Breast Cancer: Diagnosis by Image Processing & Deep Learning Techniques

Breast Cancer: Diagnosis by Image Processing & Deep Learning Techniques

Authors: Aqilah Baseri Huddin
Publisher: UKM Press
ISBN: 9789672513254
Weight: 0.23kg
Pages: 133pp
Year: 2021
Price: RM45

Breast cancer is a common threat to women's lives worldwide. Early detection of breast cancer through mammography screening may increase the survival rate. The limitations in reading mammogram images manually by radiologists have motivated interest to the use of computerized systems to aid the process. Computer-aided diagnosis (CAD) systems have been widely used to assist radiologists in making decisions; either for detection, CADe or for diagnosis, CADx, of the anomalies in mammograms. The sensitivity of the CADx system is improved by novel feature extraction techniques. Multiple resolution images provide useful features for classification and wavelet transform is one of the techniques commonly used to produce multiple resolution images. However, the fixed directionality produced by the transform limits the opportunity to extract further useful features that may contain information associated with the malignancy of the detected anomalies. Multiple orientations and multiple resolution images provide features for more complex images of anomalies for classification purposes. Recent advancement in the depth of images has demonstrated the ability to extract high-level complex representations. Thus, the advantage of DBN in being able to analyze complex patterns is exploited for the classification of microcalcification clusters for malignancy analysis. These new findings may contribute to the identification of the microcalcification clusters in mammograms.

AQILAH BASERI HUDDIN, Ph.D., is a senior lecturer in the Department of Electrical, Electronics and System Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia. Her research interests are mainly in the field of image processing and artificial intelligence, specifically in medical imaging applications.