A VGG16 feature-based transfer learning evaluation for the diagnosis of Oral Squamous Cell Carcinoma (OSCC)

• Main Authors: Arzmi, Mohd Hafiz, P.P. Abdul Majeed, Anwar, Musa, Rabiu Muazu, Mohd Razman, Mohd Azraai, Gan, Hong-Seng, Mohd Khairuddin, Ismail, Ab. Nasir, Ahmad Fakhri
• Format: Book Chapter
• Language: English
• Published: Springer 2023
• Subjects: R Medicine (General); RK Dentistry
• Online Access: http://irep.iium.edu.my/103896/2/103896_A%20VGG16%20feature-based%20transfer%20learning%20evaluation%20for%20the%20diagnosis.pdf; http://irep.iium.edu.my/103896/; https://link.springer.com/chapter/10.1007/978-981-19-8937-7_2

Oral Squamous Cell Carcinoma (OSCC) is the most prevalent type of oral cancer. Early detection of such cancer could increase a patient’s survival rate by 83%. This chapter shall explore the use of a feature-based transfer learning model, i.e., VGG16 coupled with different types of conventional machine learning models, viz. Support Vector Machine (SVM), Random Forest as well as k-Nearest Neighbour (kNN) as a means to identify OSCC. A total of 990 evenly distributed normal and OSCC histopathological images are split into the 60:20:20 ratio for training, testing and validation, respectively. A testing accuracy of 93% was recorded via the VGG16- RF pipeline from the study. Consequently, the proposed architecture is suitable to be deployed as artificial intelligence-driven computer-aided diagnostics and, in turn, facilitate clinicians for the identification of OSCC.