Image analysis with machine learning algorithms to assist breast cancer treatment

No Thumbnail Available
Authors
Asaduzzaman, Abu
Sibai, Fadi N.
Kanaya, Shigehiko
Altaf-Ul-Amin, Md
Jashim Uddin, Md
Chidella, Kishore K.
Mitra, Parthib
Issue Date
2021-06-06
Type
Book chapter
Language
en_US
Keywords
Biomedical engineering , Feature extraction , Image processing , Machine learning , Mammography , Surgical procedure
Research Projects
Organizational Units
Journal Issue
Alternative Title
Abstract

Real-time imaging technology has the potential to be applied to many complex surgical procedures such as those used in treating people with breast cancer. Key delaying factors for the successful development of real-time surgical imaging solutions include long execution time due to poor medical infrastructure and inaccuracy in processing mammogram images. In this work, we introduce a novel imaging technique that identifies malignant cells and supports breast cancer surgical procedures by analyzing mammograms in real-time with excellent accuracy. According to this method, hidden attributes of a target breast image are extracted and the extracted pixel values are analyzed using machine learning (ML) tools to determine if there are malignant cells. A malignant image is divided into contours and the rate of change in pixel value is calculated to pinpoint the regions of interest (ROIs) for a surgical procedure. Experimental results using 1500 known mammograms show that the imaging mechanism has the potential to identify benign and malignant cells with more than 99% accuracy. Experimental results also show that the rate of change in pixel values can be used to determine the ROIs with more than 98% accuracy.

Description
Click on the DOI link to access this book chapter at the publishers website (may not be free).
Citation
Asaduzzaman, A., Sibai, F. N., Kanaya, S., Altaf-Ul-Amin, M., Jashim Uddin, M., Chidella, K. K., & Mitra, P. (2021). Image analysis with machine learning algorithms to assist breast cancer treatment doi:10.1007/978-3-030-75490-7_12
Publisher
Springer, Cham
License
Journal
Volume
Issue
PubMed ID
DOI
ISSN
EISSN