neoplasms
Learning
Sigmoid Colon
Neoplasms

Cite this

@article{b4d7f88dc10a47ab874a81faf1a1a42e,
title = "A novel enhanced softmax loss function for brain tumour detection using deep learning",
abstract = "Background and Aim: In deep learning, the sigmoid function is unsuccessfully used for the multiclass classification of the brain tumour due to its limit of binary classification. This study aims to increase the classification accuracy by reducing the risk of overfitting problem and supports multi-class classification. The proposed system consists of a convolutional neural network with modified softmax loss function and regularization. Results: Classification accuracy for the different types of tumours and the processing time were calculated based on the probability score of the labeled data and their execution time. Different accuracy values and processing time were obtained when testing the proposed system using different samples of MRI images. The result shows that the proposed solution is better compared to the other systems. Besides, the proposed solution has higher accuracy by almost 2 {\%} and less processing time of 40∼50 ms compared to other current solutions. Conclusion: The proposed system focused on classification accuracy of the different types of tumours from the 3D MRI images. This paper solves the issues of binary classification, the processing time, and the issues of overfitting of the data.",
keywords = "Brain tumour detection, Deep learning, Loss function, Multiclass classification, Neural network, Softmax function",
author = "Sunil Maharjan and Abeer Alsadoon and Prasad, {P. W.C.} and Thair Al-Dalain and Alsadoon, {Omar Hisham}",
year = "2020",
month = "1",
day = "15",
doi = "10.1016/j.jneumeth.2019.108520",
language = "English",
volume = "330",
journal = "Journal of Neuroscience Methods",
issn = "0165-0270",
publisher = "Elsevier",

}

A novel enhanced softmax loss function for brain tumour detection using deep learning. / Maharjan, Sunil; Alsadoon, Abeer; Prasad, P. W.C.; Al-Dalain, Thair; Alsadoon, Omar Hisham.

In: Journal of Neuroscience Methods, Vol. 330, 108520, 15.01.2020.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A novel enhanced softmax loss function for brain tumour detection using deep learning

AU - Maharjan, Sunil

AU - Alsadoon, Abeer

AU - Prasad, P. W.C.

AU - Al-Dalain, Thair

AU - Alsadoon, Omar Hisham

PY - 2020/1/15

Y1 - 2020/1/15

N2 - Background and Aim: In deep learning, the sigmoid function is unsuccessfully used for the multiclass classification of the brain tumour due to its limit of binary classification. This study aims to increase the classification accuracy by reducing the risk of overfitting problem and supports multi-class classification. The proposed system consists of a convolutional neural network with modified softmax loss function and regularization. Results: Classification accuracy for the different types of tumours and the processing time were calculated based on the probability score of the labeled data and their execution time. Different accuracy values and processing time were obtained when testing the proposed system using different samples of MRI images. The result shows that the proposed solution is better compared to the other systems. Besides, the proposed solution has higher accuracy by almost 2 % and less processing time of 40∼50 ms compared to other current solutions. Conclusion: The proposed system focused on classification accuracy of the different types of tumours from the 3D MRI images. This paper solves the issues of binary classification, the processing time, and the issues of overfitting of the data.

AB - Background and Aim: In deep learning, the sigmoid function is unsuccessfully used for the multiclass classification of the brain tumour due to its limit of binary classification. This study aims to increase the classification accuracy by reducing the risk of overfitting problem and supports multi-class classification. The proposed system consists of a convolutional neural network with modified softmax loss function and regularization. Results: Classification accuracy for the different types of tumours and the processing time were calculated based on the probability score of the labeled data and their execution time. Different accuracy values and processing time were obtained when testing the proposed system using different samples of MRI images. The result shows that the proposed solution is better compared to the other systems. Besides, the proposed solution has higher accuracy by almost 2 % and less processing time of 40∼50 ms compared to other current solutions. Conclusion: The proposed system focused on classification accuracy of the different types of tumours from the 3D MRI images. This paper solves the issues of binary classification, the processing time, and the issues of overfitting of the data.

KW - Brain tumour detection

KW - Deep learning

KW - Loss function

KW - Multiclass classification

KW - Neural network

KW - Softmax function

UR - http://www.scopus.com/inward/record.url?scp=85075323399&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075323399&partnerID=8YFLogxK

U2 - 10.1016/j.jneumeth.2019.108520

DO - 10.1016/j.jneumeth.2019.108520

M3 - Article

C2 - 31734325

AN - SCOPUS:85075323399

VL - 330

JO - Journal of Neuroscience Methods

JF - Journal of Neuroscience Methods

SN - 0165-0270

M1 - 108520

ER -