An Approach for Breast Cancer Diagnosis Classification Using Neural Network

Abstract

Artificial neural network has been widely used in various fields as an intelligent tool in recent years, such as artificial intelligence, pattern recognition, medical diagnosis, machine learning and so on. The classification of breast cancer is a medical application that poses a great challenge for researchers and scientists. Recently, the neural network has become a popular tool in the classification of cancer datasets. Classification is one of the most active research and application areas of neural networks. Major disadvantages of artificial neural network (ANN) classifier are due to its sluggish convergence and always being trapped at the local minima. To overcome this problem, differential evolution algorithm (DE) has been used to determine optimal value or near optimal value for ANN parameters. DE has been applied successfully to improve ANN learning from previous studies. However, there are still some issues on DE approach such as longer training time and lower classification accuracy. To overcome these problems, island based model has been proposed in this system. The aim of our study is to propose an approach for breast cancer distinguishing between different classes of breast cancer. This approach is based on the Wisconsin Diagnostic and Prognostic Breast Cancer and the classification of different types of breast cancer datasets. The proposed system implements the island-based training method to be better accuracy and less training time by using and analysing between two different migration topologies.

Breast Cancer Classification Utilizing Deep Learning Techniques on Medical Images: A Comprehensive Review

“…Another study used the WDBC datasets to train an artificial neural network to categorize BC cases as benign or malignant with the express intention of identifying malignancy. To cut down on training time and increase accuracy, the scientists suggested an island-based training approach [22]. These studies underscore the importance of classification in breast cancer diagnosis, as it can facilitate early detection, inform treatment decisions, and ultimately improve patient outcomes.…”

Enhancing Breast Cancer Detection with Ensemble Machine Learning Models

“…It identified the best model from the decision tree purposed and discussed its utility in the clinical setup and for the betterment of automated medical systems. Further, in [32], a methodology based on the neural network of diagnosing breast cancer is proposed.…”

Identification of Breast Tumor Using Hybrid Approach of Independent Component Analysis and Deep Neural Network

“…The sluggish convergence and constant being caught at the local minima are two major drawbacks of the artificial neural network (ANN) classifier. To solve this challenge, the differential evolution algorithm (DE) was employed by authors [6] to find the best or near-best ANN parameter values. DE can significantly increase ANN learning.…”

“…The migration size, which shows the number of individuals travelling and governs the quantitative aspect of migration, and the migration interval, which denotes the frequency of migration, are the two essential and most sensitive elements of the island model approach. Using the WBC dataset, the authors proposed and tested an island-based training model with an ANN, in which they found 99.97% classification accuracy [6]. The author of [7], proposed fusion at classification level between MLP and Gradient based classifiers to get the most suitable multiclassifier approach for each data set like WBC, WDBC and WPBC.…”

“…The author of [7], proposed fusion at classification level between MLP and Gradient based classifiers to get the most suitable multiclassifier approach for each data set like WBC, WDBC and WPBC. On a WBC dataset, the authors of [6] employed a mixture of MLP and the feature reduction technique PCA. Results showed that Learning Machine Neural Networks (LM ANN) has better generalization classifier model than BP ANN.…”