Research Article
Transfer Learning with Pre-trained CNNs for Breast Cancer Stage Identification
@INPROCEEDINGS{10.1007/978-3-031-28725-1_8, author={Tesfahunegn Minwuyelet Mengistu and Birtukan Shegaw Arega and Birhanu Hailu Belay}, title={Transfer Learning with Pre-trained CNNs for Breast Cancer Stage Identification}, proceedings={Artificial Intelligence and Digitalization for Sustainable Development. 10th EAI International Conference, ICAST 2022, Bahir Dar, Ethiopia, November 4-6, 2022, Proceedings}, proceedings_a={ICAST}, year={2023}, month={3}, keywords={Breast cancer Pre-trained model CNN Segmentation Transfer learning}, doi={10.1007/978-3-031-28725-1_8} }- Tesfahunegn Minwuyelet Mengistu
Birtukan Shegaw Arega
Birhanu Hailu Belay
Year: 2023
Transfer Learning with Pre-trained CNNs for Breast Cancer Stage Identification
ICAST
Springer
DOI: 10.1007/978-3-031-28725-1_8
Abstract
Breast cancer stage identification is an important prerequisite for early treatment to increase the chance of survival, and predict the recurrence of cancer. Research works done so far were mainly focused on the classification of breast cancer types while many of them are neglecting to stage of breast cancer. Obtaining an adequate labeled breast cancer image dataset for training machine learning algorithms is challenging. In this paper, we propose a pre-trained Convolutional Neural Network (Pretrained-CNN) model for Breast Cancer Stage Identification. The proposed method is designed by leveraging transfer learning techniques. Further, the performance of the pre-trained model is compared with CNN-based models that are trained from scratch. The performance of the proposed model is tested using a publicly available breast cancer-image dataset taken and achieved a promising result with an overall classification accuracy of 90%
