SSDA_AOA: Stacked Sparse Denoising Autoencoder With Archimedes Optimization Algorithm Based Oral Cancer Detection on...

Area of Science:

• Oncology
• Computer Science
• Medical Imaging

Background:

• Oral cancer diagnosis presents significant challenges, impacting patient outcomes.
• Early detection is crucial for effective treatment and improved survival rates.
• Current diagnostic tools require enhancement for greater accuracy and efficiency.

Purpose of the Study:

• To develop a novel computer-aided diagnostic (CAD) framework for oral cancer.
• To leverage deep learning (DL) and optimization techniques for enhanced early detection.
• To improve diagnostic accuracy and patient outcomes in oral cancer cases.

Main Methods:

• Histopathological images were preprocessed and analyzed using the NASNet-Large model for feature extraction.
• The Archimedes Optimization Algorithm (AOA) performed simultaneous dimensionality reduction and hyperparameter tuning.
• A Stacked Sparse Denoising Autoencoder (SSDA) classifier was employed for robust latent representation learning.

Main Results:

• The proposed SSDA-AOA framework achieved high performance metrics.
• Accuracy: 95.38%
• Precision: 95.15%, Sensitivity: 91.78%, Specificity: 91.85%, F1-Score: 93.72%.

Conclusions:

• The SSDA-AOA framework shows significant potential for early and precise oral cancer detection.
• This approach can facilitate timely intervention, leading to improved patient outcomes.
• The study highlights the need for advanced computational tools in oncology diagnostics.