Supervised vessel segmentation with minimal features

Author

Azemin, Mohd Zulfaezal Che ; Tamrin, Mohd Izzuddin Mohd

Author_Institution

Kulliyyah of Allied Health Sci., Int. Islamic Univ. Malaysia, Kuantan, Malaysia

fYear

2014

fDate

17-19 Sept. 2014

Firstpage

1

Lastpage

4

Abstract

Current state-of-the art supervised vessel segmentation methods require large number of feature vectors to construct a good model. In this paper, we propose a framework to optimally search for optimal features as inputs to Artificial Neural Network (ANN) trained by Scaled Conjugate Gradient (SCG). SCG is known to speed-up the learning stage in a supervised learning especially when error reduction is critical. The proposed framework is able to reduce features from 16 to 4 dimensions and the overall performance is only decreased by 1% in average.

Keywords

biomedical optical imaging; eye; feature selection; image segmentation; learning (artificial intelligence); medical image processing; neurophysiology; vision; ANN; SCG; artificial neural network; error reduction; feature vectors; optimal features; scaled conjugate gradient; state-of-the art supervised vessel segmentation methods; supervised learning; Accuracy; Artificial neural networks; Feature extraction; Image color analysis; Image segmentation; Retina; Sensitivity; artificial neural network scaled conjugate gradient backpropation; feature selection; vessel segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Functional Electrical Stimulation Society Annual Conference (IFESS), 2014 IEEE 19th International

Conference_Location

Kuala Lumpur

Print_ISBN

978-1-4799-6482-6

Type

conf

DOI

10.1109/IFESS.2014.7036744

Filename

7036744