الفهرس 
COVEROnly 14 pages are availabe for public view
 |  | from | 125 |  |  |
| | | from | 125 | | |
Abstract
Despite an increased global effort to end breast cancer, it continues to be the most common cancer and the second leading cause of cancer deaths in women in worldwide. We frame the diagnosis problem as that of determining whether a previously detected breast lump is benign or malignant. There are three popu-lar methods for diagnosing breast cancer: mammography, Fine Needle Aspira-tion (FNA) with visual interpretation, and surgical biopsy. The reported sensi-tivity (i.e., the ability to correctly diagnose cancer when the disease is present) of mammography varies from 68% to 79%, of FNA with visual interpretation from 65% to 98%, and of surgical biopsy close to 100%. Therefore, mammog-raphy lacks sensitivity, FNA sensitivity varies widely, and surgical biopsy, alt-hough accurate, is invasive, time consuming, and costly. The goal of the diag-nostic aspect of our research is to aid physicians in detection and diagnosis of breast cancer by developing a relatively objective system that diagnoses FNAs with an accuracy that approaches the best results achieved visually.
First, this thesis presents a fully automated method for cell nuclei detection and segmentation in breast cytological images. FNAC images were enhanced with histogram stretching and contrast-limited adaptive histogram equalization (CLAHE). The locations of the cell nuclei in the image were detected with circu-lar Hough transform (CHT) and local maximum filtering. The elimination of false positive findings (noisy circles and blood cells) was achieved using Otsu’s thresholding method and fuzzy C-means clustering technique. The segmenta-tion of the nuclei boundaries was accomplished with the application of the marker controlled watershed transform in the gradient image, using the nuclei markers extracted in the detection step. The proposed computer-aided detection (CADe) system was evaluated using 92 breast cytological images containing 11,502 cell nuclei. The studied dataset obtained in cooperation with specialists from the archive of Early Cancer Detection Unit-Obstetrics and Gynecology Department, Ain shams University Hospitals. Experimental evidence shows that the proposed method has very effective results even in the case of images with high degree of blood cells, noisy circles.