Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195258,
        author = {Sumaiya Abul Hasan and Ms. Shaik Asha},
        title = {CLASSIFICATION OF ORAL CANCER INTO PRE-CANCEROUS STAGES USING MOBILENETV2},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {243-250},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195258},
        abstract = {Almost 10 million deaths from cancer were recorded in 2020, making it one of the world's major causes of death. Oral cancer is the sixth most common kind globally. Its lethality is mostly ascribed to late-stage diagnoses, when therapy becomes more difficult. However, mortality rates can be considerably decreased by early identification, especially in precancerous phases. Enhancing survival rates requires early screening and treatment, underscoring the necessity of effective diagnostic techniques. This study proposes a way to identify oral cavity lesions in their pre-cancerous phases and differentiate between benign and malignant lesions. To extract colour and texture-based information from oral cavity images which are essential for recognizing different lesion stages this method investigates five distinct colour spaces. Using MobileNetV2 for enhanced speed and accuracy, the suggested approach combines deep learning classification with handcrafted feature extraction, making it unique. The model provides an effective tool for detecting oral cancer by extracting intricate colour and texture patterns from the photos, surpassing conventional techniques in terms of computational efficiency and time. The model is a great option for affordable, mobile-based diagnostic tools because of its capacity to operate with constrained resources. The technique successfully distinguishes between benign, malignant, and precancerous lesions, showing encouraging results in both binary and multi-class classifications. This method could significantly improve the early identification of oral cancer, particularly in areas with limited access to cutting-edge healthcare facilities.},
        keywords = {},
        month = {April},
        }