Ir papnet and fine grained classifier based cervical cancer classification via pap smear images

Abstract

The early recognition and therapy of cervical cancer are imperative newlinedue to its elevated death rate. Human Papillomavirus (HPV) stands out as the newlineprincipal cause of Cervical Cancer (CC) risk in women, initiating infections newlinethat spread across the cervix. The inability of CC patients to self-examine newlineimpedes early recognition, with symptoms only manifesting in the terminal newlinestage. This global concern significantly impacts the lives of women, newlineparticularly in economically disadvantaged nations. In this context, the newlinesusceptibility of women in this country is particularly pronounced. The World newlineHealth Organization (WHO) evaluates that there are between 2 and 2.5 million newlinecancer patients in India. with an annual expectation of detecting over seven newlinelakh new cases. Automated systems for the identification and categorization newlineof CC play a vital function in aiding physicians with early detection. Clinical newlinedecision support systems, particularly those utilizing Machine Learning (ML) newlineapproaches, rely heavily on accurately recognizing CC at its early stages. newlineWithin the framework of this thesis, using pap smear images, new techniques newlinefor identifying CC have been created. These innovative approaches newlineunderscore the significance of leveraging automated detection systems to newlineimprove CC early detection and make a positive impact on healthcare newlinepractices. newline