Enhancing cervical precancerous lesion detection using African Vulture Optimization Algorithm with Deep Learning model

作者：Jiayu Song, Le Wang, Jiazhuo Yan, Yue Feng, Yunyan Zhang · 发表于：Biomedical Signal Processing and Control · 年份：2024 · DOI：10.1016/j.bspc.2024.106665 · 被引用次数：9 · 研究领域：AI in cancer detection、Medical Imaging and Analysis

• Cervical cancer is a leading cause of death, especially in emerging countries like India. • Deep learning techniques have the potential to improve screening and treatment for cervical cancer. • The proposed AVOADL-CCD technique utilizes advanced filtering and CapsNet for better feature extraction. • Experimental results demonstrate significant improvements in detection accuracy using the AVOADL-CCD algorithm. Cervical cancer (CC) is a major cause of mortality among women, mainly in emerging countries, including India. Recent technological advancements have the potential to enable quick, cost-effective, and more sensitive screening and treatment measures for CC. Consequently, deep learning (DL)-based approaches have gained importance in classifying CC patients into distinct risk groups. This study proposes an African Vulture Optimization Algorithm with Deep Learning assisted Cervical Cancer Detection (AVOADL-CCD) technique. The AVOADL-CCD technique follows Wiener filtering for image pre-processing. Besides, the AVOADL-CCD technique applies Capsule Networks (CapsNet) model for deriving feature vectors. The AVOA can be applied to the CapsNet model for the hyperparameter tuning process. For the automated recognition and classification of CC, an adaptive neuro-fuzzy inference system (ANFIS) model can be exploited. The experimental outcomes of the AVOADL-CCD method take place on the Herlev dataset. The simulation values show the enhanced detection results of the AVOADL-CCD algori...