DocumentCode :
3489106
Title :
Application of neural network based hybrid system for lung nodule detection
Author :
Chiou, Y.-S.P. ; Lure, Y. M Fleming ; Freedman, Matthew T. ; Fritz, Steve
Author_Institution :
Caelum Res. Corp., Silver Spring, MD, USA
fYear :
1993
fDate :
13-16 Jun 1993
Firstpage :
211
Lastpage :
216
Abstract :
A hybrid lung nodule detection (HLND) system based on artificial neural network architectures is developed for improving diagnostic accuracy and speed for lung cancerous pulmonary radiology. The configuration of the HLND system includes the following processing phases: data acquisition and pre-processing, in order to reduce and to enhance the figure-background contrast; quick selection of nodule suspects based upon the most prominent feature of nodules, the disc shape; and complete feature space determination and neural classification of nodules. Nodule suspects are captured and stored in 32×32 images after first two processing phases. Eight categories including true nodule, rib-crossing, rib-vessel crossing, end vessel, vessel cluster, bone, rib edge, and vessel are identified for further neural analysis and classification. Extraction of shape features is performed through the edge enhancement self-organized Kohenen feature map, histogram equalization, and evaluation of marginal distribution curves. A supervised back-propagation-trained neural network is developed for recognition of the derived feature curve, a normalized marginal distibution curve
Keywords :
backpropagation; diagnostic radiography; feature extraction; medical image processing; self-organising feature maps; HLND; HLND system; artificial neural network architectures; bone; complete feature space determination; data acquisition; diagnostic accuracy; disc shape; edge enhancement; end vessel; figure-background contrast; histogram equalization; hybrid lung nodule detection; lung cancerous pulmonary radiology; marginal distribution curves; neural classification; nodule suspects; normalized marginal distibution curve; pre-processing; processing phases; rib edge; rib-crossing; rib-vessel crossing; self-organized Kohenen feature map; supervised back-propagation-trained neural network; true nodule; vessel cluster; Artificial neural networks; Bones; Cancer detection; Data acquisition; Feature extraction; Lungs; Neural networks; Performance evaluation; Radiology; Shape;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer-Based Medical Systems, 1993. Proceedings of Sixth Annual IEEE Symposium on
Conference_Location :
Ann Arbor, MI
Print_ISBN :
0-8186-3752-8
Type :
conf
DOI :
10.1109/CBMS.1993.263017
Filename :
263017
Link To Document :
بازگشت