AN L2-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images

Grace Billingsley; Julia Dietlmeier; Vivek Narayanaswamy; Andreas Spanias; Noel E. O'Connor

doi:10.1109/ICIP49359.2023.10222887

AN L₂-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images

Grace Billingsley, Julia Dietlmeier, Vivek Narayanaswamy, Andreas Spanias, Noel E. O'Connor

Engineering, Ira A. Fulton Schools of (IAFSE)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

We propose an accurate and fast classification network for classification of brain tumors in MRI images that outperforms all lightweight methods investigated in terms of accuracy. We test our model on a challenging 2D T1-weighted CE-MRI dataset containing three types of brain tumors: Meningioma, Glioma and Pituitary. We introduce an l₂-normalized spatial attention mechanism that acts as a regularizer against overfitting during training. We compare our results against the state-of-the-art on this dataset and show that by integrating l₂-normalized spatial attention into a baseline network we achieve a performance gain of 1.79 percentage points. Even better accuracy can be attained by combining our model in an ensemble with the pretrained VGG16 at the expense of execution speed. Our code is publicly available at https://github.com/juliadietlmeier/MRI-image-classification.

Original language	English (US)
Title of host publication	2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings
Publisher	IEEE Computer Society
Pages	1895-1899
Number of pages	5
ISBN (Electronic)	9781728198354
DOIs	https://doi.org/10.1109/ICIP49359.2023.10222887
State	Published - 2023
Event	30th IEEE International Conference on Image Processing, ICIP 2023 - Kuala Lumpur, Malaysia Duration: Oct 8 2023 → Oct 11 2023

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
ISSN (Print)	1522-4880

Conference

Conference	30th IEEE International Conference on Image Processing, ICIP 2023
Country/Territory	Malaysia
City	Kuala Lumpur
Period	10/8/23 → 10/11/23

Keywords

attention mechanisms
brain tumor classification
Deep learning
MRI image analysis

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition
Signal Processing

Access to Document

10.1109/ICIP49359.2023.10222887

Cite this

Billingsley, G., Dietlmeier, J., Narayanaswamy, V., Spanias, A., & O'Connor, N. E. (2023). AN L₂-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images. In 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings (pp. 1895-1899). (Proceedings - International Conference on Image Processing, ICIP). IEEE Computer Society. https://doi.org/10.1109/ICIP49359.2023.10222887

AN L₂-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images. / Billingsley, Grace; Dietlmeier, Julia; Narayanaswamy, Vivek et al.
2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings. IEEE Computer Society, 2023. p. 1895-1899 (Proceedings - International Conference on Image Processing, ICIP).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Billingsley, G, Dietlmeier, J, Narayanaswamy, V, Spanias, A & O'Connor, NE 2023, AN L₂-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images. in 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings. Proceedings - International Conference on Image Processing, ICIP, IEEE Computer Society, pp. 1895-1899, 30th IEEE International Conference on Image Processing, ICIP 2023, Kuala Lumpur, Malaysia, 10/8/23. https://doi.org/10.1109/ICIP49359.2023.10222887

Billingsley G, Dietlmeier J, Narayanaswamy V, Spanias A, O'Connor NE. AN L₂-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images. In 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings. IEEE Computer Society. 2023. p. 1895-1899. (Proceedings - International Conference on Image Processing, ICIP). doi: 10.1109/ICIP49359.2023.10222887

Billingsley, Grace ; Dietlmeier, Julia ; Narayanaswamy, Vivek et al. / AN L₂-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images. 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings. IEEE Computer Society, 2023. pp. 1895-1899 (Proceedings - International Conference on Image Processing, ICIP).

@inproceedings{a3d1120202f744f9bd1e1e97e7e7acc4,

title = "AN L2-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images",

abstract = "We propose an accurate and fast classification network for classification of brain tumors in MRI images that outperforms all lightweight methods investigated in terms of accuracy. We test our model on a challenging 2D T1-weighted CE-MRI dataset containing three types of brain tumors: Meningioma, Glioma and Pituitary. We introduce an l2-normalized spatial attention mechanism that acts as a regularizer against overfitting during training. We compare our results against the state-of-the-art on this dataset and show that by integrating l2-normalized spatial attention into a baseline network we achieve a performance gain of 1.79 percentage points. Even better accuracy can be attained by combining our model in an ensemble with the pretrained VGG16 at the expense of execution speed. Our code is publicly available at https://github.com/juliadietlmeier/MRI-image-classification.",

keywords = "attention mechanisms, brain tumor classification, Deep learning, MRI image analysis",

author = "Grace Billingsley and Julia Dietlmeier and Vivek Narayanaswamy and Andreas Spanias and O'Connor, {Noel E.}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 30th IEEE International Conference on Image Processing, ICIP 2023 ; Conference date: 08-10-2023 Through 11-10-2023",

year = "2023",

doi = "10.1109/ICIP49359.2023.10222887",

language = "English (US)",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "1895--1899",

booktitle = "2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings",

}

TY - GEN

T1 - AN L2-Normalized Spatial Attention Network for Accurate and Fast Classification of Brain Tumors in 2D T1-Weighted CE-MRI Images

AU - Billingsley, Grace

AU - Dietlmeier, Julia

AU - Narayanaswamy, Vivek

AU - Spanias, Andreas

AU - O'Connor, Noel E.

PY - 2023

Y1 - 2023

N2 - We propose an accurate and fast classification network for classification of brain tumors in MRI images that outperforms all lightweight methods investigated in terms of accuracy. We test our model on a challenging 2D T1-weighted CE-MRI dataset containing three types of brain tumors: Meningioma, Glioma and Pituitary. We introduce an l2-normalized spatial attention mechanism that acts as a regularizer against overfitting during training. We compare our results against the state-of-the-art on this dataset and show that by integrating l2-normalized spatial attention into a baseline network we achieve a performance gain of 1.79 percentage points. Even better accuracy can be attained by combining our model in an ensemble with the pretrained VGG16 at the expense of execution speed. Our code is publicly available at https://github.com/juliadietlmeier/MRI-image-classification.

AB - We propose an accurate and fast classification network for classification of brain tumors in MRI images that outperforms all lightweight methods investigated in terms of accuracy. We test our model on a challenging 2D T1-weighted CE-MRI dataset containing three types of brain tumors: Meningioma, Glioma and Pituitary. We introduce an l2-normalized spatial attention mechanism that acts as a regularizer against overfitting during training. We compare our results against the state-of-the-art on this dataset and show that by integrating l2-normalized spatial attention into a baseline network we achieve a performance gain of 1.79 percentage points. Even better accuracy can be attained by combining our model in an ensemble with the pretrained VGG16 at the expense of execution speed. Our code is publicly available at https://github.com/juliadietlmeier/MRI-image-classification.

KW - attention mechanisms

KW - brain tumor classification

KW - Deep learning

KW - MRI image analysis

UR - http://www.scopus.com/inward/record.url?scp=85180750194&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85180750194&partnerID=8YFLogxK

U2 - 10.1109/ICIP49359.2023.10222887

DO - 10.1109/ICIP49359.2023.10222887

M3 - Conference contribution

AN - SCOPUS:85180750194

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1895

EP - 1899

BT - 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings

PB - IEEE Computer Society

T2 - 30th IEEE International Conference on Image Processing, ICIP 2023

Y2 - 8 October 2023 through 11 October 2023

ER -