Ensemble Based Deep Learning Model for Prediction of Oral Squamous Cell Carcinom

The oral squamous cell carcinoma (OSCC) is an extremely difficult and fatal form of cancer. Deep learning algorithms can discover tiny buccal tissue patterns in massive datasets, making them promising mouth cancer predictors. This research shows a deep learning network that uses groups to help predict mouth cancer. The suggested method uses several deep learning algorithms, including RNN, Bi-LSTM, and Bi-GRU, to identify features. Merging RNN and bi-GRU features concatenates bi-LSTM output. The ensemble model uses a vote classifier that has four standard classifiers built into it. The logistic regression (LR), random forest (RF), k-means clustering, and multi-SVNN classifiers are all part of this baseline. Amazingly, the suggested ensemble-based classification model achieves 98.34% accuracy on the first dataset and 98.76% accuracy on the second, thanks to its effective use of deep features. This shows how well it can predict outcomes and how promising it is for accurate oral cancer prediction. The combination of deep learning algorithms and classical classifiers improves oral cancer prediction models, according to this study.