Abstract
As breast cancer can be very aggressive, only early detection can prevent mortality. The proposed system is to eliminate the unnecessary waiting time as well as reducing human and technical errors in diagnosing breast cancer. The correct diagnosis of breast cancer is one of the major problems in the medical field. From the literature it has been found that different pattern recognition techniques can help them to improve in this domain. This paper uses the neural networks with an incremental learning algorithm as a tool to classify a mass in the breast (benign and malignant) using selection of the most relevant risk factors and decision making of the breast cancer diagnosis To test the proposed algorithm we used the Wisconsin Breast Cancer Database (WBCD). ANN with an incremental learning algorithm performance is tested using classification accuracy, sensitivity and specificity analysis, and confusion matrix. The obtained classification accuracy of 99.95%, a very promising result compared with previous algorithms already applied and recent classification techniques applied to the same database.
Highlights
Breast cancer is a leading fatality cancer for woman
Masses of 2 cm in diameter are palpable with regular breast self-examination while mammogram images can capture it from 5 mm in diameter
Our algorithm enabled us to achieve a better percentage compared to what existed in previous work with the same database (WBCD), mainly that of Marcano [5]
Summary
Breast cancer is a leading fatality cancer for woman. According to epidemiological data, breast cancer accounts for 20-25% of female malignant tumor, with is expected to increase. In our research process about the network structure that best fits our application, we use an incremental algorithm-based technique [12] performed in 4 steps: The first step consists in training a minimum network made up of a single neuron on its hidden layer. The variable selection technique adopted here consists, once the best architecture is determined, in removing au input variable each time proceeding to learn the network and evaluate the new performance. The weaker this performance is, compared to that of the starting network, the more relevant the variable in the chosen model is.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
