Background:Artificial Intelligence (AI) is a term that implies using a computer to model intelligent behavior with minimal human intervention. Their potential to exploit the meaningful relationship with a dataset can be used in diagnosis, treatment, and outcome prediction in many clinical scenarios. The aim was to evaluate the use of artificial intelligence algorithms to successfully differentiate histopathologic images of grades of OSCC and healthy oral mucosa.
Material & Methods: In this cross-sectional study, Inception-ResNet-V2, a recently created artificial intelligence system, was used to analyze 844 pictures captured from the histopathological view of the connective tissues from three groups, low-grade OSCC, high-grade OSCC and normal mucosa.
Result:The results obtained from this research and comparable articles emphasize that deep learning-based systems have a high ability to analyze histopathological images and can be very useful and effective in cancer diagnosis and grading.According to the results of the ROC analysis from this research, Inception-ResNet-V2 has shown robust results in successfully differentiating Low-Grade OSCC, High-Grade OSCC and normal mucosa with over 95% accuracy.
Conclusion: According to the results of the present and previous studies, it can be concluded that CNN, and particularly Inception-ResNet-V2 have immense potential in analyzing histopathology pictures and could be very helpful for pathologists in cancer diagnosis.