Communication science et technologie
Volume 23, Numéro 2, Pages 14-26
2025-11-18

Oral Cancer Detection Using Deep Learning: Feature Extraction And Activation Function Analysis



Authors : Rouabhi Sarra . Hakiki Racha Ikram . Tlemsani Redouane .

Abstract

The classification of oral diseases, particularly distinguishing between cancerous and non-cancerous lesions, is a critical task for early diagnosis and treatment planning. In this study, we propose a deep learning-based approach that utilizes Inception ResNet V2 for feature extraction and a simple dense neural network for classification. To optimize model performance, we experimented with different activation functions in the dense layers, specifically ReLU, Sigmoid, and Softmax. Our results demonstrate that this configuration significantly improves classification accuracy, achieving a test accuracy of 92.11%. The separation of feature extraction and classification enhances interpretability and allows for better optimization of network parameters. This study highlights the potential of deep learning in improving the accuracy and robustness of AI-assisted oral disease diagnosis.

Keywords

Classification of oral lesions, Feature extraction, Activation Function, Deep Learning.

Les articles similaires

De La Littérature Orale à La Didactique De L’oral. Le Conte Et L’apprentissage De La Production Orale. (cas De La 5ème Année Primaire) From Oral Literature To The Teaching Of Oral Communication: Folktales And Learning Oral Production. (case Study: 5th Year Of Primary School)

Brahmi Sara . 
pages 96-110.

Predictive Modeling Of Ozone Dosing In Drinking Water Treatment Plant Using Deep Learning Comparative Study Between Deep Neural Networks And Convolutional Neural Networks

Hellal Aouatef .  Djeddou Messaoud .  Loukam Imed .  A. Hameed Ibrahim .  Al Dallal Jehad .  Shawaqfah Moayyad . 
pages 69-83.

Object Multi‐detection And Multi‐tracking Via Machine Learning And Deep Learning

Merouane Labeni .  Chaouki Boufenar .  Mokhtar Taffar . 
pages 80-88.

Prediction And Feature Analysis Of Self-compacting Concrete Strength Using Machine Learning With Recycled Coarse Aggregate And Supplementary Cementitious Materials

Pulkit Kavendra .  Nigam Shivi . 
pages 513-534.

الألكستيميا لدى النساء المبتورات الثدي من جراء الإصابة بالسرطان دراسة ميدانية بمصلحة الأورام السرطانية مستشفى معلم لحسن-الغزوات-the Alkemstemia In Women With Breast Cancer Due To Cancer A Field Study At The Department Of Cancer Oncology Moualem Lahcen Hospital - Ghazaouet -

مسعودي محمد رضا .  دراز خليدة . 
ص 449-480.

Formats de citation