Detecting leukaemia (AML) blood cells using cellular automata and heuristic search

Authors:
Waidah Ismail;Rosline Hassan;Stephen Swift
Affiliations:
Brunel University, West London, UK;Haematology Department, Universiti Sains Malaysia, Kubang Kerian, Kelantan, Malaysia;Brunel University, West London, UK
Venue:
IDA'10 Proceedings of the 9th international conference on Advances in Intelligent Data Analysis
Year:
2010

Citing 6
Cited 1

How to Solve It: Modern Heuristics

How to Solve It: Modern Heuristics
Circle detection on images using genetic algorithms

Pattern Recognition Letters
Segmentation and border identification of cells in images of peripheral blood smear slides

ACSC '07 Proceedings of the thirtieth Australasian conference on Computer science - Volume 62
Using evolvable genetic cellular automata to model breast cancer

Genetic Programming and Evolvable Machines
Biomedical image analysis on a cooperative cluster of GPUs and multicores

Proceedings of the 22nd annual international conference on Supercomputing
Fast training of SVM via morphological clustering for color image segmentation

ICIC'05 Proceedings of the 2005 international conference on Advances in Intelligent Computing - Volume Part I

Use of general purpose GPU programming to enhance the classification of leukaemia blast cells in blood smear images

IDA'12 Proceedings of the 11th international conference on Advances in Intelligent Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a method for the identification of leukaemia cells within images of blood smear microscope slides, which is currently a time consuming manual process. The work presented is the first stage of a procedure aimed at classifying the sub-types of Acute Myeloid Leukaemia. This paper utilises the techniques of Otsu, Cellular Automata and heuristic search and highlights a comparison between random and seeded searches. We present a novel Cellular Automata based technique that helps to remove noise from the images and additionally locates good starting points for candidate white blood cells. Our results are based on real world image data from a Haematology Department, and our analysis shows promising initial results.