Improving Oral Cancer Detection Using Pretrained Model

Oral Cancer is a kind of cancer, if found early enough, has a great chance of survival. Artificial Intelligence techniques is helping in a great way by automating the cancer prediction at a faster pace and low cost. Deep learning algorithms are extremely helpful in automating the oral cancer detecti...

Full description

Saved in:
Bibliographic Details
Published in2022 IEEE 6th Conference on Information and Communication Technology (CICT) pp. 1 - 5
Main Authors Kavyashree, C, Vimala, H S, Shreyas, J
Format Conference Proceeding
LanguageEnglish
Published IEEE 18.11.2022
Subjects
Cancer detection
Computational modeling
Convolutional Neural Networks
Costs
Deep learning
Deep Learning methods
DenseNet models
Feature extraction
Oral Cancer
Prediction algorithms
Training
Online AccessGet full text

Cover

More Information
Summary:Oral Cancer is a kind of cancer, if found early enough, has a great chance of survival. Artificial Intelligence techniques is helping in a great way by automating the cancer prediction at a faster pace and low cost. Deep learning algorithms are extremely helpful in automating the oral cancer detection. These algorithms are helpful in extracting the features and classifying the images. The Convolutional Neural Network is used in this paper to compare with pretrained DenseNet201, DenseNet169 and DenseNet121 model. The results of the DenseNet201 has shown a good improvement in detection accuracy with 85% and significantly reduces the loss whereas the DenseNet169 provides excellent training accuracy of 98.96%. DenseNet201 also performs well at recall and F1-score 93.93% and 93.22%. The pretrained model also improves computational efficiency by using predefined weights.
DOI:10.1109/CICT56698.2022.9997897