ROI Extraction and Nuclei Classification of Pap Smear Images for Cervical Cancer Detection

Automated nuclei classification methods are increasingly important in improving the accuracy and efficiency of cervical cancer screening. These methods utilize artificial intelligence algorithms such as machine learning and deep learning to identify and classify cellular nuclei from Pap smear images. However, the implementation of these methods presents various challenges such as the need for high-quality images, manual annotation of training datasets, and the management of changes in lighting conditions. This paper proposes a novel automated nuclear classification method that employs an enhanced Alexnet model to overcome the aforementioned challenges. In addition, this paper presents a Region of Interest (ROI) Extraction method to improve the accuracy of Pap smear image classification and reduce the computational burden of the proposed AlexNet model. The proposed system is evaluated using a publicly available dataset, and the results indicate a classification accuracy of 95.78%, precision of 94.86%, recall of 96.86%, F1-score of 95.82%, and specificity of 94.82%. These outcomes suggest that the system can effectively automate the process of nuclei segmentation and accurately identify and classify cellular nuclei from Pap smear images with minimal human intervention. The findings demonstrate that the proposed system is an effective approach to automated nuclear classification.