This study presents a deep learning model based on EfficientNetB3 for the detection of Oral Squamous Cell Carcinoma (OSCC) using histopathological images. The model was trained on a dataset of 1,224 images from 230 patients, captured at varying magnifications. The model achieved a high accuracy of 99.13% on the test dataset, with precision, recall, and F1-score all exceeding 99%. The model's ability to distinguish between normal epithelium and OSCC tissues demonstrates its potential as a reliable diagnostic tool for OSCC. The study highlights the importance of deep learning in improving the accuracy and efficiency of OSCC diagnosis, particularly in the absence of precise diagnostic tools. The model's performance was validated using a publicly available dataset, showcasing its clinical applicability. The research also addresses the challenges of OSCC diagnosis, including the complexity of histopathological analysis and the need for accurate and efficient diagnostic methods. The study emphasizes the significance of automated detection in medical imaging, particularly in the context of OSCC. The model's high accuracy and performance metrics suggest that it could serve as a valuable tool in clinical settings for the early and accurate detection of OSCC. The study also acknowledges the limitations and challenges of the model, including potential biases and the need for further research to enhance its generalization capabilities. Future research directions include expanding the dataset to include diverse populations and oral cancer subtypes, as well as improving the model's interpretability and clinical integration. The study concludes that the proposed model represents a significant advancement in the field of OSCC detection, offering a reliable and efficient solution for automated diagnosis.This study presents a deep learning model based on EfficientNetB3 for the detection of Oral Squamous Cell Carcinoma (OSCC) using histopathological images. The model was trained on a dataset of 1,224 images from 230 patients, captured at varying magnifications. The model achieved a high accuracy of 99.13% on the test dataset, with precision, recall, and F1-score all exceeding 99%. The model's ability to distinguish between normal epithelium and OSCC tissues demonstrates its potential as a reliable diagnostic tool for OSCC. The study highlights the importance of deep learning in improving the accuracy and efficiency of OSCC diagnosis, particularly in the absence of precise diagnostic tools. The model's performance was validated using a publicly available dataset, showcasing its clinical applicability. The research also addresses the challenges of OSCC diagnosis, including the complexity of histopathological analysis and the need for accurate and efficient diagnostic methods. The study emphasizes the significance of automated detection in medical imaging, particularly in the context of OSCC. The model's high accuracy and performance metrics suggest that it could serve as a valuable tool in clinical settings for the early and accurate detection of OSCC. The study also acknowledges the limitations and challenges of the model, including potential biases and the need for further research to enhance its generalization capabilities. Future research directions include expanding the dataset to include diverse populations and oral cancer subtypes, as well as improving the model's interpretability and clinical integration. The study concludes that the proposed model represents a significant advancement in the field of OSCC detection, offering a reliable and efficient solution for automated diagnosis.