TY  - BOOK
AU  - Wafa{u2019}a Abdulhameed Abdullah Alolofi
AU  - Ahmed Mohamed Badawi , 
AU  - Mohammed Ahmed Islam , 
AU  - Muhammad Ali Rushdi , 
TI  - Effects of scaling, noise, and compression on breast cancer detection in whole slide images / 
PY  - 2018///
CY  - Cairo : 
PB  - Wafa{u2019}a Abdulhameed Abdullah Alolof , 
KW  - Digital pathology
KW  - Virtual microscopy
KW  - Whole-slide imaging
N1  - Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of Systems and Biomedical Engineering; Issued also as CD
N2  - Whole slide imaging (WSI) is a recent technology introduced in medical pathology practices. WSI images are created using a computerized system that scans, changes, and stitches pathology specimen glass slides into digital images that have a multi-resolution pyramid construction of a huge gigabyte size. Therefore, digital whole slide imaging brings major challenges in data storage, transmission (telepathology), processing and interoperability.Whole-slide imaging enables histopathological analysis of biological tissues at very high levels of magnification, and hence the early detection of anomalies such as breast cancer can be achieved. In thiswork, we investigate the effects of scaling, compression, and noise on anomaly detection in whole slide imaging. ThusThus, we analyze the effects of image scale on anomaly detection performance. We propose a learning-based approach to find the scale mappings between WSI levels using partial least-square (PLS) regression.The learned scale mapping can be used to detect anomalies in lower-resolution images and small magnification hence reduce the computational cost of anomaly detection.Then we explore the effect of different levels of noise on anomaly detection. We simulate different scenarios where WSI images are contaminated with Gaussian noise and several de-noising algorithms were applied, namely de-noising with PLS, Block Matching 3D (BM3D) and the combination of PLS and BM3D. We show how these different de-noising techniques can help to reduce the noise severity on anomaly detection.Our results lead to useful conclusions on how to handle whole slide images under scaling, compression, and noise conditions
ER  -