Abstract:
Leukemia or blood cancer is one of the most common types of cancer in humans,
which being in the bone marrow and start to produce abnormal cells.
For detecting blood cancer (leukemia) the computer aided diagnoses is a vital tool
which can accelerate the producer of detection.
The database that have been used in this system is the public database
(ALL_IDB1) that include 108 images. From the 108 images only seven image
have been chosen for the system and 100 ROIs are extracted, 50 normal and 50
abnormal. The developed system includes five stages as follows, The pre processing methodology is Segmentation, and the post-processing is feature
extraction using first order statistical features, higher order statistical features and
shape features, feature selection stage using sequential forward selection, the
classification stage using SVM (support vector machine) and KNN (k-nearest
neighbor) gave accuracy of the system 84%.
