Date of Publication :7th February 2026

Abstract: Early detection of malignant tumors remains a significant challenge in healthcare, as conventional diagnostic procedures are often time-consuming and not uniformly accessible, particularly in resource-limited settings. To address this issue, this study proposes an automated cancer detection framework that leverages computational linguistics and deep learning techniques to analyze unstructured clinical narratives. Convolutional, recurrent, and transformer-based models were evaluated using standard medical datasets. Experimental results indicate that biomedical transformer models outperform traditional approaches, achieving an accuracy of 92.5% and an F1-score of 91.1%. The proposed system demonstrates scalability and has the potential to support clinicians by enabling faster and more informed diagnostic decision-making in clinical environments.

Reference :