Automated classification of Pap smear images to detect cervical dysplasia

• An automated Pap smear classifier is proposed to detect cervical dysplasia.• Study performed on both cell as well as smear level.• 1610 single cell and 1320 complete indigenous samples were collected to perform this study.• Analysis is being performed on Shape, Texture and Colour features which includes 121 total numbers of features.• An ensemble classifier is designed using Least Square Support Vector Machine, Multilayer Perceptron's and Random Forest using weighted majority voting.• Samples are divided into two and three classes which will reflect the established Bethesda patho logical classification of cervical cancer.

Source: Computer Methods and Programs in Biomedicine - October 18, 2016 Category: Bioinformatics Authors: Kangkana Bora, Manish Chowdhury, Lipi B Mahanta, Malay Kumar Kundu, Anup Kumar Das Source Type: research