Classification Of White Blood Cells Using ModularNeural Network

Abstract

BLOOD tests are frequently employed to evaluate human health. One of the most straightforward blood tests is to quantify and identify the blood cell types. A complete blood count (CBC) is primarily a measure of these cellular components and is one of the most routinely ordered blood tests by clinicians. CBCs, especially white blood cell (WBC) count, provide physicians with key information valuable for diagnosing many different disease states including: anemia, leukemia, autoimmune disorders, fungal, and bacterial infections as well as Recognition and inspection of white blood cellsin peripheral blood can assist hematologists in diagnosing many diseases such as AIDS, Leukemia, and blood cancer. Thus,this process is assumed as one of the most prominent steps in the
hematological procedure. There are five main phases involved in the system. They are image preprocessing, extraction classifying and segmenting the Five Types of White BloodCells. For classification neural classifiers in HISTOGRAM are used and also be using a more Efficient supervised learning approaches for more accurate and computationally efficient segmentation. The features are extracted
from the Five Typesof White Blood Cells using matlab program approach and these accurate features are used to train the neural classifier. Classification of Five Types of White Blood Cells is an essential research topic as it may be advantageous in monitoring many diseases. Therefore the need for fast, automatic, less expensive and accurate method to classify Five Types of White Blood Cells is of great realistic significance. The main aim of our project work to develop a Computer Aided diagnosed system for classification of Five Types of White Blood Cells.