Cervical Cancer Risk Prediction Using One-Dimensional Convolutional Neural Network Enhanced by Snake Optimization Algorithm

Abstract

Cancer of the cervix is one of the most prevalent malignances in the global, hence early diagnosis is vital in the healthcare delivery systems. Growth in medical imaging as well as enhanced computational methods has now opened up possibilities for efficient and accurate diagnosis apart from other clinical investigations such as the Pap smear. Here, we introduced CNN-SOA, a deep learning-based method using One-Dimensional Convolutional Neural Network (1D-CNN) supported by the Snake Optimization Algorithm to estimate cervical cancer risk. The experiments presented show that the proposed method greatly improves prediction performance and stability, which confirms its suitability for medical classification problems. The results indicate that CNN-SOA can be generalized to other healthcare domains, which will opportunity development of precise and earlier diagnosing technologies in AI-based medical diagnosis.