Early Cervical Cancer Diagnostics Detection Based on RMSE and Extreme Learning Machine
Aditya Kumar Singh Pundir, Ullal Akshatha Nayak, D. Gayathri, M. Sucharitha, Bhavana Jamalpur, Ravi Rastogi · 2023
Among Indian women, cervical cancer continues to be the most prevalent form of the disease. The death and incidence rates from cervical cancer are dramatically higher in India than in any other country. Cervical cancer is unique among malignancies in that it can be prevented through screening by detecting and treating precancerous lesions at any point in the disease's lengthy natural history. Women can be screened for cervical precancers and malignancies using a number of different procedures, both older and more cutting-edge technologies. Given the limitations of any given screening tool, selecting the appropriate tool for a given situation is essential. Cervical precancers can be treated in a number of ways; the one chosen will rely on a number of factors, including the patient's preferences, the available budget, the quality of the biopsy specimens, etc. Cervical cancer screening guidelines for Indian women are examined. The proposed method consists of three main stages: preprocessing, feature extraction, and model training. In the preprocessing phase, image processing techniques like HE and sharpening and smoothing are applied. The Fourier Transform is used for feature extraction in GLCM. The RMSE-ELM technique is utilized for model evaluation. Better results are achieved with the proposed technique than with CNN or ELM.