Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network
One of the major limiting factors of the application of deep learning in automating malaria diagnosis is the insufficiency of labeled data for network training which leads to overfitting of the model on training data. Network regularization and data augmentation are two major techniques employed dur...
Saved in:
Main Authors: | , , , |
---|---|
Format: | Conference or Workshop Item |
Published: |
2022
|
Subjects: | |
Online Access: | http://eprints.utm.my/id/eprint/101443/ http://dx.doi.org/10.1109/IAICT55358.2022.9887391 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.utm.101443 |
---|---|
record_format |
eprints |
spelling |
my.utm.1014432023-06-14T10:27:04Z http://eprints.utm.my/id/eprint/101443/ Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network Abdullahi Muhammad, Fatima Sudirman, Rubita Ariffin, Ismail Ramli, Norhafizah TK Electrical engineering. Electronics Nuclear engineering One of the major limiting factors of the application of deep learning in automating malaria diagnosis is the insufficiency of labeled data for network training which leads to overfitting of the model on training data. Network regularization and data augmentation are two major techniques employed during network design and training to mitigate the effect of overfitting. To investigate and compare the effect of these two techniques, this study developed three deep learning models based on convolutional neural network which were trained on 27558 images of equal instances of malaria (P.Falciparum specie) infected and uninfected cells (NIH malaria dataset). The first model which consist of three convolution layers and one hidden layer achieved a validation accuracy of 95%. A second model was developed which employed network regularization, by training the model on fewer neurons to increase its robustness. Drop out regularization of 20%, 30% and 50% were embedded on successive convolution layers which resulted in a final validation accuracy of 95.1%. Data augmentation in the form of various image transformation was employed. This resulted in a validation accuracy of 93.7%. 2022 Conference or Workshop Item PeerReviewed Abdullahi Muhammad, Fatima and Sudirman, Rubita and Ariffin, Ismail and Ramli, Norhafizah (2022) Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network. In: 2022 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2022, 28 July 2022 - 30 July 2022, Bali, Indonesia. http://dx.doi.org/10.1109/IAICT55358.2022.9887391 |
institution |
Universiti Teknologi Malaysia |
building |
UTM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Teknologi Malaysia |
content_source |
UTM Institutional Repository |
url_provider |
http://eprints.utm.my/ |
topic |
TK Electrical engineering. Electronics Nuclear engineering |
spellingShingle |
TK Electrical engineering. Electronics Nuclear engineering Abdullahi Muhammad, Fatima Sudirman, Rubita Ariffin, Ismail Ramli, Norhafizah Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
description |
One of the major limiting factors of the application of deep learning in automating malaria diagnosis is the insufficiency of labeled data for network training which leads to overfitting of the model on training data. Network regularization and data augmentation are two major techniques employed during network design and training to mitigate the effect of overfitting. To investigate and compare the effect of these two techniques, this study developed three deep learning models based on convolutional neural network which were trained on 27558 images of equal instances of malaria (P.Falciparum specie) infected and uninfected cells (NIH malaria dataset). The first model which consist of three convolution layers and one hidden layer achieved a validation accuracy of 95%. A second model was developed which employed network regularization, by training the model on fewer neurons to increase its robustness. Drop out regularization of 20%, 30% and 50% were embedded on successive convolution layers which resulted in a final validation accuracy of 95.1%. Data augmentation in the form of various image transformation was employed. This resulted in a validation accuracy of 93.7%. |
format |
Conference or Workshop Item |
author |
Abdullahi Muhammad, Fatima Sudirman, Rubita Ariffin, Ismail Ramli, Norhafizah |
author_facet |
Abdullahi Muhammad, Fatima Sudirman, Rubita Ariffin, Ismail Ramli, Norhafizah |
author_sort |
Abdullahi Muhammad, Fatima |
title |
Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
title_short |
Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
title_full |
Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
title_fullStr |
Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
title_full_unstemmed |
Comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
title_sort |
comparison of regularization and data augmentation as means of improving malaria classification using convolutional neural network |
publishDate |
2022 |
url |
http://eprints.utm.my/id/eprint/101443/ http://dx.doi.org/10.1109/IAICT55358.2022.9887391 |
_version_ |
1769842055870152704 |
score |
13.18916 |