Abstract

SummaryBreast cancer is a very hazardous disease that mainly affects women and leads to a high mortality rate. Early detection of this disease only can reduce the mortality rate. Previously, several methods were used to detect this cancer, but none of them provides sufficient accuracy. To deal this issue, a pulse coupled neural network optimized with chaotic grey wolf algorithm is proposed in this article for the classification of breast cancer using mammogram images. The breast cancer images are gathered from MAMMOSET dataset. The images are preprocessed and segmented with the help of Tsalli's entropy based multilevel 3D Otsu (TE‐3D‐Otsu) thresholding method. Then the parameters of the TE‐3D‐Otsu are optimized using the improved shuffled shepherd optimization algorithm. The features are extracted using moment invariant wavelet feature extraction technique. The images are classified using pulse coupled neural network (PCNN). Then, PCNN mass parameters are optimizes utilizing chaotic grey wolf optimization algorithm (chaotic‐GWOA) for categorizing the mammogram breast cancer imageries, such as malignant, benign, and normal. The proposed method is activated on MATLAB, its performance is analyzed with existing methods. The simulation results show that the proposed BC‐PCNN‐CGWOA attains the accuracy 57.86%, 85.94%, and 53.45% is greater than the existing methods.

Full Text
Published version (Free)

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 call