Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images

For both women and men over 60, liver cancer is the primary cause of cancer-related deaths. To help physicians to diagnose patients more accurately, computer-assisted imaging techniques have become increasingly important in recent years. Recent, deep Convolutional Neural Network (CNN) research has p...

Full description

Saved in:
Bibliographic Details
Main Authors: Bilga Jacob, R. S. Vinod Kumar, S. S. Kumar
Format: Article
Language:English
Published: Bulgarian Academy of Sciences 2024-09-01
Series:International Journal Bioautomation
Subjects:
Online Access:http://www.biomed.bas.bg/bioautomation/2024/vol_28.3/files/28.3_03.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1841549886358552576
author Bilga Jacob
R. S. Vinod Kumar
S. S. Kumar
author_facet Bilga Jacob
R. S. Vinod Kumar
S. S. Kumar
author_sort Bilga Jacob
collection DOAJ
description For both women and men over 60, liver cancer is the primary cause of cancer-related deaths. To help physicians to diagnose patients more accurately, computer-assisted imaging techniques have become increasingly important in recent years. Recent, deep Convolutional Neural Network (CNN) research has produced amazing improvements in image segmentation and classification. The same issue of diagnosing liver nodules in computed tomography (CT) scans is addressed in this research by introducing a novel Computer-Aided Detection (CAD) system that makes use of an Efficient Network (EfficientNet) image classification algorithm. Unlike CNN, which adjusts its network parameters arbitrarily, a set of predetermined scaling coefficients is used in the EfficientNet scaling technique to reliably scale the network’s breadth, depth, and resolution. Here the EfficientNet models are assessed by varying the input dimensions of the CT scans from The Liver Tumor Segmentation (LiTs) dataset. Finally, the performance evaluation shows that the input dimension 224x224 effectively classified the images and is superior to the other models evaluated with 0.991 AUC and 99.37% F1-Score, precision 99.44%, recall 99.30%, specificity 99.43%, and accuracy 99.36% for Kaggle datasets.
format Article
id doaj-art-b2b64bf8c9d0472288800ea7be217cb3
institution Kabale University
issn 1314-1902
1314-2321
language English
publishDate 2024-09-01
publisher Bulgarian Academy of Sciences
record_format Article
series International Journal Bioautomation
spelling doaj-art-b2b64bf8c9d0472288800ea7be217cb32025-01-10T12:57:08ZengBulgarian Academy of SciencesInternational Journal Bioautomation1314-19021314-23212024-09-0128315116010.7546/ijba.2024.28.3.001001Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT ImagesBilga Jacob0R. S. Vinod KumarS. S. KumarDepartment of Electronics and Communication Engineering, Noorul Islam Center for Higher Education, Kumarakoil, Kanyakumari, Tamil Nadu, IndiaFor both women and men over 60, liver cancer is the primary cause of cancer-related deaths. To help physicians to diagnose patients more accurately, computer-assisted imaging techniques have become increasingly important in recent years. Recent, deep Convolutional Neural Network (CNN) research has produced amazing improvements in image segmentation and classification. The same issue of diagnosing liver nodules in computed tomography (CT) scans is addressed in this research by introducing a novel Computer-Aided Detection (CAD) system that makes use of an Efficient Network (EfficientNet) image classification algorithm. Unlike CNN, which adjusts its network parameters arbitrarily, a set of predetermined scaling coefficients is used in the EfficientNet scaling technique to reliably scale the network’s breadth, depth, and resolution. Here the EfficientNet models are assessed by varying the input dimensions of the CT scans from The Liver Tumor Segmentation (LiTs) dataset. Finally, the performance evaluation shows that the input dimension 224x224 effectively classified the images and is superior to the other models evaluated with 0.991 AUC and 99.37% F1-Score, precision 99.44%, recall 99.30%, specificity 99.43%, and accuracy 99.36% for Kaggle datasets.http://www.biomed.bas.bg/bioautomation/2024/vol_28.3/files/28.3_03.pdfdeep learningconvolution neural networkcomputed tomographyliver tumor classificationthe liver tumor segmentation
spellingShingle Bilga Jacob
R. S. Vinod Kumar
S. S. Kumar
Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images
International Journal Bioautomation
deep learning
convolution neural network
computed tomography
liver tumor classification
the liver tumor segmentation
title Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images
title_full Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images
title_fullStr Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images
title_full_unstemmed Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images
title_short Multiple Scaling Based EfficientNet Modelling for Liver Tumor Classification on CT Images
title_sort multiple scaling based efficientnet modelling for liver tumor classification on ct images
topic deep learning
convolution neural network
computed tomography
liver tumor classification
the liver tumor segmentation
url http://www.biomed.bas.bg/bioautomation/2024/vol_28.3/files/28.3_03.pdf
work_keys_str_mv AT bilgajacob multiplescalingbasedefficientnetmodellingforlivertumorclassificationonctimages
AT rsvinodkumar multiplescalingbasedefficientnetmodellingforlivertumorclassificationonctimages
AT sskumar multiplescalingbasedefficientnetmodellingforlivertumorclassificationonctimages