Journal
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
Volume 21, Issue 1, Pages 691-701Publisher
SPRINGER
DOI: 10.1007/s10586-017-0978-1
Keywords
RBC; SVM; Feature extraction; Segmentation; ELM
Ask authors/readers for more resources
The digitalization of blood slides introduced pathology to a new era. Despite being the most powerful prognostic tool; automated analysis of microscopic blood smear images is still not used in routine clinical practices as manual pathological image analysis methods are still in use that is tedious, time consuming and subjective to technician dependent variation, furthermore it also needs training and skills. In this work, we present novel method based on extreme machine learning approach for the classification of red blood cells (RBC) images. Segmentation of RBC is initiated with statistical based thresholding to retrieve those pixels which are most relevant to RBC followed by Fuzzy C-means for the image segmentation and boundary detection. Different texture and geometrical features are extracted for the classification of normal and abnormal cells. The classification technique is rigorously evaluated against the dataset to evaluate the accuracy of classifier. We have compared the results with state of the art techniques. So far the proposed technique has produced more promising results as compared to the existing techniques.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available