DOIONLINE

DOIONLINE NO - IJASEAT-IRAJ-DOIONLINE-17392

Publish In
International Journal of Advances in Science, Engineering and Technology(IJASEAT)-IJASEAT
Journal Home
Volume Issue
Issue
Volume-8,Issue-3  ( Jul, 2020 )
Paper Title
Breast Cancer Prediction using Different Machine Learning Algorithms
Author Name
Abhijeet Ahiwale, Prajwal Choudhary, Omkar Gujar, Satej Todkar, Shital Bachpalle
Affilition
Department of Information Technology, Zeal College of Engineering & Research, Pune, India Professor, Department of Information Technology, Zeal College of Enginerring & Research, Pune, India
Pages
15-17
Abstract
According to a report, 8% of world „s women population suffers from breast cancer, which ranks second in the number of deaths of patients after Lung cancer. The condition is same for both developed and undeveloped nations. The symptoms of breast cancer are changes in the size of breasts, redness, constant pain, change of genes, skin texture of breasts. Breast cancer can be identified by biopsy in which a tissue of the affected region is removed and examined by a microscope . The condition is identified by histopathology which looks for irregular cells. This traditional method for the identification of breast cancer has some flaws. If the histopathologist is not well qualified and lacks experience in the field then this can lead to wrong analysis. Due to advancement in Machine Learning and Image processing, there have been several attempts to develop a system which can identify the pattern in histology images to get results which can be more reliable than the traditional methods of analysis. Here we have tried to study two different machine learning approaches and find the best algorithm which can be used to classify the breast cancer histology images into benign and malignant classes and then further subdivide these two classes in subclasses. In the first approach, the support vector machine is used to train the model while the second approach uses Convolutional neural networks. To enhance the accuracy of the convolutional neural network, we have augmentation techniques for testing the dataset. The result shows that Convolutional neural networks outperformed the other approach. Keywords - Image Processing, Images, Convolutional Neural Networks, Computer-Aided Diagnosis, Locality Constrained Linear Coding, Engineered Features, Histopathology, Images.
  View Paper