Automated Leukemia Detection using Contour Signature in Blood Microscopic Images
Journal Title: International Journal of Research in Computer and Communication Technology - Year 2014, Vol 3, Issue 3
Abstract
Out of four major types of leukemia acute lymphoblastic leukemia (ALL) is most common blood disorder in children’s of age ranging from 1-9 years. With lymphoblastic leukemia cancerous change begins in marrow cell where rapid growth of abnormal lymphocytes (a type of WBC) takes place. ALL often leads to wrong diagnosis because of its nonspecific nature. The present techniques that are generally used for diagnosis are time consuming and very costly. Microscopic examination of blood smear and bone marrow aspirate is the most economical technique but manual examination of microscopic images often leads to bias because of human error. Thus, there is a need to automate this procedure. Most convenient way is to do analysis of microscopic images using image processing techniques. The proposed work focuses on automation of microscopic study. In which preprocessing using median filtering is carried out to remove the noise from images. Automatic thresholding gives region of interest. Shape based features of separated nucleus are evaluated and then boundary roughness is evaluated using contour signature which can be used for identification of cancer cells.
Authors and Affiliations
Amruta Deshmukh , Prof. Mrs. Anagha Panditrao
Keywords
Related Articles
- EP ID EP27850
- DOI -
- Views 265
- Downloads 0
Voice and Gesture Based Electric-Powered Wheelchair Using ARM
A voice and gesture based system has been developed to control a wheelchair using voice commands and moment of hand i.e. Mems sensor is connected to hand. The system is divided into two main components: speech recogni...
Secure Data Aggregation In Wireless Networks
Security threat is originated by node capture attacks in hierarchical data aggregation where a hacker achieves full control over a sensor node through direct physical access in wireless sensor networks. It makes a hig...
UM: A Novel Seclusion Model for Defending Structural Information of Social Network Data
Publishing or sharing the social network data for social science research and business analysis lack of privacy. Existing technique k-anonymity is used to prevent identification of micro data. Even though an attacke...
An Efficient Searching and File Replication Consistency in Peer-to-Peer Networks
Unstructured Network with file sharing properties with peer-to-peer (P2P) is very popular in the market. Though it is so popular it produce flooding query messages and result a heavy network traffic and which may not...
http://www.ijrcct.org/index.php/ojs/article/download/1245/pdf
It is pleasing to offer differentiated access services such thatdata access policies are distinct over user attributes or roles,which are direct by the key authorities. We propose a securedata recovery scheme using C...